Alkaptonuria

Homogentisate 1,2-dioxygenase (HGD) deficiency — homogentisic aciduria with ochronosis and ochronotic osteoarthropathy.

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

1. Introduction

1.1. Definition

Alkaptonuria is an inborn error of phenylalanine and tyrosine catabolism caused by deficiency of homogentisate 1,2-dioxygenase (HGD), the third enzyme of the tyrosine degradation pathway, which converts homogentisic acid (HGA) into maleylacetoacetate. The enzymatic block, resulting from biallelic pathogenic variants in the HGD gene, causes accumulation of HGA in body fluids and its urinary excretion in quantities of the order of grams per day. Oxidation of HGA gives rise to a melanin-like pigmented polymer that is deposited in collagen-rich connective tissue, especially cartilage, in a process called ochronosis [1, 2, 20]. The disease manifests with three cardinal elements, ordered by their usual chronology of appearance: urine that darkens on standing or after alkalinisation, ochronotic pigmentation of connective tissue and arthropathy of the spine and large joints [2, 8]. The term "ochronosis" refers to the pigmentation process and should not be used as a synonym for the disease, because ochronosis may also be of exogenous origin [2].

1.2. Historical overview

Alkaptonuria occupies a special place in the history of human genetics. In 1859, Boedeker precipitated from a patient's urine a substance capable of reducing copper, which he called "alkapton"; in 1891, Wolkow and Baumann identified it as 2,5-dihydroxyphenylacetic acid and named it homogentisic acid, because of its structural similarity to gentisic acid [21]. In 1867, Virchow described the generalised pigmentation of cartilage and called it ochronosis, because the bluish-grey or black pigment appeared ochre-coloured under the microscope [21]. In 1902, Archibald Garrod, at the suggestion of William Bateson, proposed that alkaptonuria was transmitted as a recessive Mendelian trait, one of the first human conditions interpreted in this way; alkaptonuria was part of Garrod's initial group of four "inborn errors of metabolism" [1, 22]. From the study of these patients and their families, Garrod derived the idea that each inborn error results from the alteration of a single enzyme by a single genetic event, a forerunner of the "one gene–one enzyme" hypothesis formulated by Beadle and Tatum in 1941 [21]. Biochemical analysis of tissue from an Egyptian mummy of about 1500 BC (Harwa) identified HGA pigment, documenting the disease in antiquity [1, 23].

1.3. Milestones of the molecular discovery

In 1958, La Du and colleagues demonstrated deficiency of homogentisic acid oxidase in the liver, and the defect was later also documented in the kidney [21]. The locus was mapped to the long arm of chromosome 3 by two independent groups: Pollak and colleagues, in 1993, by homozygosity mapping in consanguineous families (a 16 cM region at 3q2), and Janocha and colleagues, in 1994, by linkage in six Slovak pedigrees selected from 191 patients registered in Martin, Slovakia [1, 24, 25]. In 1996, Fernández-Cañón and colleagues cloned the human gene on the basis of homology with the hmgA gene of the fungus Aspergillus nidulans, showed that it lay in the same region and identified missense variants cosegregating with the disease (p.Pro230Ser and p.Val300Gly), with biochemical proof of loss of function [4, 26]. There followed the characterisation of the Slovak, Finnish and Central European variants between 1997 and 2000, resolution of the crystal structure of the enzyme in 2000 and the reviews of the mutational spectrum of 2009 and 2019 [27–32]. In the mouse, an ethylnitrosourea-induced mutation was recognised by the black wood shavings in the cages and located on chromosome 16, in a region syntenic with human 3q [1].

1.4. From symptomatic therapy to HGA reduction

For almost a century, treatment was exclusively symptomatic. Attempts to reduce HGA formation with diets low in phenylalanine and tyrosine and to prevent its oxidation with ascorbic acid did not demonstrate clinical benefit [2, 21]. Nitisinone, a triketone developed as a herbicide and approved for hepatorenal tyrosinaemia, inhibits the enzyme that produces HGA and was proposed for alkaptonuria at the end of the 1990s; the first treated patients were described in 2002 [11, 21]. The randomised trial of the National Institutes of Health (NIH), published in 2011, and the European DevelopAKUre programme, with the SONIA 1 and SONIA 2 trials, led to authorisation of nitisinone for adults with alkaptonuria in the EU in 2020 and, in the USA, to approval of a 2 mg formulation in 2025 [14, 15, 33–35]. In 2026, the HAS published the first national protocol dedicated to the disease, which extends the indication, by expert opinion, to adolescents from age 16 and defines follow-up by age [3].

2. Molecular Genetics and Pathophysiology

2.1. The tyrosine catabolic pathway

Phenylalanine is hydroxylated to tyrosine, which is transaminated to 4-hydroxyphenylpyruvate; 4-hydroxyphenylpyruvate dioxygenase (HPPD) converts this compound into homogentisic acid; HGD opens the aromatic ring of HGA and forms maleylacetoacetate, which is isomerised to fumarylacetoacetate and hydrolysed by fumarylacetoacetate hydrolase (FAH) into fumarate and acetoacetate [21, 22]. HGA is therefore a normal intermediate of aromatic amino acid catabolism, and its accumulation depends on the flux of phenylalanine and tyrosine entering the pathway, determined mainly by protein intake and endogenous protein catabolism [32, 36]. The position of HPPD immediately upstream of HGD explains the rationale for nitisinone: by inhibiting HPPD, the drug suppresses HGA formation, but shifts the accumulation to tyrosine, the substrate of the preceding step; the same mechanism is used, at a more distal point of the pathway, in tyrosinaemia type 1, in which the FAH defect generates toxic downstream metabolites [13, 22].

2.2. The HGD gene and the enzyme

The HGD gene is located at 3q13.33 (historically mapped to 3q21-q23), spans about 60 kb and contains 14 exons; it encodes a 445-amino-acid protein [4, 21]. The MANE Select transcript is NM_000187.4, encoding NP_000178.2; the 2021 GeneReviews uses the previous version NM_000187.3, with the same numbering for the variants listed [2, 6]. By Northern blot, expression is highest in the prostate, small intestine, colon and liver; the functional activity relevant to HGA elimination is concentrated in the liver and kidney, where the enzymatic defect has been demonstrated [4, 21, 22]. The enzyme is organised as a hexamer, in the form of a dimer of trimers; the crystal structure showed that intra- and intersubunit interactions and the extensive surfaces required for folding and assembly can be inactivated at several levels by a single amino acid substitution [4, 30]. Analysis of 172 patients from the SONIA 2 trial and other cohorts, from 39 countries, showed that the predicted residual activity of the variants ranges from 1% to more than 30% [32]. HGD is also expressed in the brain, where HGA may accumulate, without a specific neurological disease having been demonstrated [3].

2.3. From HGA to ochronosis

In the absence of HGD, HGA accumulates and is excreted so efficiently that the plasma concentration, although clearly above normal, remains low relative to the urinary load: an adult excretes 1 to 8 g of HGA per day, versus 20 to 30 mg in a control 24-hour urine [2, 21]. After intravenous administration of carbon-14-labelled HGA, individuals without the disease rapidly oxidise it to carbon dioxide, whereas in patients about 90% is excreted unchanged in the urine [21]. HGA is oxidised, spontaneously and through the action of a polyphenol oxidase present in skin and cartilage, to benzoquinone acetic acid (BQA), a reactive compound that forms adducts with sulfhydryl and amino groups and polymerises into a black pigment; this polymer binds to collagen and other connective tissue macromolecules [21, 22]. The damage results from the direct and indirect effects of polymer binding, with protein oxidation and oxidative stress, and increased osteoclastic activity contributes to the osteoarthropathy [22]. Deposition is selective: cartilage is affected early, whereas the liver and brain remain macroscopically spared, for reasons not yet clarified; in vitro and murine models have confirmed a relationship between HGA concentration and the extent of ochronosis and its modulation by HPPD inhibition [20]. Ochronosis is considered the central event from which cartilage fragility, with fissures and early degeneration, tendon ruptures, valvular calcification and lithiasis derive; once established, it is not fully reversible [3, 20]. Vascular deposition, documented at autopsy in all layers of the aortic wall, is associated with amyloid substance, and secondary amyloidosis has been described in several tissues [3, 22].

2.4. Role of the kidney in HGA elimination and production

The kidney is the main organ of HGA elimination. In an analysis of 225 patients compared with 52 controls, HGA clearance and fractional excretion exceeded the theoretical maximum renal plasma flow, indicating renal production of HGA, and tubular secretion contributed more than glomerular filtration to elimination; circulating HGA concentration increased with age, in parallel with the decline in elimination [37]. A 2026 study in mice with kidney-specific deletion of Hgd concluded that local tyrosine and HGA metabolism in the kidney is negligible [38]. The clinical consequence has been documented in patients with kidney failure: in a 46-year-old man with diabetic nephropathy, plasma HGA was double that of any other patient in the series and ochronosis progressed faster than in his two affected siblings; after kidney transplantation, plasma HGA normalised and urinary excretion decreased by 2 to 3 g/day, suggesting that the graft restores excretion and provides HGD activity [39]. A Portuguese report of two sisters with similar comorbidities showed more severe ochronosis and a fatal course due to aortic stenosis in the sister with chronic kidney disease on peritoneal dialysis, and milder ochronosis, with greater urinary HGA excretion, in the sister with normal kidney function [40].

2.5. Variant spectrum

The number of variants described increased from 17 in 1999 to 91 in 2009 (62 missense, 13 splicing, 10 frameshift, 5 nonsense and 1 stop-loss), 129 in 380 patients in 2016 and more than 200, or more than 300 according to the PNDS, at present, mostly missense and private [3, 4, 22, 31, 41]. Variants cluster in exons 3, 6, 8 and 13 and have been found in all exons except exon 14 [21, 31]. Unlike many genes, substitutions do not preferentially involve CpG dinucleotides: the CCC motif and its reverse complement GGG account for 34.5% of mutated bases, which led to the proposal of the CCC triplet as a mutational hotspot [4, 21]. In 2019, the analysis of 172 patients from 39 countries added 28 novel variants, including three larger genomic deletions detected by multiplex ligation-dependent probe amplification (MLPA) with purpose-designed probes, and showed with minigene assays that three of eight candidate splice-altering variants caused exon skipping or activation of cryptic sites [32]. In Brazil, the first series studied by genome sequencing identified, in addition to missense variants, a deletion of exon 13 in trans with the deep intronic variant c.1007-172A>G, proposed as causal [42].

The following table brings together the recurrent or founder variants described in the sources, with the nomenclature normalised in VariantValidator on the MANE transcript and the historical designation used in the literature.

Historical numbering differs from current numbering because it was calculated on other complementary DNA references; for example, c.183-1G>A in Müller and colleagues (1999) corresponds to c.16-1G>A, and c.648G>A in Srsen and colleagues (2002) corresponds to c.481G>A. The correspondences used are based on the identity of the protein change and the intron name given in the sources themselves and on validation of the current forms in VariantValidator [2, 6, 9, 28]. The p.His80Gln (H80Q) variant is considered a common polymorphism and should not be counted as a pathogenic allele [4, 44].

2.6. Genotype-phenotype correlation, expressivity and modifiers

There is no clinically useful correlation between variant type and HGA excretion or severity [2, 22]. In the 2019 study, variants with about 1% and with more than 30% predicted residual activity differed slightly in urinary HGA excretion corrected for protein intake, with no difference in serum HGA, absolute excretion or symptoms, which led the authors to conclude that protein intake weighs more than the variant in the amount of HGA accumulated [32]. The PNDS states that severity "may vary" with the variants, a statement not supported by the available correlation data [3, 32]. HGA excretion and severity may vary between siblings with the same genotype [2]. The documented modifiers are sex, with earlier joint disease and faster progression in men, age, kidney function and mechanical load on the spine and large joints [2, 11, 39]. Biochemical penetrance is complete: all individuals with biallelic pathogenic variants excrete excess HGA; heterozygotes are asymptomatic and have normal urinary HGA [2, 3].

2.7. Allelic disorders and atypical inheritance mechanisms

No other phenotypes are known to be associated with biallelic HGD variants, nor are there phenotypes in heterozygotes [2]. The mechanism is loss of function. Three probands with three potentially deleterious variants have been described, a situation that requires a segregation study to assign phase [21]. When only one parent is a carrier, the hypotheses are a de novo variant in the proband, a post-zygotic variant in a mosaic parent or uniparental isodisomy [2]. The coexistence of alkaptonuria with other recessive disorders mapped nearby, such as sucrase-isomaltase deficiency or severe neonatal hyperparathyroidism in a family with familial hypocalciuric hypercalcaemia, was used historically to guide mapping to 3q [1].

3. Clinical Spectrum and Manifestations by System

3.1. Natural history cohorts

Knowledge of the natural history rests on a few cohorts, recruited in reference centres and biased towards symptomatic patients. The NIH cohort assessed 58 patients aged 4 to 80 years with clinical, radiographic, biochemical and molecular methods; by life-table analysis, joint replacement occurred at a mean age of 55 years, kidney stones at 64 years, cardiac valve involvement at 54 years and coronary calcification at 59 years, and the radiographic severity score began to rise after age 30, faster in men than in women [1, 11]. The same team studied 53 patients aged 10 to 80 years (mean 43.6 years) with regard to musculoskeletal manifestations and disability [45]. The UK National Alkaptonuria Centre (NAC), in Liverpool, followed 87 patients between 2007 and 2020, 57 of them on nitisinone 2 mg/day, and published analyses of bone, joints, the aorta and nutrition [46–49]. The SONIA 2 trial included 138 adults aged 25 years or over with clinical manifestations, followed for 4 years, and provided the largest available untreated arm [35, 50]. Outside these centres, there are series from France (96 patients published since 1882), Turkey (66 patients), China (91 patients in the literature), Portugal (10 patients in a reference centre) and Brazil (5 patients studied by genome sequencing) [8, 42, 43, 51, 52]. The data that follow refer to the untreated natural history, unless otherwise stated.

3.2. Childhood and adolescence

The only constant manifestation in childhood is urine that darkens after voiding. Urine is normal when passed and darkens over minutes to hours, on contact with air, or immediately on addition of an alkali; cloth nappies washed with alkaline detergent turn black, and current disposable nappies turn pink or reddish; sweat may stain underwear [3, 21]. Paediatric diagnosis is rare, because the sign is not given weight: in the illustrative Portuguese series and in the Brazilian series, dark urine had been present since childhood and the diagnosis was made only in adulthood in some of the cases [42, 53]. In Turkey, 56.1% of 66 patients presented with isolated dark urine, which reflects a higher proportion of paediatric diagnoses [43]. Spurious proteinuria, detected by laboratory assay methods with normal dipstick and albuminuria, may lead to the diagnosis [3]. Urinary lithiasis may occur in childhood: Zibolen and colleagues described 5 Slovak patients younger than 15 years, one diagnosed with lithiasis at age 2 [21, 54]. This finding diverges from the PNDS statement that lithiasis has not been described in children [3]. Ochronosis is practically not described before adulthood; the PNDS mentions a 16-year-old adolescent with a small ocular spot, and Sloboda and colleagues described HGA deposition in the teeth of a 5-year-old girl [3, 55]. In Turkey, 7 of 66 patients had joint pain, low back pain or kidney stones in childhood [43]. The PNDS considers childhood osteoarticular involvement unlikely and does not attribute the paediatric arthralgias described to the disease, and osteoarticular and cardiac imaging is normal in children [3]. In a Turkish case-control study of 16 children, bone turnover markers did not differ from controls [56].

3.3. Ochronosis: eyes, ears and skin

Pigmentation usually becomes visible after age 30, and the PNDS accepts it from late adolescence [2, 3]. In the eye, three locations are distinguished: scleral, visible to the naked eye, with brown patches in the horizontal meridian, on both sides of the cornea, at the insertion of the medial and lateral rectus muscles; conjunctival, with vermiform pigmentation requiring a slit lamp; and limbal, with "oil-drop" lesions in the horizontal meridian, considered pathognomonic [3, 57]. Ocular pigmentation does not affect vision; there was a historical case of enucleation due to confusion with melanosarcoma [2]. The late ocular complications discussed are earlier cataract, glaucoma due to pigment accumulation in the iridocorneal angle and progressive astigmatism due to deformation of the globe [3]. In the ear, the cartilage of the concha and antihelix becomes slate-blue or grey, irregular or thickened, and may calcify; the tympanic membrane and cerumen may also become pigmented [2, 21]. In the skin, pigmentation appears later, as spots or diffusely, especially on the back of the hands, over the tendons and in the web between the thumb and index finger, and also on the nose, cheeks, axillae and genital region; the nails may have bluish-grey longitudinal streaks and sweat may be dark [2, 21, 22]. In the NAC cohort, 90.5% of patients had photographically documented ochronosis [47]. In the Portuguese cohort of 10 patients, nailfold capillaroscopy showed blue-black deposits in 6, mostly invisible to the naked eye and absent in the control groups, and a higher frequency of abnormally shaped capillaries (70% versus 20%) [52].

3.4. Spine

Ochronotic spondyloarthropathy is the dominant manifestation. Low back pain is the earliest and most frequent symptom: in the NIH cohort it appeared before age 30 in 49% and before age 40 in 94% of patients [2, 11]. Without treatment, pain appears after age 25 in men and 30 in women [3]. Lumbar and thoracic symptoms precede cervical ones, with pain of mechanical or inflammatory pattern, stiffness and progressive loss of lumbar lordosis, thoracic hyperkyphosis and loss of height, detectable from age 40 and with a mean reduction of about 8 cm [2, 3, 21]. In the NAC cohort, spinal pain was lumbar in 77.4%, cervical in 59.5% and thoracic in 46.4% of patients, and positron emission tomography (PET) showed thoracic uptake in 74.4%, lumbar in 70.7%, cervical in 64.6% and sacroiliac in 47.8% [47]. Neurological complications result from multilevel disc degeneration: radicular pain, disc herniation, lumbar spinal stenosis and, more rarely, myelopathy, which in a review of surgical cases was most often cervical, then thoracic and lumbar [2, 3]. Limitation of forward flexion of the spine (positive Schober test) correlates with functional loss and disability: of 53 patients, 18 had kyphosis, 16 scoliosis, 16 marked limitation of at least one large joint, 15 joint replacements and 11 tendon ruptures [45].

The radiographic appearance may be pathognomonic. Multilevel disc calcifications begin in the central part of the annulus fibrosus, extend to the whole disc and predominate in the lumbar spine; they become less visible as the disc space is destroyed [3]. They are associated with multilevel disc space narrowing, intradiscal vacuum phenomenon, sclerosis of the vertebral endplates and mild marginal osteophytosis; progressive ossification of the discs with bony bridges and vertebral fusion may give a "bamboo" appearance and mimic ankylosing spondylitis [3, 21]. In the SONIA 2 trial, narrowing of the intervertebral spaces, osteophytes or hyperostosis and calcifications were the most frequent baseline findings, and progression over 4 years, slow but significant, followed the order narrowing, calcifications, vacuum, osteophytes and fusions [50]. In 2025 the NIH developed a 13-level spinal radiographic score (C2–C7 and T10–S1), based on disc narrowing, calcification and vacuum (maximum 78 points), applied to 409 radiograph series from 136 participants, with high intra- and interobserver reliability (intraclass correlation coefficient above 0.95) and a minimal detectable change of 2.6 points; the score correlated with age, sex, pain and physical function [58].

The sources diverge regarding the sacroiliac joints: GeneReviews and the ATLAS describe them as usually spared, and this is a criterion distinguishing it from ankylosing spondylitis, whereas the PNDS accepts sacroiliac and pubic symphysis involvement and NAC PET showed sacroiliac uptake in 47.8% of patients [2, 3, 21, 47]. As for osteophytosis, the ATLAS describes it as scant and GeneReviews and the PNDS as present but mild [2, 3, 21].

3.5. Peripheral joints, tendons and muscles

The large proximal joints are the most affected: knees, hips and shoulders; the small joints of the hands and feet tend to be spared [2, 22]. The arthropathy is degenerative, with mechanical or inflammatory pain, flares with synovial effusion mimicking crystal arthritis, suprapatellar bursa effusions, intra-articular loose bodies and progressive stiffness [3, 21]. Radiographically, there is global joint space narrowing, moderate osteophytosis, subchondral sclerosis, collapse of the articular surfaces, fragmentation and tendon and bursal calcifications; pain may precede imaging changes [3]. In the NAC cohort, joint pain was present in 85.7% of patients and PET showed joint or spinal uptake in all, sometimes from age 25 and before clinical signs [3, 47]. The prevalence of arthroplasty was 36.8%, mainly knee, hip and shoulder, with multiple arthroplasties in 33.3%, with no difference between sexes, and was associated with spondyloarthropathy (R = 0.5) and ochronosis (R = 0.54) [48]. Half of the NIH patients had at least one joint replacement by age 55, no patient required a joint replacement before age 40, and the prevalence of joint replacements is about three times that of osteoarthritis [2, 3]. The supraspinatus, long head of biceps and Achilles tendons show tendinosis, calcification and ruptures after minimal trauma, more severe with higher HGA excretion; Achilles tendon thickening is palpable and demonstrable on magnetic resonance imaging (MRI); ligament and muscle ruptures have been described [2, 3, 21]. There may be raised erythrocyte sedimentation rate (55 to 110 mm/h in the NIH series) and bone formation and resorption markers during inflammatory phases [21].

3.6. Bone: osteoporosis and fractures

At the NAC, 39 of 87 patients had 48 fractures at baseline; the prevalence of fragility fractures was 44.8% and that of osteoporosis was 3.1% at the femoral neck, 10.8% at the total hip and 24.7% at the lumbar spine [46]. The PNDS cites the same cohort (mean age 48 years, 63% men) with 47.1% of patients with low-energy fractures and an incidence of 5.1% per year, increasing with age and associated with greater severity of ochronosis [3]. Bone loss is associated with accelerated turnover, with raised procollagen type 1 N-terminal propeptide (P1NP) and C-terminal telopeptide (CTX), more marked in men and in severe disease [3]. Vertebral fractures in an ankylosed spine create a risk of instability and neurological injury [3]. Lumbar dual-energy X-ray absorptiometry (DXA) is unreliable, because disc calcifications overestimate density, and quantitative computed tomography (CT) correlated better with fractures [3, 46].

3.7. Heart and vessels

Cardiovascular involvement results from pigment deposition in the valves, endocardium and vascular wall. Of 76 consecutive adults studied by echocardiography at the NIH between 2000 and 2009, 6 already had an aortic valve prosthesis and, of the remaining 70, 12 had aortic sclerosis and 7 aortic stenosis; there was no aortic valve lesion before age 40, cusp calcifications appeared from a median age of 52 and, above age 60, only 30% had normal cusps and 25% had already undergone surgery [3, 59]. Unlike degenerative valve disease, there was no correlation with classic cardiovascular risk factors or with urinary HGA, and there was a modest association with joint severity [59]. On non-contrast CT of 40 of these patients, calcifications were observed in the coronary arteries, valves, aortic root, descending aorta and iliac arteries [59]. No patient younger than 40 had coronary calcification, which was present in 50% by age 59 [3, 11]. In 81 NAC patients (mean age 48.6 years), the prevalence of aortic stenosis was 22.2% and that of valve replacement 6.2%; peak aortic transvalvular velocity (Vmax) correlated with age (R = 0.58), systolic pressure (R = 0.32), serum HGA (R = 0.28), ochronosis (R = 0.72) and the severity index (R = 0.58), and age and ochronosis were independent predictors [49]. In SONIA 2, 13.0% of the 138 patients had aortic stenosis and 18.1% aortic sclerosis at baseline [60]. Distensibility of the descending thoracic aorta, measured by MRI in 76 patients, was reduced relative to reference values (5.2 versus 6.2 × 10⁻³), began to decline at age 35, was significantly lower from age 50 and was associated with aortic calcification, independently of hypertension (38% of the cohort) and dyslipidaemia (42%) [3, 61]. Aortic dilatation is occasional, mitral calcification rarely has consequences and the electrocardiogram and Holter monitoring were normal in 16 adults with a median age of 55.5 years [2, 3]. There are no studies of the risk of acute cardiac events; in SONIA 2, there were two deaths over 4 years, one from heart failure of unspecified cause and the other from myocardial infarction [3].

3.8. Kidney and urinary tract, prostate, gallbladder and salivary glands

Renal HGA deposition is associated with interstitial fibrosis and tubular atrophy, but kidney function is usually preserved [3]. Kidney stones are frequent: half of the NIH patients had a history of lithiasis by age 64, and the PNDS cites a first occurrence at a median age of 48 and considers that probably all patients eventually develop lithiasis, symptomatic or not [2, 3]. The stones are black or dark brown and contain calcium oxalate, phosphate and carbonate or weddellite, and their formation has been related to increased calcium and phosphate excretion due to active bone resorption associated with urinary HGA [3]. Prostatic stones are black and frequent: 8 of 27 men aged 31 to 60 in the NIH cohort, up to half of men according to the PNDS and very frequent after age 60; they may cause recurrent infection or obstruction [2, 3, 21]. Factors that may impair kidney function are repeated obstruction by stones, the pro-inflammatory effect of the deposits, non-steroidal anti-inflammatory drugs (NSAIDs) used for pain and comorbidities such as diabetes and hypertension [3]. Severe kidney failure is rare, but accelerates ochronosis and joint destruction, through increased plasma HGA [39, 40]. Biliary and salivary lithiasis has been described in isolated cases, with no frequency established in the large cohorts [3].

3.9. Metabolic crises with methaemoglobinaemia and haemolysis

At least 11 acute episodes of methaemoglobinaemia and/or severe haemolysis associated with acute kidney injury or acute-on-chronic kidney disease have been published, all fatal [3, 62]. The proposed mechanism is the rise in plasma HGA, which has oxidant properties, through loss of renal clearance, in a probable context of altered red cell redox status, genetic or acquired, since many patients with kidney failure never developed this complication [3, 62]. The presentation is respiratory distress with low peripheral oxygen saturation and normal arterial oxygen tension, with significant methaemoglobinaemia from 20 to 30% [3]. Treatment is covered in section 7.

3.10. Thyroid and other endocrine manifestations

In a cohort of 125 NIH patients (median age 45 years, 57.6% men), the prevalence of primary hypothyroidism was 16.0% (20/125), versus 3.7% in the general population (difference 0.12; 95% confidence interval [CI] 0.10–0.24), one third subclinical, with the youngest patient aged 8; women (odds ratio [OR] 10.99; 95% CI 3.13–38.66) and patients with thyroid peroxidase antibodies (OR 7.36; 95% CI 1.89–28.62) were more likely to have hypothyroidism, whereas the prevalence of hyperthyroidism (0.8%), nodules and cancer did not differ from the general population [3, 12]. There was no relationship with HGA, tyrosine or nitisinone; pigmented deposits in follicular cells have been described in thyroidectomy specimens, and the hypothesis of competition between HGA and tyrosine for transport into the thyroid remains unproven [3]. There is no demonstrated increase in the prevalence of type 2 diabetes, and the two published cases of multiple autoimmunity have no known mechanistic basis [3].

3.11. Nervous system, cognition and mental health

Alkaptonuria does not cause developmental delay or cognitive impairment [2]. Headaches, seizures and syncope are not more frequent than in the general population [3]. Ten patients with a parkinsonian syndrome have been described, 4 of them in a cohort of 88 NAC patients (4.5%), with a prevalence estimated at about 20 times that of the age-adjusted general population; two of these patients developed Parkinson's disease before nitisinone and two during treatment, and the authors proposed prolonged exposure to redox-active HGA and unmasking of a dopaminergic deficit by tyrosinaemia as mechanisms [63]. The PNDS considers that, in the absence of another series confirming the observation, the association is not established [3]. Chronic pain, disability and occupational limitations are associated with depressive symptoms: 8 of 66 Turkish patients had diagnosed depression [3, 43].

3.12. Treated disease

Data on patients treated with nitisinone come from adults who started the drug at a mean age close to 48 years, with established lesions. In these patients, nitisinone reduced ocular and ear ochronosis, with partial reversal documented by photographs and ear cartilage biopsies, reduced progression of the clinical severity index and pain, but had no demonstrated effect on established joint lesions, the incidence of new arthroplasties or fractures [3, 35, 46, 48, 64]. There are no long-term data on patients treated from adolescence or young adulthood, the period in which the expected preventive effect would be greatest [3, 65].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age

In infants and children, suspicion rests on urine that darkens in nappies or on clothing minutes to hours after voiding, or on sweat that stains clothing black [3, 21]. Many patients never observe the dark urine, because darkening may take hours and urine is rarely left standing, so the absence of such a report does not exclude the disease [2, 21]. At any age, laboratory proteinuria discordant with a normal dipstick and albuminuria, or a positive test for reducing substances without glycosuria, should prompt a request for urinary HGA [3, 21]. An unexplainedly low or unstable urinary creatinine measured by an enzymatic method may also result from HGA interference [3, 66].

In young adults, the warning signs are low back pain or spinal stiffness from age 25 to 30, especially with multilevel disc calcifications on radiography, tendon ruptures after minimal trauma, kidney or prostatic stones and scleral or ear patches [2, 3]. In middle-aged adults, the usual presentation is early osteoarthritis of the hips, knees or shoulders disproportionate to age, a spondylopathy mistaken for ankylosing spondylitis, or the discovery of black cartilage and bone during arthroplasty or spinal surgery [3, 53]. Calcific aortic stenosis without classic risk factors, in a patient with joint disease, should also suggest the disease [59]. The PNDS summarises the diagnosis as made at two periods of life: in early childhood, through the nappies, or in adulthood, by the surgeon or the rheumatologist [3]. In the Chinese series of 91 patients, the mean age at diagnosis was 37 years (standard deviation 23 years) [51].

4.2. Targeted questions and observations

When the disease is suspected, the history includes the colour of urine in childhood, the colour of sweat, cerumen and clothing, a history of lithiasis, tendon ruptures and low-energy fractures, the age at onset of low back pain, a family history of siblings with early joint disease and consanguinity [2, 3]. The examination looks for scleral pigmentation in the horizontal meridian, pigmentation of the concha and antihelix, discolouration of the tympanic membrane, pigmentation of the back of the hands and nails, Achilles tendon thickening, loss of lumbar lordosis with hyperkyphosis, reduced Schober test and an aortic stenosis murmur [2, 3, 45]. Observing freshly voided urine that darkens with a few drops of sodium hydroxide is a quick pointer, but does not replace measurement [3, 22].

4.3. Emergency signs

The most serious acute situation is acute kidney injury, or acute-on-chronic kidney disease, in a patient with alkaptonuria, which may trigger intravascular haemolysis and methaemoglobinaemia with a fatal outcome [3, 62]. Dyspnoea with peripheral desaturation and normal arterial oxygen tension, a greyish skin colour, dark urine and plasma or rapidly developing haemolytic anaemia require blood gas analysis with co-oximetry and assessment of kidney function [3, 67]. Other emergencies are spinal cord or nerve root compression due to disc herniation or fracture in an ankylosed spine, obstructive renal colic, urinary retention due to prostatic stones and syncope or exertional chest pain due to aortic stenosis [2, 3]. On nitisinone, eye pain, photophobia, watering, blurred vision or red eye require slit-lamp examination and tyrosine measurement without delay [3, 13, 15].

4.4. Newborn screening and cascade screening

The sources consulted do not describe any current newborn screening programmes for alkaptonuria. The Slovak experience, with urinary screening of about 509,000 newborns since 1968, allowed the incidence to be estimated at 1:19,000 and 208 patients to be registered, 110 of them children [9]. Screening of the siblings of an index case is done by urinary HGA, which reliably diagnoses or excludes the disease, without the need for molecular study [2, 3]. The children of a patient may be tested by the same method during childhood, if the parents so wish [3].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Criteria

No published consensus clinical criteria exist [2]. The biochemical diagnosis is established in a patient with suggestive findings and a significant amount of HGA in the urine, detected by GC-MS or LC-MS/MS; the molecular diagnosis is established by identification of biallelic pathogenic or likely pathogenic variants in HGD [2]. The PNDS considers raised urinary HGA pathognomonic and sufficient for a definitive diagnosis, without the need for genetic study in the index case when the clinical and biochemical findings are concordant [3]. Alkaptonuria is therefore a disease in which the biochemical test takes precedence over the genetic test: biallelic variants of uncertain significance (VUS) neither establish nor exclude the diagnosis, whereas HGA excretion in gram quantities establishes it [2, 5].

5.2. Measurement of homogentisic acid

HGA can be measured specifically by GC-MS or quantified in the urinary organic acid profile by GC-MS, in which it appears as an isolated peak; methods by high-performance liquid chromatography (HPLC), by LC-MS/MS with simultaneous measurement of tyrosine and by enzymatic methods have also been described [2, 3, 22, 68]. A random urine sample is sufficient, with no particular timing or duration, collected at any laboratory and sent to a specialised laboratory, because excretion is massive and permanent [2, 3]. Reference values depend on the unit of expression, as the following table shows.

Screening tests have historical or triage value: alkalinisation of urine with sodium hydroxide causes immediate darkening, urine reduces reducing-substance reagents and the silver of photographic emulsions, and the reducing-substance test is negative when the laboratory uses glucose oxidase only [3, 21, 22]. These tests do not quantify HGA and do not replace chromatography.

5.3. Laboratory interference

HGA interferes with several routine assays, depending on the method. Urinary creatinine by an enzymatic method may be underestimated; chemical measurement of urinary proteins gives spurious proteinuria; on the other hand, the dipstick and urinary protein electrophoresis are not affected [3, 66, 69]. On nitisinone, with HGA close to normal, the interference disappears [3]. These interferences should be communicated to the laboratory when analyte/creatinine ratios or proteinuria are interpreted in a patient with untreated alkaptonuria.

5.4. Molecular study

Molecular study is requested for genetic counselling and testing of relatives, and is necessary when an incidental finding of HGD variants arises on exome or genome sequencing, a situation in which confirmation is by urinary HGA [2, 3]. Sequencing of HGD, alone, in a panel or on exome/genome, detects about 97% of pathogenic variants, and gene-targeted deletion/duplication analysis (MLPA, quantitative or long-range PCR, targeted microarray) detects the remaining 3% [2, 32]. When sequencing identifies only one variant or none in a patient with a definite biochemical diagnosis, the next step is a search for copy-number variation and then analysis of the promoter and non-coding regions by genome sequencing, discussed at a multidisciplinary meeting [2, 3]. The Brazilian series illustrates this limit: in one patient, the deletion of exon 13 was in trans with a deep intronic variant (c.1007-172A>G) that exonic sequencing would not have detected [42]. The figure of 97% refers to the variants detectable by the technique and not to the clinical sensitivity of the test, which also depends on coverage of intronic and regulatory regions [2]. A complete molecular report states the gene, the transcript and version (NM_000187.4), the HGVS DNA and protein nomenclature, zygosity, classification according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) and the recommendation of a segregation study in the parents, indispensable to confirm the trans phase of two heterozygous variants [2, 6]. Any genetic study requires informed consent [3].

5.5. Diagnostic algorithm

The algorithm that follows from the sources is sequential. In the presence of clinical suspicion (dark urine, ochronosis, early spondyloarthropathy, black cartilage at surgery, spurious proteinuria), urinary HGA is requested on a random sample; a raised value confirms the diagnosis, and a normal value, with a careful drug and exposure history, excludes it and points to the causes of exogenous ochronosis [2, 3, 70]. Once the diagnosis is confirmed, a staging assessment and referral to a metabolic disease centre follow, together with molecular study of the index case and parents for counselling [2, 3]. Clinical action on the genetic result was classified by ClinGen as of moderate actionability in adults, with no paediatric assessment [5].

5.6. Initial assessment after diagnosis

The initial assessment recommended by GeneReviews and the PNDS comprises a complete history and examination with range of motion of the spine and large joints, physical medicine and rehabilitation assessment if there is limitation or pain, electrocardiogram and echocardiogram (GeneReviews indicates them from age 40; the PNDS includes echocardiography in the initial work-up before nitisinone), renal ultrasound or helical abdominal CT for stones, TSH and free thyroxine, ophthalmological assessment with slit lamp, densitometry, dental examination and genetics consultation [2, 3]. Before nitisinone, the PNDS adds full blood count with platelets, prothrombin time, electrolytes, creatinine, ferritin, folate, vitamin B12, total homocysteine, 25-hydroxyvitamin D, parathyroid hormone, liver enzymes, bilirubin, lipid profile, plasma amino acid chromatography and urinary HGA, and, optionally, spinal MRI and bone scintigraphy to calculate the severity score [3].

5.7. Prenatal and preimplantation diagnosis

Once the two familial variants are known, prenatal diagnosis and preimplantation genetic testing are technically possible [2]. The PNDS does not routinely offer them, given the normal life expectancy and cognitive development, the existence of treatment and onset in adulthood, and refers couples' requests to a multidisciplinary decision in a prenatal diagnosis centre [3].

6. Differential Diagnosis

The differential is organised around the three typical presentations: dark urine, ochronotic pigmentation and early spondyloarthropathy or arthropathy. In all, the discriminating test is urinary HGA, which is normal in all the alternatives [2, 3].

Exogenous ochronosis is the differential that most often leads to a false diagnosis. At the NIH, among 88 patients referred with suspected alkaptonuria between 2000 and 2004, five women with ear pigmentation and mild degenerative disease of the spine and large joints had taken minocycline for prolonged periods for dermatological or rheumatological conditions, and the diagnosis was withdrawn because urinary HGA was normal [70]. The PNDS adds that minocycline may irreversibly blacken the craniofacial skeleton, which may mimic the surgical finding of black bone [3]. Ochronosis induced by carbolic acid dressings on chronic ulcers is now historical, and pigmentation due to quinacrine and hydroquinone creams has a similar distribution, with possible associated osteoarthritis in the case of quinacrine [1, 2].

In the spine, confusion with ankylosing spondylitis is frequent because both affect the spine and large joints and may produce bony bridges and vertebral fusion; multilevel disc calcification, the vacuum phenomenon and, according to GeneReviews and the ATLAS, sacroiliac sparing favour alkaptonuria [2, 3, 21]. The PNDS points out, however, that there may be sacroiliac and pubic symphysis involvement, so the presence of sacroiliitis does not exclude alkaptonuria [3]. Early osteoarthritis, with hip and shoulder involvement more frequent than in common osteoarthritis, should prompt examination of the sclerae and ears and a request for urinary HGA, as illustrated by the Portuguese case in which a complete rheumatological assessment with negative serology ended without a diagnosis until arthroplasty [21, 53]. Among metabolic arthropathies of children and young adults, alkaptonuria belongs to the rheumatological differential of inborn errors of metabolism, alongside homocystinuria, Gaucher disease, Farber disease and the mucolipidoses, whose pictures differ by the absence of HGA and by their own systemic findings [71].

7. Treatments and Risk Management Strategies

7.1. Principles and organisation of care

Treatment has two levels. The first is reduction of HGA production with nitisinone, whose effect is essentially preventive: it slows progression of ochronosis and of the lesions that depend on it, but does not repair established joint lesions, so the expected benefit is greater the earlier treatment starts, ideally at a presymptomatic or mildly symptomatic stage [3, 13]. The second level is symptomatic and surgical treatment of osteoarticular, cardiac, urological, bone and endocrine complications, which is necessary even in patients treated with nitisinone [2, 3]. The PNDS reserves initiation and follow-up of nitisinone and the associated diet to a hospital metabolic disease physician within the rare disease network, and assigns to the general practitioner coordination, screening for complications, first-line symptomatic treatments and referral to orthopaedics, rheumatology, urology, the pain clinic, physiotherapy, psychology and social services [3]. The European label requires nitisinone to be initiated and supervised by a physician experienced in the treatment of these patients [13].

7.2. Nitisinone: mechanism, formulations and doses

Nitisinone, 2-(2-nitro-4-trifluoromethylbenzoyl)-1,3-cyclohexanedione, is a competitive and reversible inhibitor of HPPD, the enzyme that forms HGA from 4-hydroxyphenylpyruvate [1, 15]. It reduces plasma and urinary HGA almost immediately, with near-normalisation of excretion within a few days, and increases plasma tyrosine in proportion to the dose [3, 34]. The pharmacokinetics described in the US label, in healthy volunteers, indicate a time to maximum concentration of 3.5 hours, plasma protein binding above 95%, a volume of distribution of 8.2 L and a terminal half-life of 59.3 hours, with no relevant effect of food on exposure [15]. In the EU, the product authorised for alkaptonuria is Orfadin® (capsules and 4 mg/mL oral suspension; the capsule may be opened and the contents suspended in water immediately before taking) [13]. In the USA, the product approved for alkaptonuria is Harliku™, as 2 mg tablets; the other nitisinone formulations registered in the USA (Orfadin®, Nityr® and generics) are indicated only for tyrosinaemia type 1 [15]. In Brazil, nitisinone is registered as Orfadin® and Nitikabs® [72]. The healthcare professional package insert for Orfadin® approved by the Brazilian Health Regulatory Agency (ANVISA) in 2024 (registration MS 1.3840.0002, Innovative Medicines Brasil, under licence from Swedish Orphan Biovitrum) restricts the indication to the treatment of adult and paediatric patients of any age with confirmed hereditary tyrosinaemia type 1, in combination with dietary restriction of tyrosine and phenylalanine; alkaptonuria is not among the indications, and the text states that it is based on the European SmPC in the version of 20 December 2019, prior to the 2020 European extension [14, 16]. The registered presentation is a bottle of 60 capsules of 2, 5, 10 or 20 mg, stored between 2 and 8 °C and usable for up to 2 months after opening [16]. In Portugal, Orfadin® benefits from the European centralised authorisation, including the alkaptonuria indication, and INFARMED's Infomed database lists it as 2, 5, 10 and 20 mg capsules; the same database lists a generic, Nitisinona Dipharma® (5, 10 and 20 mg capsules), authorised in November 2018 by the decentralised procedure, before the extension of Orfadin® to alkaptonuria [13, 14, 73]. The current Portuguese SmPC of the generic already includes alkaptonuria: treatment of adults, with 10 mg once daily, initiation and supervision by a physician experienced in the disease, safety and efficacy not established from 0 to 18 years, and tyrosine monitoring with a tyrosine- and phenylalanine-restricted diet (target below 500 µmol/L) only in patients who develop keratopathy; the capsule may be opened and suspended in a little water or in formula immediately before taking [74]. The dosing coincides with that of the Orfadin® SmPC.

The reference dose is that of the 2026 PNDS: 10 mg orally, once daily, in adults and in adolescents from age 16, always combined with a protein-restricted diet controlled in phenylalanine and tyrosine; the dose may be reduced during follow-up if urinary HGA remains on target, ideally below 10–20 µmol/mmol creatinine in adults and below 10 µmol/mmol in adolescents (normal value below 2) [3]. The other published doses differ from one another and are shown in the following table for comparison.

The choice between 10 mg and 2 mg is based on different data. The 10 mg dose was that of the SONIA 2 trial, which demonstrated an effect on the clinical severity index, and a post hoc modelling analysis estimated that 10 mg slowed progression about four times more than 2 mg, at the cost of more keratopathy (14.5% versus 4.9%) [3, 35, 76]. The 2 mg dose is that of the NIH trial, which supported US approval on the basis of reduction of urinary HGA [15, 33]. In the SONIA 1 dose-response trial, 24-hour urinary HGA fell in a dose-dependent manner, from 31.53 mmol untreated to 3.26, 1.44, 0.57 and 0.15 mmol with 1, 2, 4 and 8 mg/day after 4 weeks [34]. A 2021 review argued that doses of 0.2 to 2 mg/day, which already reduce urinary HGA by more than 90%, could be used preventively in younger patients with fewer adverse effects, and identified as open questions the minimum individual dose, the age at initiation and the urinary HGA level that prevents complications [65].

7.3. Treatment by age according to the PNDS

In adults, even young and asymptomatic ones, nitisinone is considered consensual and early initiation is recommended to avoid, delay or limit osteoarticular lesions [3]. From age 16, by expert opinion, the PNDS proposes nitisinone 10 mg/day with the associated diet, in a reference or competence centre, with the aim of reducing urinary HGA by more than 90% and preventing complications; the dose may then be reduced if the biochemical target is maintained [3]. Between ages 12 and 16, introduction of nitisinone at a lower dose, with the diet, may be considered after discussion at a multidisciplinary meeting of a reference centre; the PNDS does not set this dose [3]. Before age 12, nitisinone is not recommended, and the possible approaches, to be discussed case by case, are no treatment, a protein-limited diet alone or an oral antioxidant (vitamin C, with or without N-acetylcysteine); none has demonstrated clinical efficacy [3]. The rationale for this caution is the absence of paediatric studies in alkaptonuria, the difficulty of adhering to a very restrictive diet in childhood, the risk of nutritional deficiencies and the unknown effect of prolonged hypertyrosinaemia on cognitive development, given that children with tyrosinaemia type 1 treated with nitisinone show cognitive difficulties of unexplained cause [3]. In tyrosinaemia type 1, nitisinone has been used since the 1990s from the neonatal period, at doses of 1 to 2 mg/kg/day, much higher than those used in alkaptonuria [3, 75]. The European and US labels do not include patients younger than 18 [13, 15]. The ClinicalTrials.gov registry includes a single study of nitisinone in different age groups (8 participants, completed in 2016), with no results in the sources consulted [77].

7.4. Nutritional treatment and tyrosine targets

The diet has two distinct roles. On its own, without nitisinone, restriction of phenylalanine and tyrosine reduces urinary HGA only with the most restrictive regimens, by up to 90% of the previous value, without ever falling below 500 µmol/mmol creatinine (usually 1000–2000), with difficult long-term adherence; GeneReviews considers it ineffective and potentially dangerous when severe, and the PNDS does not recommend it as sole treatment [2, 3, 78, 79]. Combined with nitisinone, the diet controls drug-induced hypertyrosinaemia and prevents keratopathy: in the mouse model, tyrosine and phenylalanine restriction reduced tyrosine from 813 µmol/L to 144–389 µmol/L in 3 days, whereas phenylalanine restriction alone was ineffective; in 10 NAC patients with tyrosine above 700 µmol/L, protein restriction, with or without tyrosine- and phenylalanine-free amino acid mixtures, significantly reduced tyrosine, and 4 reached values below 700 µmol/L [36]. Comparison between NAC patients, with formal dietetic follow-up, and those in SONIA 2, who received only advice on self-restriction of protein, showed that dietary intervention stabilised lean mass and grip strength, moderated tyrosine and reduced the frequency of keratopathy, but that protein restriction was associated with fat mass gain [80].

Tyrosine targets diverge between sources. The PNDS sets, in adolescents on nitisinone, a fasting plasma tyrosine below 400 µmol/L, accepting 400–800 µmol/L with monitoring for keratitis and cognitive development when adherence is difficult; in adults, it considers a target below 800 µmol/L reasonable in the absence of symptoms, by analogy with tyrosinaemia type 1, in which the national and international target is below 400 µmol/L but rarely achieved by adults [3]. The European and US labels do not impose a routine diet and require it only in the presence of keratopathy, with the aim of keeping tyrosine below 500 µmol/L [13, 15]. In the NIH trial, without diet, mean tyrosine on 2 mg/day was about 800 µmol/L, approximately ten times normal, and the 2025 meta-analysis estimated a mean increase of 708.77 µmol/L (95% CI 649.32–768.22) [2, 81].

The diet recommended by the PNDS is protein-restricted and controlled in phenylalanine and tyrosine, supervised by a dietitian from a metabolic disease centre. It uses tyrosine-free, phenylalanine-limited amino acid mixtures, in 2 to 3 doses per day, designed for tyrosinaemia types 1 and 2 and suitable for alkaptonuria, including recent glycomacropeptide-based mixtures with better palatability, to maintain total protein intake at population recommendations (0.8 g/kg/day in young adults and 1 g/kg/day in the elderly); low-protein foods are rarely needed in adults; and it adds vitamin supplements, especially vitamin D, and minerals, including calcium [3]. If the patient refuses the mixtures and protein intake falls below recommendations, the benefit of restriction should be reassessed against the reduced risk of keratitis [3]. On a vegan diet, the most likely deficiencies are of iron, vitamin B12, calcium, zinc, iodine and omega-3 fatty acids [3]. Protein substitutes in spheres or tablets have been evaluated in acceptability studies registered on ClinicalTrials.gov [77].

7.5. Safety of nitisinone and management of adverse effects

The dominant adverse effect is keratopathy due to hypertyrosinaemia. In the trials, without dietary prevention, the 4-year incidence was 14.5% with 10 mg/day and 4.9% with 2 mg/day, and no case occurred without nitisinone [3, 76]. In the NIH trial, 3 of 20 treated patients (15%) had keratitis, including irritation, eye pain and photophobia, with permanent discontinuation in one (5%) [15]. In SONIA 2, which only advised reduction of protein intake, 9 of the 69 treated patients (13%) had unilateral or bilateral keratopathy confirmed by slit lamp, and a tenth patient, who could not attend the appointment, was withdrawn because of suggestive ocular symptoms; eight of the nine had symptoms (pain, blurred vision or other signs), and one was detected only at the scheduled examination [35]. The protocol provided for discontinuation of nitisinone, reintroduction at 2 mg/day after resolution and at least 2 months' break, and permanent discontinuation if symptoms recurred; resolution was complete in all nine patients, verified at least 2 months after discontinuation, and of the eight who resumed 2 mg/day, five had recurrence of symptoms and three were asymptomatic at the end of the study [35]. All cases arose in the first 3 years, with no new cases in the 4th year, and ocular adverse events had an incidence of 1.0 per 10 patient-years with nitisinone versus 0.3 in controls (65 events in 25 patients versus 12 in 8) [35]. Serum tyrosine was above 500 µmol/L in all treated patients (median 925 µmol/L at 12 months; range 563–1530), dose reduction to 2 mg lowered it little, and there was no direct relationship between serum tyrosine, measured only at visits, and keratopathy; the authors accept that the causal factor may be the ocular concentration of tyrosine [35]. Overall, keratopathy led to temporary or permanent discontinuation in 14% of treated patients [13]. Patients with keratopathy were more often men (11 of 13), younger (44.6 versus 51.4 years) and had higher plasma HGA; tyrosine at the time of diagnosis of keratopathy was between 609 and 1236 µmol/L, values similar to those of patients without keratopathy [3]. The picture is that of the hypertyrosinaemias: chain-like keratopathy, of pseudodendritic topography, which may be mistaken for herpetic keratitis, initially raised and then ulcerated, unilateral or bilateral, with pain, blurred vision, watering and red eye, but which may be asymptomatic, as in a 22-year-old man on 2 mg/day in whom only routine slit-lamp examination detected it, with mild, asymptomatic recurrence one year after resuming the drug with a stricter diet [3, 82]. Keratopathy is reversible in less than a month when tyrosine is kept below 400 µmol/L; the simplest approach is to stop nitisinone for about a month, during which tyrosine normalises within a few days (below 100 µmol/L) and the cornea regenerates, and then resume treatment, with a stricter diet [3]. An episode of keratitis does not contraindicate reintroduction [3]. The European and US labels recommend slit-lamp examination before initiation and regularly thereafter, at least once a year according to the European SmPC, and immediate examination with tyrosine measurement in the presence of any ocular symptom or tyrosine above 500 µmol/L [13, 15]. The US label also mentions painful palmoplantar hyperkeratotic plaques with tyrosine above 500 µmol/L [15].

Other adverse effects are less frequent. In the NIH trial, 2 of 20 patients (10%) had thrombocytopenia, and in tyrosinaemia type 1 leucopenia (3%), thrombocytopenia (3%) or both (1.5%) have been described, reversible; both labels recommend regular monitoring of platelets and white cells [13, 15]. In SONIA 2, 400 adverse events occurred in 59 of 69 treated patients (86%) and 284 in 57 of 69 controls (83%), with incidences of 2.3 and 2.1 per 10 patient-years; serious events affected 27 treated patients (39%; 57 events) and 26 controls (38%; 52 events), mainly arthroplasties, fractures and other manifestations of the disease, and none was attributed to the drug [35]. The investigators considered 48 events in 18 patients (26%) related to nitisinone; adverse events led to study withdrawal in 9 treated patients (13%) and 1 control (1%) and dose reduction in 8 treated patients (12%) [35]. Infections were more frequent in treated patients (56 events in 27 patients versus 24 in 11), in particular pneumonia and bronchitis (11 events in 9 patients versus 1 in 1), with no known mechanism and no similar signal in the tyrosinaemia type 1 experience; the authors attributed the weight gain to replacement of protein by carbohydrate and fat in patients advised to reduce protein intake [35]. The European SmPC also lists headache, pruritus and rash as common [13]. There were two deaths, both in the treated group (heart failure and myocardial infarction), considered unrelated to the drug [35]. The 2025 meta-analysis found no overall increase in adverse events (relative risk 1.03; 95% CI 0.90–1.19) or serious events (2.47; 95% CI 0.24–25.91; low certainty) [81]. In 63 adults assessed with the Wechsler Adult Intelligence Scale (WAIS-IV) and 74 with the Beck Depression Inventory (BDI-II) over 6 to 7 years on 2 mg/day, there was no cognitive decline or worsening of depression, and tyrosine did not correlate with these measures [83]. Two of the four NAC patients with Parkinson's disease developed it while on nitisinone, and the authors raised the hypothesis of unmasking through inhibition of brain tyrosine hydroxylase by tyrosinaemia; a causal relationship has not been established [3, 63].

Nitisinone is a moderate inhibitor of cytochrome P450 2C9 (CYP2C9) and an inhibitor of the organic anion transporters OAT1 and OAT3. In volunteers, it increased the area under the curve of tolbutamide by 131% and of furosemide by 72%, and decreased that of chlorzoxazone by 27% [15]. The US label recommends halving the dose of sensitive CYP2C9 substrates and avoiding OAT1/OAT3 substrates; the European SmPC recommends careful monitoring and possible adjustment of narrow-therapeutic-index drugs metabolised by CYP2C9, such as warfarin and phenytoin [13, 15]. Serum nitisinone concentration increased over time in treated patients, which the authors attributed to possible autoinhibition of its catabolism by cytochrome P450 [76].

7.6. Evidence of efficacy: trials and observational studies

The following table brings together the available randomised trials and the meta-analysis.

The SONIA 2 secondary endpoints refine the effect. Serum HGA fell by 98.8% relative to controls at 12 months (ratio of geometric means 0.01; 95% CI 0.01–0.02), and the reduction in urinary HGA was maintained at 48 months (ratio 0.005; 0.003–0.008), with means of 1569 versus 33,207 µmol/24 h [35]. The cAKUSSI increased by a mean of 15.1 points in controls and 6.7 in treated patients, a reduction of about 56% in progression, which the authors equated with two arthroplasties or one fracture or rupture, if concentrated in a single component; the version without the pigmentation items (mAKUSSI) progressed less in treated patients, but without statistical significance (−3.6; 95% CI −9.6 to 2.4; p=0.23) [35]. At 48 months, the adjusted difference was −2.5 (−3.9 to −1.0) in the ocular ochronosis score, −0.5 (−0.9 to −0.1) in the ear score, +0.14 (0.00 to 0.28) in the hip bone mineral density T-score and −0.5 (−0.9 to 0.0) in spinal pain; joint pain differed at 12 months (−0.9; −1.6 to −0.1; p=0.023), but not at 48 months (−0.7; −1.6 to 0.1; p=0.10) [35]. There was no difference in aortic velocity (−0.030 m/s; −0.149 to 0.089), in the number of joints or spinal regions with osteoarticular disease, or in the Cobb angles of kyphosis and scoliosis; fractures and ruptures of tendons, ligaments and muscles were numerically less frequent in treated patients, without statistical significance, and quality of life (SF-36) and joint range of motion showed favourable trends [35]. Although the primary endpoint was favourable, the interim analysis of secondary endpoints at 12 months was not considered sufficient for an application for authorisation of the new indication [35]. The authors interpreted the results as arrest of ocular ochronosis and partial reversal of ear ochronosis, starting early and maintained throughout the study, without complete depigmentation [35].

The NAC observational studies, with nitisinone 2 mg/day and dietetic follow-up, showed reduction of ocular and ear ochronosis, with partial reversal documented photographically and by ear cartilage biopsy, reduction of the severity index and of the change in aortic Vmax, and a similar but non-significant pattern in spondyloarthropathy, with no effect on the incidence of arthroplasty (6.5% with nitisinone versus 7.1% without) or on fractures [46–49, 64]. In the modelling comparison between the NAC (2 mg; 37 treated and 23 controls) and SONIA 2 (10 mg; 69 and 69) cohorts, the monthly slope of the severity index was 0.55 and 0.19 points in NAC controls and treated patients and 0.30 and 0.06 in SONIA 2 controls and treated patients [76]. In French practice, a dose of 0.2 mg/day reduced urinary HGA by more than 90% in three patients [55]. In the mouse model, lifelong nitisinone completely prevented pigment deposition in articular cartilage, which in untreated animals appeared at 15 weeks, supporting the hypothesis of prevention with early initiation [84].

The reading of these data has limits that the sources acknowledge. The trials included adults with a mean age close to 48 to 52 years and established disease, and the SONIA 2 primary endpoint was a biomarker; the clinical benefit rests on a composite score (cAKUSSI) that includes components directly modified by HGA reduction, such as pigmentation, and on an absolute difference of 8.6 points over 4 years [35, 85]. The smaller NIH trial was negative on its clinical primary endpoint [33]. The comparison between 2 mg and 10 mg is based on a post hoc analysis of different cohorts, with different dietary intervention, and not on randomisation between doses [76]. The trials were open-label, because urine colour reveals treatment; in SONIA 2, only assessments without patient contact, such as image reading, were blinded, which may have biased pain, quality of life and adverse event reporting, possibly under-reported in the control group [35]. The same trial included a population not selected for osteoporosis, aged from 26 to over 70, had more dropouts than expected (mainly withdrawal of consent in controls and adverse events in treated patients), had neither continuous tyrosine monitoring nor a strict diet, and its duration was limited by the European funding period [35]. The meta-analysis found no effect on joint range of motion, with high certainty [81]. There are no controlled data on starting treatment in adolescents or young adults, nor on hard outcomes such as arthroplasty, valve replacement or fractures [3, 65].

7.7. Osteoarticular treatment, pain and rehabilitation

Treatment of the arthropathy follows, by analogy, the recommendations for osteoarthritis [3]. Non-pharmacological measures are physiotherapy, for assessment, physical treatment and teaching of exercises that relieve pain and improve function; adapted physical activity, with a preference for non-weight-bearing exercise such as swimming, associated with better physical performance, better quality of life and fewer changes in spinal alignment; weight loss in case of excess weight; occupational therapy; and, in young patients with social and occupational impact, rehabilitation in a specialised centre [2, 3, 45]. Patients should avoid physical stress on the spine and large joints, including heavy manual work and impact or contact sports; young people should be guided towards non-contact, low-impact sports and occupations without intense physical exertion [2].

Analgesia follows a therapeutic ladder: paracetamol as the first-line oral analgesic for mild to moderate pain; NSAIDs at the lowest effective dose, avoiding prolonged use, also because of nephrotoxicity in a patient whose kidney is essential for HGA elimination; weak opioids for refractory pain and strong opioids reserved for severe refractory pain, with assessment at a pain centre in case of prolonged use, resistance or dependence; drugs for neuropathic pain in radicular pain; and, as alternatives, transcutaneous electrical nerve stimulation (TENS) and nerve blocks [2, 3]. The sources consulted do not set specific doses for alkaptonuria, and those of the respective SmPCs apply. Intra-articular corticosteroid injections are useful in moderate to severe refractory pain and in effusions; hyaluronic acid injections may be used in the knees and hips, with a later effect [3]. Nitisinone 10 mg/day partially reduced spinal pain in SONIA 2 [3, 13].

Knee, hip or shoulder arthroplasty is indicated when conservative treatment is no longer sufficient, with the main aim of relieving pain rather than increasing range of motion [2, 3]. Prosthesis survival is comparable to that in osteoarthritis, function improves relative to the preoperative state, and surgical site infections, loosening and revisions do not appear more frequent, with follow-up of 3 to 24 years after hip arthroplasty [2, 3, 86]. Surgery is technically more difficult, because the black discolouration of cartilage and periarticular tissues alters anatomical landmarks and the bone is osteoporotic, and blood loss appears greater owing to more inflammatory synovitis [3]. Spinal surgery is less frequent (decompression in 3.4% in the cohorts), in the form of discectomy for radiculopathy or spinal cord decompression [2, 3]. Tendon ruptures are treated according to the usual orthopaedic rules, taking tissue fragility into account [3].

7.8. Osteoporosis and fractures

There are no specific recommendations for alkaptonuria, and those for the general population apply: correction of calcium and vitamin D deficiencies, smoking cessation, physical exercise and fall prevention [3]. Risk assessment considers the usual factors (age, menopause especially before age 40, heredity, corticosteroid therapy, endocrine diseases, smoking, alcohol, sedentary lifestyle) and densitometry, which should be measured at peripheral sites, especially the proximal femur, because disc calcifications overestimate lumbar density and the T-score is not validated before age 50 [3, 46]. The decision on pharmacological treatment is based on densitometry and low-energy fractures, possibly with a rheumatology opinion [3]. There are no trials in alkaptonuria; cases treated with oral and intravenous bisphosphonates, denosumab and teriparatide have been described, the last with a good response and no worsening of calcifications in isolated cases [3, 87, 88]. At the NAC, antiresorptive therapy increased bone mineral density without reducing fragility fractures, and nitisinone did not appear to modify bone loss [46]. In the absence of a specific dose, the doses are those of the osteoporosis labels, with the extrapolation declared, as the table summarises.

7.9. Cardiovascular, urological, renal and endocrine complications

Significant aortic stenosis is treated by valve replacement, according to the usual indications [2, 3]. Surgical replacement has been described in several patients since 1978 [1, 21]. Control of cardiovascular risk factors, especially hypertension and dyslipidaemia, frequent in the NIH cohort, is part of follow-up [3, 61]. Nitisinone may slow the progression of aortic stenosis, but the data are observational or post hoc, with a small absolute effect [49, 60]. Kidney and prostatic stones are treated as in the general population, with surgery when necessary [2, 3]. Kidney function should be protected: avoid prolonged NSAID use, treat obstructions early, control diabetes and hypertension and recognise acute kidney injury as a risk situation [3, 62]. Primary hypothyroidism is treated with levothyroxine; the label indicates, in adults, a full replacement dose of 1.6 µg/kg/day, with a lower starting dose in the elderly and in patients with heart disease, and titration by 12.5–25 µg every 4 to 6 weeks according to TSH [2, 93].

7.10. Metabolic crisis with methaemoglobinaemia and haemolysis

The conventional treatments for methaemoglobinaemia, methylene blue, vitamin C and N-acetylcysteine, have not shown efficacy in patients with alkaptonuria, and the PNDS accepts that they may worsen haemolysis in the presence of HGA; the role of dialysis has not been evaluated [3, 62]. Since 2020 it has been proposed to start nitisinone urgently, 10 mg daily, as the only treatment capable of rapidly reducing plasma HGA [3, 67]. This recommendation is theoretical: since its publication no treated case has been reported, and efficacy has not been confirmed [3]. Supportive care, including treatment of kidney failure, follows general rules. Prevention, by maintaining kidney function and reducing HGA with nitisinone in patients with chronic kidney disease, follows from the same pathophysiological reasoning [3, 62].

7.11. Organ transplantation

Kidney transplantation in a patient with diabetic nephropathy and liver transplantation in a patient with hepatitis B virus cirrhosis markedly reduced HGA excretion, improved ochronosis and slowed progression, because the graft provides HGD activity [3, 39, 94]. The PNDS considers that alkaptonuria is not in itself an indication for transplantation, given the preserved life expectancy and the need to intervene before symptoms; the indications for transplantation in these patients are those of the general population [3].

7.12. Fertility, pregnancy and breastfeeding

There are no published data on fertility, and unpublished French experience does not suggest an impact of the disease [3]. Published pregnancies are rare and there are no studies on the fetal toxicity of maternal HGA or on the effect of pregnancy on the disease; the experience of French centres does not suggest HGA toxicity for the embryo or fetus [3]. In an untreated 34-year-old woman followed at the NAC, urinary bone and cartilage turnover markers increased at the beginning and end of pregnancy, without immediate worsening of the severity index [3, 95]. In a 33-year-old woman treated with nitisinone 0.2 mg/day, continued during pregnancy with a protein intake of about 1.1 g/kg/day, tyrosine remained at 364 ± 104 µmol/L, urinary HGA rose from 1067 to 1456 µmol/mmol creatinine, and zinc and vitamin B12 deficiencies were corrected; the newborns of these two pregnancies were born at term, appropriately grown and with a normal examination at follow-up [3, 22, 55].

Reproductive toxicity data for nitisinone come from animal studies. According to the PNDS, in rabbits nitisinone induced dose-dependent malformations (umbilical hernia, gastroschisis) from 2.5 times the maximum human dose for tyrosinaemia (2 mg/kg/day), and in mice it reduced pup survival and growth at 125 and 25 times that dose [3]. The US label, which uses the 2 mg/day dose as reference, describes incomplete ossification and reduced pup survival in mice from 12 times the maximum dose, maternal toxicity and incomplete ossification in rabbits from 48 times, and reduced litter size and postnatal survival in rats at 486 times, and considers human data insufficient to estimate the risk [15]. The European SmPC advises against use in pregnancy unless the clinical situation requires it [13]. The tyrosinaemia experience suggests that maternal hypertyrosinaemia is the toxic factor: in mothers with tyrosinaemia type 2 and tyrosine of 1300–1650 µmol/L, two newborns had intrauterine growth restriction and microcephaly and two had psychomotor delay (one with epilepsy), whereas six pregnancies with tyrosine below 232 µmol/L had normal children; in mothers with tyrosinaemia type 1 who continued nitisinone, with tyrosine of 600–850 µmol/L, the children were healthy at short-term follow-up [3].

The PNDS recommends stopping nitisinone and the associated protein-restricted diet from the time conception is planned and throughout pregnancy, with the aim of having normal tyrosine levels, because interruption carries no vital or functional risk for the mother (the drug's effect is preventive and long term), because tyrosine on nitisinone may reach high values and because the diet carries a risk of deficiencies [3]. During pregnancy, protein intake should meet recommendations, if necessary with protein or milk powders in women who do not wish to reintroduce meat, with dietetic follow-up by the metabolic team; folic acid 0.4 mg/day is recommended from before conception and during the first trimester, as in any pregnancy, and iron supplementation if necessary [3]. Obstetric follow-up is that of the general population, because pregnancies usually occur in women who are still mildly symptomatic [3]. The position of Saudubray (avoid nitisinone in pregnancy) is concordant; the French very-low-dose series, which continued the drug, is an alternative described in a single case [55, 75].

There is no contraindication to breastfeeding related to HGA [3]. As for nitisinone, the European SmPC prohibits breastfeeding, the US label has no human data and mentions ocular toxicity and lower weight in rat pups exposed through milk, and the PNDS does not recommend use during breastfeeding; the only detailed report, of a mother with tyrosinaemia type 1, described two infants with raised nitisinone and tyrosine at birth that normalised in about two weeks with the mother on treatment, suggesting transplacental passage and absence in milk; if the drug is used, clinical monitoring and measurement of the newborn's tyrosine in the first weeks are required [3, 13, 15].

7.13. Perioperative care and transition

Orthopaedic surgery should anticipate the alteration of anatomical landmarks by black cartilage, osteoporosis and possibly greater blood loss [3]. Intraoperative discovery of black cartilage or bone in a patient without a diagnosis should prompt a request for urinary HGA and metabolic assessment [3, 53]. In a patient with alkaptonuria undergoing surgery, maintaining kidney function (hydration, prevention of nephrotoxicity) reduces the risk of a rise in plasma HGA [3, 62]. The transition from paediatric to adult care should be anticipated and integrated into therapeutic education programmes addressing smoking, drugs, sexuality, fertility, educational and career guidance and, in particular, the choice of occupations without intense physical exertion [2, 3]. In France, registration as long-term condition no. 17 (code E70.2), made by a reference or competence centre, guarantees access to nitisinone and to foods for special medical purposes [3].

7.14. Ineffective and unvalidated therapies and agents to avoid

High-dose ascorbic acid reduces BQA excretion, by up to 99.6% with 250 to 1000 mg/day, but does not alter HGA excretion, and no credible study has shown clinical efficacy; in two infants, it may have doubled urinary HGA, possibly through an effect on immature HPPD [1–3, 96]. The combination of N-acetylcysteine and vitamin C normalised in vitro serum amyloid A protein and inflammatory cytokines in patients' chondrocytes, with no clinical data [3]. In a 48-year-old man, Mayatepek and colleagues tested ascorbic acid (1 and 10 g/day) and a protein-restricted diet (0.3 and 1.3 g/kg/day) for one month each, with reduction of BQA without substantial reduction of HGA; extreme protein restriction would not be acceptable in the long term [1, 78]. Diet alone is covered in section 7.4. Heavy manual work and impact or contact sports should be avoided [2]. NSAIDs should be used at the lowest dose and for short periods [3]. Methylene blue, vitamin C and N-acetylcysteine should not be used to treat methaemoglobinaemia associated with alkaptonuria, because of the risk of worsening haemolysis [3].

7.15. Regulatory status

In Brazil, Orfadin® and Nitikabs® have active registration with ANVISA, but the 2024 professional package insert for Orfadin® indicates nitisinone only in hereditary tyrosinaemia type 1, so prescription in alkaptonuria is off-label use [16, 72]. The regulatory status thus differs across the three jurisdictions: in the EU, Orfadin® has had an indication for adults with alkaptonuria since the Commission decision of 22 October 2020; in the USA, Orfadin® and the other nitisinone formulations retain the indication in tyrosinaemia type 1 alone, and the only product with an indication in alkaptonuria is Harliku™, approved in June 2025; in Brazil, none of the sources consulted shows an approved indication in alkaptonuria [13–16, 97]. The practical implications in Brazil are the absence of alkaptonuria-specific posology, warnings and tyrosine targets in the approved text, the need to document in the clinical record the justification and the patient's informed consent, and access that does not follow from an approved indication, for a medicine manufactured in Sweden and imported by the marketing authorisation holder [16]. The general warnings of the package insert apply to this use: slit-lamp examination before initiation and at least once a year, immediate ophthalmological assessment in the event of visual disturbance, measurement of plasma tyrosine with tightening of tyrosine and phenylalanine restriction above 500 µmol/L, regular monitoring of platelets and white cells, contraindication of breastfeeding, and monitoring of narrow therapeutic index medicines metabolised by CYP2C9, such as warfarin and phenytoin [16]. In a Portuguese case from 2025, nitisinone was proposed and the request for authorisation through the national route (hospital drug and therapeutics committee) was under consideration, which illustrates that European authorisation does not by itself guarantee access [53].

7.16. Research and developments of the past 12 months

The regulatory development of the period is the United States approval of Harliku™ in June 2025, described above [15, 98]. In trials, the post hoc echocardiographic analysis of SONIA 2, published in December 2025, found a small difference of borderline significance in the aortic gradient (level of evidence: exploratory analysis of a randomised trial) [60]. In assessment methods, the NIH published in 2025 a validated vertebral radiographic score for following progression and response to treatment (cohort study) [58]. Paediatric studies remain scarce: a Turkish case-control study from 2026 found no changes in bone remodelling markers in 16 children (observational study) [56]. The trial registries consulted contain no recruiting interventional trials in alkaptonuria, apart from the NIH natural history study, ongoing since 2000 with 300 planned participants [77].

Gene therapy is at the preclinical stage. The strategy under evaluation is somatic gene replacement targeted to the liver, with an adeno-associated viral (AAV) vector of serotype 8 with an AAV2 genome (AAV2/8), carrying the codon-optimised complementary DNA of human HGD under a hepatocyte-specific promoter; the rationale is to restore enzyme activity in the main organ of tyrosine catabolism, eliminating HGA at its source without blocking the pathway upstream, unlike nitisinone [99]. In a study published as a preprint in June 2026, not yet peer-reviewed, Hgd−/− mice received 1 × 10¹² vector genomes per animal; the reporter vectors showed predominantly hepatic, dose-dependent transduction, of 89 to 93% of hepatocytes at the highest dose, and the therapeutic vector produced human HGD transcripts about 33 times higher than those of the endogenous murine gene [99]. Twelve weeks after injection, plasma HGA approached the values of animals without the disease and urinary HGA fell; metabolomics showed a reduction in phase I and phase II metabolites derived from HGA and modulation of the tricarboxylic acid cycle, and the reduction in plasma HGA was comparable to that obtained with nitisinone, without elevation of upstream tyrosine metabolites [99]. There are no results on prevention of ochronosis, durability beyond 12 weeks, immunogenicity, loss of episomal expression with hepatic proliferation or the risk of integration, nor studies in large animals or in humans, and no gene therapy clinical trial is registered [77, 99]. The kidney-specific knockout study, which showed a negligible contribution of the kidney to local tyrosine and HGA metabolism, strengthens the liver as the target [38]. Two other lines are at the preclinical stage, still without data in animals or in patients. The first seeks pharmacological chaperones, molecules that stabilise HGD carrying missense variants, which account for 64.2% of causal variants and readily destabilise the enzyme hexamer: a high-throughput screening assay in Escherichia coli expressing variants of human HGD, validated for robustness, tested 2320 FDA-approved medicines and identified 30 that increased the catalytic activity of the p.(Gly161Arg) variant at least threefold; one of them doubled activity, in a concentration-dependent manner, at 100 and 250 µM, with binding predicted by molecular docking to the active site loop and to the C-terminal β-sheet, and the same assay allows variants to be ranked by residual activity, which is of interest for genotype-phenotype correlation [100]. The second seeks HPPD inhibitors with a better profile than nitisinone: in a bacterial model expressing human HPPD, complemented by docking and molecular dynamics simulations, seven triketones (three commercial herbicides, mesotrione, tembotrione and sulcotrione, and four synthesised analogues) bound at the same site as nitisinone, with coordination of the iron ion of the active centre, and five of them had a longer target residence time than nitisinone; the authors propose residence time as a selection criterion, but the study did not assess induced hypertyrosinaemia, toxicity or in vivo pharmacokinetics [101].

7.17. Divergences between the sources of recommendation

The main recommendations diverge on four points. On dose, the PNDS and the European SmPC use 10 mg/day and the United States label 2 mg/day, and there is experience with 0.2 mg/day [3, 13, 15, 55]. On age, the PNDS extends treatment to 16 years and allows it between 12 and 16 years after a multidisciplinary meeting, whereas the labels restrict it to adults [3, 13, 15]. On diet, the PNDS requires it in every case, with tyrosine targets of 400 to 800 µmol/L, and the labels require it only in the presence of keratopathy, with a target below 500 µmol/L [3, 13, 15]. On cardiac surveillance, the text of the PNDS indicates annual echocardiography in adults, its annexed table indicates it every 2 years from 40 years of age, and GeneReviews recommends it every 1 to 2 years from 40 years of age [2, 3]. In this article, the reference for doses and targets is the 2026 PNDS, as the most recent guideline and the only one dedicated to the disease; the labels are given for comparison [3].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

Alkaptonuria is transmitted in an autosomal recessive manner. The parents of an affected child are, with rare exceptions, obligate heterozygotes and asymptomatic, and each pregnancy of the couple carries a 25% risk of an affected child, 50% of an asymptomatic carrier and 25% of a child neither affected nor a carrier [2, 3]. Testing of the parents is recommended to confirm heterozygosity and the phase of the variants; if only one is a carrier, a de novo variant in the proband, a postzygotic variant in a mosaic parent or uniparental isodisomy should be considered, situations in which the recurrence risk is below 25% [2, 3]. The siblings of the parents have a 50% probability of being carriers [2]. HGA excretion and severity may vary between siblings with the same genotype, so genotype does not allow the clinical course to be predicted [2]. Consanguinity has been noted since Garrod and is frequent in populations with clustered cases; in France, it was found in only 5 of 81 families, and 84% of sibships had a single affected member [8, 21].

The children of a patient are obligate carriers [2, 3]. The risk that a patient will have an affected child depends on whether the partner is a carrier: the PNDS estimates the carrier frequency in the general population at about 1/250 and the resulting risk at about 1/500, higher in consanguineous unions or with partners from populations of high frequency, such as Slovakia or the Dominican Republic [3, 9, 10]. When the partner is tested, a negative sequencing result reduces but does not abolish the risk, because about 3% of variants are deletions or duplications and some are deep intronic [2, 42].

8.2. Testing of family members

Screening of the siblings, older or younger, of an index case is recommended in order to identify early those who may benefit from joint protection measures and, according to the PNDS, from treatment [2, 3]. The test of choice is urinary HGA, which diagnoses or excludes the disease with certainty, without the need for molecular study [3]. Carrier testing requires molecular study with the known familial variants, because urinary HGA in heterozygotes is normal and biochemical testing is not reliable for detecting carriers [2]. The children of a patient may be tested by urinary HGA in childhood if the parents request it, and an affected infant is usually suspected from the nappies [3].

8.3. Reproductive options and testing in minors

Once the two familial variants have been identified, prenatal diagnosis and preimplantation genetic testing are possible [2]. The ideal time to discuss risk and options is before pregnancy, and counselling should be offered to affected, carrier or at-risk young adults [2]. The PNDS does not, as a rule, propose prenatal diagnosis, given the vital and cognitive prognosis and the late onset, and refers such requests for multidisciplinary decision in a prenatal diagnosis centre; GeneReviews acknowledges differences of perspective between professionals and families [2, 3]. Testing of minors has direct clinical benefit when it is diagnostic (urinary HGA), because it allows physical activity and career choice to be guided and, from adolescence, treatment to be considered [2, 3]. A woman with alkaptonuria on nitisinone should plan pregnancy with the metabolic team, so as to stop the medicine before conception (section 7.12) [3].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency in the general population

Alkaptonuria occurs worldwide and in all ethnic groups [2, 3]. Orphanet classifies the worldwide and European point prevalence as 1–9/1,000,000, and records a birth prevalence of 0.25/100,000 in the USA, derived from GeneReviews, and of 0.15/100,000 in France [7]. Incidence in the USA is estimated at between 1:250,000 and 1:1,000,000 live births; the PNDS uses the value of about 1:250,000, and the review by Zatkova gives 1:100,000 to 1:250,000 in most ethnic groups [2, 3, 102]. These estimates are based on published cases and not on population screening, and all sources consider them underestimated owing to late diagnosis and lack of awareness of the disease [2, 3]. At least 1000 patients had been described in the literature by 2021 [2]. In France, the compilation of 96 patients from 81 families, since 1882, led to an incidence of about 1:680,000, with a slight male predominance (54.8%) [8]. In O'Brien's historical series (1963), Czechoslovakia contributed 126 published cases, Germany 108 and the USA 90 [1].

9.2. Founder populations

In Slovakia, the estimated incidence is 1:19,000, the highest described. Urinary screening of more than 611,000 inhabitants, 509,000 of them newborns, started in 1968 at the Institute of Clinical Genetics in Martin, registered 208 patients, 110 of them children, and genealogical studies linked most to mountainous localities in the north-west of the country, settled by shepherds of the Wallachian colonisation, with consanguinity and genetic isolation [9]. Unlike a classic founder effect, 10 different variants were found, of which 6, including the prevalent ones, are thought to have arisen in Slovakia itself; all are located at mutational hotspots, which led to the proposal of an increased mutation rate in the gene, alongside isolation [29]. The four founder variants (c.457dup, c.481G>A, c.808G>A and c.1111dup) account for 80% of Slovak pathogenic alleles [2]. In the Dominican Republic, the high frequency, described since 1960 in 47 cases from eight interrelated families, is due to the founder variant c.360T>G, p.(Cys120Trp) [1, 2, 10]. In Turkey, where consanguinity is frequent, the most common variants in 66 patients were c.175del, c.674G>A and c.1007-2A>T, and c.175del (designated R58fs) has been attributed to an origin in Central Asia with spread to Anatolia and Europe [21, 43]. In Finland, variants of probably recent origin have been described [4]. In Europe outside Slovakia, the most prevalent variant is c.1102A>G, p.(Met368Val) [2]. In the 2016 Italian series, 12 novel variants were identified, including one in a Brazilian patient (c.557T>A) [41].

9.3. Portugal

The sources consulted do not include prevalence estimates for Portugal. The cohort of the Lisbon rheumatology reference centre, published in 2025, included 10 patients (60% women, mean age 54.4 ± 12.2 years), all with skin or scleral pigmentation and 90% with musculoskeletal manifestations, and described capillaroscopic changes in the disease for the first time [52]. Portuguese cases have been published of two sisters with divergent courses depending on kidney function, of a 59-year-old woman with a hip prosthesis and pigmented femoral head and of a 53-year-old woman diagnosed after knee arthroplasty, with biochemical confirmation at the Instituto Nacional de Saúde Doutor Ricardo Jorge [40, 53, 103]. In the French series, two of the families had both parents of Portuguese origin [8]. Access to nitisinone in Portugal derives from the European centralised authorisation, but the 2025 case shows that it depends on case-by-case hospital authorisation [13, 53].

9.4. Brazil and Latin America

In Brazil, the first series characterised by genome sequencing, published in 2026, included five individuals from four families, aged from 8 months to 61 years at first assessment, all with dark urine since childhood and marked HGA excretion. The two consanguineous families from Bahia were homozygous for c.847A>T and c.899T>G; the two non-consanguineous families from São Paulo had c.508G>A in compound heterozygosity with c.899T>G and a deletion of exon 13 in trans with the deep intronic variant c.1007-172A>G [42]. The c.899T>G, p.(Val300Gly) variant occurred in two of the four families [42]. One of the patients also had infertility and a homozygous variant of uncertain significance in STX2, interpreted as a possible second diagnosis [42]. A 2025 Brazilian systematic review summarised the efficacy and safety of nitisinone [81]. Nitisinone has active registration with ANVISA (Orfadin® and Nitikabs®), with an approved indication only in hereditary tyrosinaemia type 1, so treatment of alkaptonuria with nitisinone in Brazil is off-label use [16, 72]. The sources consulted do not include prevalence estimates for Brazil. In the Dominican Republic, the case of a child diagnosed in a resource-limited setting was published in 2025, and in Colombia, in 2026, two patients with a novel variant [104, 105].

9.5. Other populations and sex differences

In China, the review of 91 published patients showed a male:female ratio of about 2:1, dark urine in all, arthropathy in 69.23% and pigmentation in 58.24%, more frequent in adults [51]. Autosomal recessive inheritance implies equal frequency in both sexes, but joint disease is earlier and more severe in men, which may explain a larger number of male diagnoses in clinical series [2, 21, 106]. Keratopathy on nitisinone is also more frequent in men, and hypothyroidism in women [3, 12].

9.6. Care network, registries and organisations

Alkaptonuria falls within the scope of the European Reference Network for hereditary metabolic disorders (MetabERN) [17]. The United Kingdom has a national centre (NAC, Liverpool), which follows patients on nitisinone 2 mg/day with dietetic support; the USA has the NIH natural history protocol, recruiting since 2000; France has reference and competence centres of the G2M network, listed in the PNDS [3, 46, 77]. The European DevelopAKUre consortium conducted the SONIA 1 and SONIA 2 trials [35, 41]. There is an international registry, ApreciseKUre, and gene variant databases [2]. The patient organisations are the AKU Society (United Kingdom), ALCAP (France, sister organisation of the AKU Society) and the AKU Society of North America; the PNDS recommends that teams inform patients of their existence, for the support, information and therapeutic education they provide [2, 3, 18, 19]. The sources consulted did not identify specific organisations in Portugal or Brazil.

10. Illustrative Clinical Case

10.1. Early osteoarthritis diagnosed during knee arthroplasty (Portugal, 2025)

Braga, Teixeira and Matos described a 53-year-old Caucasian woman, born in France and living in Portugal, with a history of dyslipidaemia, hypothyroidism due to Hashimoto's thyroiditis, non-obstructive nephrolithiasis and left tympanoplasty in 1999 [53]. At around age 30, she developed persistent and progressive mechanical arthralgia, first in the knees and then in the lumbar spine, shoulder girdle and ankles, with significant functional disability from age 45 [53]. Over more than a decade, L4-L5 and L5-S1 disc disease with lumbar hyperlordosis (2011), bilateral tricompartmental knee osteoarthritis, bilateral shoulder osteoarthritis and Achilles tendinosis were documented; the patient declined spinal surgery in 2016 and knee arthroplasty in 2020 [53]. A rheumatological assessment in May 2024, with negative or normal rheumatoid factor, anti-CCP antibodies, antinuclear antibodies and inflammatory markers, ended with discharge without a diagnosis [53]. Her mother had a knee prosthesis and her 51-year-old sister had disabling diffuse arthralgia, without a diagnosis; in 2023, the patient lost her job and started antidepressants [53]. In June 2024, she could only walk with two crutches [53].

On 26 March 2025, during total left knee arthroplasty, the surgeon observed black areas in the joint; histology showed black pigmentation in the bone of the tibial plateaus, marked degeneration of hyaline cartilage with fissures, fragmentation and chondrocyte clusters, and chronic synovitis [53]. When questioned afterwards, the patient reported bluish discolouration of the ears noticed since age 25 and dark urine on nappies and sheets in childhood, never given weight [53]. Examination confirmed bluish pigmentation of the sclerae, ear cartilage and tympanic membranes, asymmetrical in the latter (right normal, left bluish), and freshly voided urine darkened after 24 hours in air [53]. The published photographs show the scleral pigmentation (https://pmc.ncbi.nlm.nih.gov/articles/PMC12675957/figure/FIG2/), the ear cartilage pigmentation (https://pmc.ncbi.nlm.nih.gov/articles/PMC12675957/figure/FIG3/) and the urine before and after 24 hours of exposure to air (https://pmc.ncbi.nlm.nih.gov/articles/PMC12675957/figure/FIG5/) [53].

The staging assessment showed a transthoracic echocardiogram without valve thickening, regurgitation or aortic root dilatation; a 4 mm non-obstructive calyceal stone in the left kidney; and densitometry with 1.28 g/cm² at the lumbar spine (Z-score 1.2), 0.658 g/cm² at the femoral neck (Z-score −2.0) and 0.571 g/cm² at the total proximal femur (Z-score −3.2), interpreted as osteoporosis, with normal calcium, phosphate, alkaline phosphatase, vitamin D and parathyroid hormone, in a premenopausal woman [53]. In August 2025, she sustained a left intertrochanteric fracture after a fall from standing height, treated with surgical fixation [53]. In September 2025, at the metabolic disease clinic, urinary HGA by GC-MS at the Instituto Nacional de Saúde Doutor Ricardo Jorge was 7016 µmol/mmol creatinine (reference: undetectable), confirming the diagnosis; the erythrocyte sedimentation rate was 45 mm/h, with normal full blood count, liver and thyroid function, C-reactive protein and protein electrophoresis; molecular study of HGD was ongoing at the time of publication [53]. Treatment with nitisinone was discussed, with the application for authorisation through the national route under consideration, and right knee arthroplasty was proposed [53].

10.2. Lessons from the case

The case brings together the elements that delay diagnosis: the childhood urinary sign and the ear pigmentation from age 25 were not given weight, and two decades of early, disproportionate osteoarthritis were interpreted as common degenerative disease, despite a complete rheumatological assessment [53]. Intraoperative discovery of black cartilage is, as the PNDS notes, one of the two usual circumstances of diagnosis in adults [3]. The densitometry illustrates the recommendation to assess the femur and not the lumbar spine: the positive lumbar Z-score contrasted with the low femoral values, and the low-energy fracture that followed is consistent with the frequency of fragility fractures described in the cohorts [3, 46, 53]. The coexistence of hypothyroidism and nephrolithiasis corresponds to known complications of the disease, although the authors do not establish causality [12, 53]. At age 53, with established arthroplasty and osteoporosis, the expected effect of nitisinone is limited to future progression of ochronosis and of lesions not yet established [3]. The observation of tympanic asymmetry and slow darkening of the urine shows that bedside demonstration takes time and that the definitive test is HGA measurement [2, 53]. The sister with disabling arthralgia is a candidate for screening by urinary HGA, in line with the recommendations for testing siblings [2, 53].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

Alkaptonuria does not, in general, reduce life expectancy, but quality of life may be severely affected by chronic pain and loss of mobility [2, 3]. Disability increases with age and may be severe from the 6th decade; at the extreme, patients become confined to a wheelchair or bed [21, 22]. In the untreated natural history, half of patients have at least one joint prosthesis by age 55, half have a history of kidney stones by age 64, cardiac valve involvement appears at a mean age of 54 and coronary calcification at 59 [2, 11]. The cAKUSSI 2.0 clinical severity index increased, without treatment, by about 4 points per year in 64 patients aged 47.2 ± 10.0 years followed for 4 years [3, 85]. The causes of death described include myocardial infarction, complications of aortic stenosis and crises of haemolysis and methaemoglobinaemia associated with kidney failure, the latter fatal in all published cases [21, 40, 62]. Kidney failure accelerates the disease, and kidney transplantation may slow it [39, 40]. Nitisinone, started in adulthood with established disease, reduces progression of the severity index and partially reverses ochronosis, with no demonstrated effect on joint replacements or fractures; the PNDS expects early initiation, ideally presymptomatic, to modify the prognosis favourably, which has not yet been demonstrated [3, 35, 48].

11.2. Severity score for follow-up

The PNDS proposes cAKUSSI 2.0, an abbreviated version of the index used in SONIA 2, as the clinical follow-up score; an asymptomatic patient has 0 points, there is no defined maximum (162 points plus events), and the maximum score observed in SONIA 2 was 292 [3, 85]. The components, transcribed from the PNDS grid, are shown in the following table.

The flexible score (flex-AKUSSI), without the more costly tests, correlates with the full index, but underestimates progression and overestimates response to treatment when osteoarticular imaging assessment is removed [85]. The NIH spinal radiographic score is an alternative for monitoring the spine, with a minimal detectable change of 2.6 points [58].

11.3. Surveillance

The following table brings together the recommended tests, with the starting age, interval and source. Where the sources diverge, the divergence is indicated.

11.4. Quality of life, work and psychosocial support

Pain and disability have occupational and social consequences. Teams should screen for anxiety and mood disorders and offer psychological support, and the occupational physician can adapt the workplace or support retraining [3]. In France, the PNDS refers to support for disabled workers, the disability compensation allowance and mobility cards; in other countries, the equivalent mechanisms apply [3]. Therapeutic education, including the transition to adult care, and links with patient organisations complete the follow-up [3, 18].

12. Summary and Final Considerations

Alkaptonuria is an autosomal recessive disease caused by biallelic loss-of-function variants in HGD, which encodes homogentisate 1,2-dioxygenase. The accumulation of homogentisic acid, present from birth, manifests in childhood only as urine that darkens and, from late adolescence and especially after age 30, as ochronosis of the sclerae, ears and skin and as spondyloarthropathy and large-joint arthropathy leading to joint replacements at around age 50 to 55, tendon ruptures, fragility fractures, calcific aortic stenosis, renal and prostatic lithiasis and hypothyroidism. Life expectancy is generally preserved, but acute kidney injury may trigger fatal haemolysis and methaemoglobinaemia.

The diagnosis is biochemical: measurement of HGA in a random urine sample, by GC-MS or LC-MS/MS, is pathognomonic and sufficient, and molecular study of HGD, with sequencing and deletion/duplication analysis on transcript NM_000187.4, serves mainly family counselling. Suspicion should arise in the presence of dark urine, spurious proteinuria, early osteoarthritis or spondylopathy with multilevel disc calcifications, black cartilage at surgery or aortic stenosis without risk factors, and the main differential, exogenous ochronosis due to minocycline, hydroquinone or quinacrine, is excluded by normal urinary HGA. Prevalence lies between 1:250,000 and 1:1,000,000, with high-frequency populations in Slovakia and the Dominican Republic; recent series from Portugal and Brazil document late diagnosis and allelic heterogeneity.

Nitisinone, which inhibits HPPD and suppresses HGA formation, is the only targeted treatment. The 2026 French PNDS recommends 10 mg/day from age 16, always with a protein-restricted diet controlled in phenylalanine and tyrosine and tyrosine targets of 400 to 800 µmol/L, accepts treatment between ages 12 and 16 after a multidisciplinary decision and does not recommend it before age 12. In the EU, the label provides for 10 mg/day in adults, and in the USA, since 2025, 2 mg/day in adults. The trials reduced HGA by 88 to 99.7% and, with 10 mg/day, slowed the clinical severity index over 4 years, but were conducted in adults with established disease and showed no effect on established joint lesions, joint replacements or fractures. The main adverse effect is keratopathy due to hypertyrosinaemia, reversible with temporary discontinuation and diet, which requires regular ophthalmological monitoring. Nitisinone should be stopped before conception and during pregnancy and is not recommended during breastfeeding.

Treatment of complications (analgesia, physiotherapy and non-weight-bearing exercise, arthroplasty, valvular and urological surgery, treatment of osteoporosis and hypothyroidism) and scheduled surveillance, guided by the PNDS follow-up table and the cAKUSSI 2.0 index, remain indispensable. The open questions are the ideal age and minimum effective dose for prevention, the safety of prolonged hypertyrosinaemia in adolescents, the urinary HGA value that prevents complications and the clinical translation of hepatic gene replacement approaches, still at the preclinical stage.

13. References

1. OMIM. #203500 Alkaptonuria; AKU. Online Mendelian Inheritance in Man. Baltimore (MD): Johns Hopkins University; last edited 27 Apr 2023. https://www.omim.org/entry/203500
2. Introne WJ, Perry M, Chen M. Alkaptonuria. 2003 May 9 [updated 2021 Jun 10]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1454/
3. Haute Autorité de Santé; Filière de Santé Maladies Rares G2M; Centre de Référence Maladies Héréditaires du Métabolisme (coord. Arnoux JB). Protocole National de Diagnostic et de Soins (PNDS): Alcaptonurie. Saint-Denis La Plaine: HAS; avril 2026 [published 5 May 2026]. https://www.has-sante.fr/jcms/p_3989531
4. OMIM. *607474 Homogentisate 1,2-dioxygenase; HGD. Online Mendelian Inheritance in Man. Baltimore (MD): Johns Hopkins University; last edited 8 May 2015. https://www.omim.org/entry/607474
5. GeneGuidance. HGD — homogentisate 1,2-dioxygenase; alkaptonuria (OMIM 203500, ORPHA:56). GeneGuidance database, version of 26 September 2026. Accessed 6 November 2026.
6. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. REST query for NM_000187.4 on 6 November 2026.
7. Orphanet. Alcaptonúria (Alkaptonuria). ORPHA:56. https://www.orpha.net/en/disease/detail/56 (Orphadata: prevalence and natural history; version 2026-07-02). Accessed 6 November 2026.
8. Aquaron RR. Alkaptonuria in France: past experience and lessons for the future. J Inherit Metab Dis. 2011;34(6):1115-1126. doi:10.1007/s10545-011-9392-7
9. Srsen S, Müller CR, Fregin A, Srsnova K. Alkaptonuria in Slovakia: thirty-two years of research on phenotype and genotype. Mol Genet Metab. 2002;75(4):353-359. doi:10.1016/s1096-7192(02)00002-1
10. Goicoechea De Jorge E, Lorda I, Gallardo ME, Pérez B, Peréz De Ferrán C, Mendoza H, et al. Alkaptonuria in the Dominican Republic: identification of the founder AKU mutation and further evidence of mutation hot spots in the HGO gene. J Med Genet. 2002;39(7):E40. doi:10.1136/jmg.39.7.e40
11. Phornphutkul C, Introne WJ, Perry MB, Bernardini I, Murphey MD, Fitzpatrick DL, et al. Natural history of alkaptonuria. N Engl J Med. 2002;347(26):2111-2121. doi:10.1056/nejmoa021736
12. Avadhanula S, Introne WJ, Auh S, Soldin SJ, Stolze B, Regier D, et al. Assessment of Thyroid Function in Patients With Alkaptonuria. JAMA Netw Open. 2020;3(3):e201357. doi:10.1001/jamanetworkopen.2020.1357
13. European Medicines Agency. Orfadin (nitisinone): EPAR – Product Information (Summary of Product Characteristics). Swedish Orphan Biovitrum International AB; EMEA/H/C/000555. https://www.ema.europa.eu/en/documents/product-information/orfadin-epar-product-information_en.pdf
14. European Commission. Union Register of medicinal products for human use: Orfadin (EU/1/04/303), decision C(2020) 7446 of 22 October 2020 (procedure EMEA/H/C/000555/II/0071). https://ec.europa.eu/health/documents/community-register/html/h303.htm
15. U.S. Food and Drug Administration. HARLIKU (nitisinone) tablets, for oral use. Prescribing information. Cycle Pharmaceuticals Ltd; NDA 209449 [SPL 2025-06-02; efficacy supplement approved 10 Jun 2025]. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=7cf67e24-d46b-4376-bbd0-0b973c406d0d
16. Agência Nacional de Vigilância Sanitária (ANVISA). Orfadin® (nitisinona) cápsulas 2 mg, 5 mg, 10 mg e 20 mg: bula do profissional de saúde. Innovative Medicines Brasil SP – Distribuidora de Medicamentos Ltda.; registration MS 1.3840.0002; file no. 1112435247, published in the Bulário Eletrónico on 14 August 2024. https://consultas.anvisa.gov.br/#/bulario/q/?nomeProduto=orfadin. Accessed 7 November 2026.
17. European Reference Network for Hereditary Metabolic Disorders (MetabERN). https://metab.ern-net.eu/. Accessed 6 November 2026.
18. AKU Society. https://akusociety.org/. Accessed 6 November 2026.
19. Association pour la Lutte Contre l'Alcaptonurie (ALCAP). https://www.alcap.org/. Accessed 6 November 2026.
20. Ranganath LR, Norman BP, Gallagher JA. Ochronotic pigmentation is caused by homogentisic acid and is the key event in alkaptonuria leading to the destructive consequences of the disease-A review. J Inherit Metab Dis. 2019;42(5):776-792. doi:10.1002/jimd.12152
21. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. Chapter 14, Alkaptonuria. p. 109–115.
22. Chakrapani A, Gissen P, McKiernan P. Disorders of Tyrosine Metabolism. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 17, section 17.6 Alkaptonuria. doi:10.1007/978-3-662-63123-2_17
23. Stenn FF, Milgram JW, Lee SL, Weigand RJ, Veis A. Biochemical identification of homogentisic acid pigment in an ochronotic egyptian mummy. Science. 1977;197(4303):566-568. doi:10.1126/science.327549
24. Pollak MR, Chou YH, Cerda JJ, Steinmann B, La Du BN, Seidman JG, et al. Homozygosity mapping of the gene for alkaptonuria to chromosome 3q2. Nat Genet. 1993;5(2):201-204. doi:10.1038/ng1093-201
25. Janocha S, Wolz W, Srsen S, Srsnova K, Montagutelli X, Guénet JL, et al. The human gene for alkaptonuria (AKU) maps to chromosome 3q. Genomics. 1994;19(1):5-8. doi:10.1006/geno.1994.1003
26. Fernández-Cañón JM, Granadino B, Beltrán-Valero de Bernabé D, Renedo M, Fernández-Ruiz E, Peñalva MA, et al. The molecular basis of alkaptonuria. Nat Genet. 1996;14(1):19-24. doi:10.1038/ng0996-19
27. Gehrig A, Schmidt SR, Müller CR, Srsen S, Srsnova K, Kress W. Molecular defects in alkaptonuria. Cytogenet Cell Genet. 1997;76(1-2):14-16. doi:10.1159/000134501
28. Müller CR, Fregin A, Srsen S, Srsnova K, Halliger-Keller B, Felbor U, et al. Allelic heterogeneity of alkaptonuria in Central Europe. Eur J Hum Genet. 1999;7(6):645-651. doi:10.1038/sj.ejhg.5200343
29. Zatková A, de Bernabé DB, Poláková H, Zvarík M, Feráková E, Bosák V, et al. High frequency of alkaptonuria in Slovakia: evidence for the appearance of multiple mutations in HGO involving different mutational hot spots. Am J Hum Genet. 2000;67(5):1333-1339. doi:10.1016/s0002-9297(07)62964-4
30. Rodríguez JM, Timm DE, Titus GP, Beltrán-Valero De Bernabé D, Criado O, Mueller HA, et al. Structural and functional analysis of mutations in alkaptonuria. Hum Mol Genet. 2000;9(15):2341-2350. doi:10.1093/oxfordjournals.hmg.a018927
31. Vilboux T, Kayser M, Introne W, Suwannarat P, Bernardini I, Fischer R, et al. Mutation spectrum of homogentisic acid oxidase (HGD) in alkaptonuria. Hum Mutat. 2009;30(12):1611-1619. doi:10.1002/humu.21120
32. Ascher DB, Spiga O, Sekelska M, Pires DEV, Bernini A, Tiezzi M, et al. Homogentisate 1,2-dioxygenase (HGD) gene variants, their analysis and genotype-phenotype correlations in the largest cohort of patients with AKU. Eur J Hum Genet. 2019;27(6):888-902. doi:10.1038/s41431-019-0354-0
33. Introne WJ, Perry MB, Troendle J, Tsilou E, Kayser MA, Suwannarat P, et al. A 3-year randomized therapeutic trial of nitisinone in alkaptonuria. Mol Genet Metab. 2011;103(4):307-314. doi:10.1016/j.ymgme.2011.04.016
34. Ranganath LR, Milan AM, Hughes AT, Dutton JJ, Fitzgerald R, Briggs MC, et al. Suitability Of Nitisinone In Alkaptonuria 1 (SONIA 1): an international, multicentre, randomised, open-label, no-treatment controlled, parallel-group, dose-response study to investigate the effect of once daily nitisinone on 24-h urinary homogentisic acid excretion in patients with alkaptonuria after 4 weeks of treatment. Ann Rheum Dis. 2016;75(2):362-367. doi:10.1136/annrheumdis-2014-206033
35. Ranganath LR, Psarelli EE, Arnoux JB, Braconi D, Briggs M, Bröijersén A, et al. Efficacy and safety of once-daily nitisinone for patients with alkaptonuria (SONIA 2): an international, multicentre, open-label, randomised controlled trial. Lancet Diabetes Endocrinol. 2020;8(9):762-772. doi:10.1016/s2213-8587(20)30228-x
36. Hughes JH, Wilson PJM, Sutherland H, Judd S, Hughes AT, Milan AM, et al. Dietary restriction of tyrosine and phenylalanine lowers tyrosinemia associated with nitisinone therapy of alkaptonuria. J Inherit Metab Dis. 2020;43(2):259-268. doi:10.1002/jimd.12172
37. Ranganath LR, Milan AM, Hughes AT, Khedr M, Davison AS, Shweihdi E, et al. Homogentisic acid is not only eliminated by glomerular filtration and tubular secretion but also produced in the kidney in alkaptonuria. J Inherit Metab Dis. 2020;43(4):737-747. doi:10.1002/jimd.12181
38. Rutland DA, Norman BP, Hughes JH, Wilson PJM, Sutherland H, Brown RLW, et al. Specific knockout of kidney homogentisate 1,2-dioxygenase reveals that local metabolism of tyrosine and homogentisic acid is negligible in alkaptonuria. Hum Mol Genet. 2026;35(8):ddag036. doi:10.1093/hmg/ddag036
39. Introne WJ, Phornphutkul C, Bernardini I, McLaughlin K, Fitzpatrick D, Gahl WA. Exacerbation of the ochronosis of alkaptonuria due to renal insufficiency and improvement after renal transplantation. Mol Genet Metab. 2002;77(1-2):136-142. doi:10.1016/s1096-7192(02)00121-x
40. Faria B, Vidinha J, Pêgo C, Correia H, Sousa T. Impact of chronic kidney disease on the natural history of alkaptonuria. Clin Kidney J. 2012;5(4):352-355. doi:10.1093/ckj/sfs079
41. Nemethova M, Radvanszky J, Kadasi L, Ascher DB, Pires DE, Blundell TL, et al. Twelve novel HGD gene variants identified in 99 alkaptonuria patients: focus on 'black bone disease' in Italy. Eur J Hum Genet. 2016;24(1):66-72. doi:10.1038/ejhg.2015.60
42. Moreno CA, de Oliveira Sobrinho RP, Jorente J, Appenzeller S, Meira NGS, Costa LCC, et al. Molecular characterization of Alkaptonuria in Brazilian patients. Mol Genet Metab Rep. 2026;46:101288. doi:10.1016/j.ymgmr.2025.101288
43. Kisa PT, Gunduz M, Dorum S, Uzun OU, Cakar NE, Yildirim GK, et al. Alkaptonuria in Turkey: Clinical and molecular characteristics of 66 patients. Eur J Med Genet. 2021;64(5):104197. doi:10.1016/j.ejmg.2021.104197
44. Beltrán-Valero de Bernabé D, Granadino B, Chiarelli I, Porfirio B, Mayatepek E, Aquaron R, et al. Mutation and polymorphism analysis of the human homogentisate 1, 2-dioxygenase gene in alkaptonuria patients. Am J Hum Genet. 1998;62(4):776-784. doi:10.1086/301805
45. Perry MB, Suwannarat P, Furst GP, Gahl WA, Gerber LH. Musculoskeletal findings and disability in alkaptonuria. J Rheumatol. 2006;33(11):2280-2285.
46. Ranganath LR, Khedr M, Vinjamuri S, Gallagher JA. Frequency, diagnosis, pathogenesis and management of osteoporosis in alkaptonuria: data analysis from the UK National Alkaptonuria Centre. Osteoporos Int. 2021;32(5):927-938. doi:10.1007/s00198-020-05671-y
47. Ranganath LR, Khedr M, Vinjamuri S, Gallagher JA. Characterizing the alkaptonuria joint and spine phenotype and assessing the effect of homogentisic acid lowering therapy in a large cohort of 87 patients. J Inherit Metab Dis. 2021;44(3):666-676. doi:10.1002/jimd.12363
48. Ranganath LR, Gallagher JA, Davidson J, Vinjamuri S. Characterising the arthroplasty in spondyloarthropathy in a large cohort of eighty-seven patients with alkaptonuria. J Inherit Metab Dis. 2021;44(3):656-665. doi:10.1002/jimd.12340
49. Ranganath LR, Heseltine T, Khedr M, Fisher MF. Evaluating the aortic stenosis phenotype before and after the effect of homogentisic acid lowering therapy: Analysis of a large cohort of eighty-one alkaptonuria patients. Mol Genet Metab. 2021;133(3):324-331. doi:10.1016/j.ymgme.2021.05.007
50. Imrich R, Sedláková J, Úlehlová M, Gornall M, Jackson R, Olsson B, et al. Radiological evolution of spinal disease in alkaptonuria and the effect of nitisinone. RMD Open. 2022;8(2):e002422. doi:10.1136/rmdopen-2022-002422
51. Qiu X, Liu Y, Shao Y, Mao X, Zhang J, Chen X, et al. Clinical and molecular characteristics of Chinese patients with alkaptonuria: 4-year follow-up of a pediatric patient and literature review. Mol Genet Metab Rep. 2025;45:101276. doi:10.1016/j.ymgmr.2025.101276
52. Pereira da Costa R, Costa F, Fonseca JE, Khmelinskii N, Oliveira A, Dourado E. Rheumatic manifestations of alkaptonuria: clinical, genetic and capillaroscopic characterisation of a referral centre cohort reveals new capillaroscopic marker of the disease. Semin Arthritis Rheum. 2025;75:152856. doi:10.1016/j.semarthrit.2025.152856
53. Braga R, Teixeira F, Matos C. Early-Onset Degenerative Bone Changes as a Manifestation of Alkaptonuria: A Case Report. Cureus. 2025;17(11):e96028. doi:10.7759/cureus.96028
54. Zibolen M, Srsnova K, Srsen S. Increased urolithiasis in patients with alkaptonuria in childhood. Clin Genet. 2000;58(1):79-80. doi:10.1034/j.1399-0004.2000.580115.x
55. Sloboda N, Wiedemann A, Merten M, Alqahtani A, Jeannesson E, Blum A, et al. Efficacy of low dose nitisinone in the management of alkaptonuria. Mol Genet Metab. 2019;127(3):184-190. doi:10.1016/j.ymgme.2019.06.006
56. Enver EÖ, Soylu Eİ, Vatansever P, Zubarioglu T, Kıykım E, Akın Y, et al. Bone turnover markers in pediatric alkaptonuria: a case-control study. J Pediatr Endocrinol Metab. 2026. doi:10.1515/jpem-2026-0502
57. Lindner M, Bertelmann T. On the ocular findings in ochronosis: a systematic review of literature. BMC Ophthalmol. 2014;14:12. doi:10.1186/1471-2415-14-12
58. Rossignol F, Pan KS, Perry MB, Bryant JC, O'Brien KJ, Castillo IJ, et al. Development of a radiographic vertebral severity score for evaluation of disease progression in alkaptonuria. Mol Genet Metab. 2025;146(3):109251. doi:10.1016/j.ymgme.2025.109251
59. Hannoush H, Introne WJ, Chen MY, Lee SJ, O'Brien K, Suwannarat P, et al. Aortic stenosis and vascular calcifications in alkaptonuria. Mol Genet Metab. 2012;105(2):198-202. doi:10.1016/j.ymgme.2011.10.017
60. Bruce C, Meenamkuzhy-Hariharan P, Hussain S, Eleuteri A, Ranganath L, Fisher M, et al. Effect of Nitisinone on Aortic Stenosis Disease Progression in Patients With Alkaptonuria: An Analysis of the Suitability of Nitisinone in Alkaptonuria (SONIA) 2 Study. Cureus. 2025;17(12):e100123. doi:10.7759/cureus.100123
61. Thimmapuram R, Bandettini WP, Shanbhag SM, Yu JH, O'Brien KJ, Gahl WA, et al. Aortic distensibility in alkaptonuria. Mol Genet Metab. 2020;130(4):289-296. doi:10.1016/j.ymgme.2020.05.006
62. Davison AS, Milan AM, Gallagher JA, Ranganath LR. Acute fatal metabolic complications in alkaptonuria. J Inherit Metab Dis. 2016;39(2):203-210. doi:10.1007/s10545-015-9902-0
63. Ranganath L, Khedr M, Milan AM, Davison AS, Norman BP, Janssen MCH, et al. Increased prevalence of Parkinson's disease in alkaptonuria. JIMD Rep. 2023;64(4):282-292. doi:10.1002/jmd2.12367
64. Ranganath LR, Milan AM, Hughes AT, Khedr M, Davison AS, Wilson PJ, et al. Reversal of ochronotic pigmentation in alkaptonuria following nitisinone therapy: Analysis of data from the United Kingdom National Alkaptonuria Centre. JIMD Rep. 2020;55(1):75-87. doi:10.1002/jmd2.12137
65. Wolffenbuttel BHR, Heiner-Fokkema MR, van Spronsen FJ. Preventive use of nitisinone in alkaptonuria. Orphanet J Rare Dis. 2021;16(1):343. doi:10.1186/s13023-021-01977-0
66. Loken PR, Magera MJ, Introne W, Tortorelli S, Gavrilov D, Oglesbee D, et al. Homogentisic acid interference in routine urine creatinine determination. Mol Genet Metab. 2010;100(1):103-104. doi:10.1016/j.ymgme.2010.01.006
67. Davison AS, Luangrath E, Selvi E, Ranganath LR. Fatal acute haemolysis and methaemoglobinaemia in a man with renal failure and Alkaptonuria - Is nitisinone the solution?. Mol Genet Metab Rep. 2020;23:100588. doi:10.1016/j.ymgmr.2020.100588
68. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 3, Diagnostic Procedures, section 3.1.
69. Curtis SL, Roberts NB, Ranganath LR. Interferences of homogentisic acid (HGA) on routine clinical chemistry assays in serum and urine and the implications for biochemical monitoring of patients with alkaptonuria. Clin Biochem. 2014;47(7-8):640-647. doi:10.1016/j.clinbiochem.2013.12.016
70. Suwannarat P, Phornphutkul C, Bernardini I, Turner M, Gahl WA. Minocycline-induced hyperpigmentation masquerading as alkaptonuria in individuals with joint pain. Arthritis Rheum. 2004;50(11):3698-3701. doi:10.1002/art.20606
71. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, Clinical Approach to Inborn Errors of Metabolism in Paediatrics, sections 1.6.9 and 1.6.15.
72. Agência Nacional de Vigilância Sanitária (ANVISA). Open data on registered medicines: nitisinone (Orfadin®, Nitikabs®). https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv. Accessed 6 November 2026.
73. INFARMED. Infomed — database of medicinal products for human use: nitisinone (Orfadin®; Nitisinona Dipharma®, procedure DK/H/2726/003/E01). https://extranet.infarmed.pt/INFOMED-fo/. Accessed 7 November 2026.
74. INFARMED. Nitisinona Dipharma 5 mg e 10 mg cápsulas: resumo das características do medicamento. Dipharma Arzneimittel GmbH. Infomed, med_id 634862. https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=634862&tipo_doc=rcm. Accessed 7 November 2026.
75. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 45, Medications Used in the Treatment of Inborn Errors of Metabolism (nitisinona).
76. Ranganath LR, Milan AM, Hughes AT, Khedr M, Norman BP, Alsbou M, et al. Comparing nitisinone 2 mg and 10 mg in the treatment of alkaptonuria-An approach using statistical modelling. JIMD Rep. 2022;63(1):80-92. doi:10.1002/jmd2.12261
77. ClinicalTrials.gov. Registered studies for "alkaptonuria" (API v2). https://clinicaltrials.gov/search?cond=alkaptonuria. Accessed 6 November 2026.
78. Mayatepek E, Kallas K, Anninos A, Müller E. Effects of ascorbic acid and low-protein diet in alkaptonuria. Eur J Pediatr. 1998;157(10):867-868. doi:10.1007/s004310050956
79. de Haas V, Carbasius Weber EC, de Klerk JB, Bakker HD, Smit GP, Huijbers WA, et al. The success of dietary protein restriction in alkaptonuria patients is age-dependent. J Inherit Metab Dis. 1998;21(8):791-798. doi:10.1023/a:1005410416482
80. Ranganath LR, Khedr M, Milan AM, Davison AS, Hughes AT, Norman BP, et al. Anthropometric, Body Composition, and Nutritional Indicators with and without Nutritional Intervention during Nitisinone Therapy in Alkaptonuria. Nutrients. 2024;16(16):2722. doi:10.3390/nu16162722
81. Mayrink FD, Dorneles G, da Silva IM, Areda CA. Efficacy and safety of Nitisinone for patients with alkaptonuria: A systematic review with metanalysis. Mol Genet Metab. 2025;145(1):109099. doi:10.1016/j.ymgme.2025.109099
82. Khedr M, Judd S, Briggs MC, Hughes AT, Milan AM, Stewart RMK, et al. Asymptomatic Corneal Keratopathy Secondary to Hypertyrosinaemia Following Low Dose Nitisinone and a Literature Review of Tyrosine Keratopathy in Alkaptonuria. JIMD Rep. 2018;40:31-37. doi:10.1007/8904_2017_62
83. Davison AS, Hughes G, Harrold JA, Clarke P, Griffin R, Ranganath LR. Long-term low dose nitisinone therapy in adults with alkaptonuria shows no cognitive decline or increased severity of depression. JIMD Rep. 2022;63(3):221-230. doi:10.1002/jmd2.12272
84. Preston AJ, Keenan CM, Sutherland H, Wilson PJ, Wlodarski B, Taylor AM, et al. Ochronotic osteoarthropathy in a mouse model of alkaptonuria, and its inhibition by nitisinone. Ann Rheum Dis. 2014;73(1):284-289. doi:10.1136/annrheumdis-2012-202878
85. Cant HEO, Chatzidaki I, Olsson B, Rudebeck M, Arnoux JB, Imrich R, et al. Improving the clinical accuracy and flexibility of the Alkaptonuria severity score index. JIMD Rep. 2022;63(4):361-370. doi:10.1002/jmd2.12290
86. Pachore JA, Shah VI, Upadhyay S, Shah K, Sheth A, Kshatriya A. Primary hip arthroplasty for the treatment of alkaptonuric hip arthritis: 3- to 24-year follow-ups. Arthroplasty. 2019;1(1):8. doi:10.1186/s42836-019-0010-8
87. Ranganath LR, Khedr M, Mistry A, Vinjamuri S, Gallagher JA. Treatment of osteoporotic fractures in alkaptonuria by teriparatide stimulates bone formation and decreases fracture rate - A report of two cases. Bone Rep. 2021;15:101151. doi:10.1016/j.bonr.2021.101151
88. Ebrahim IC, Hoang TD, Vietor NO, Schacht JP, Shakir MKM. Dilemmas in the diagnosis and management of osteoporosis in a patient with alkaptonuria: Successful treatment with teriparatide. Clin Case Rep. 2022;10(12):e6729. doi:10.1002/ccr3.6729
89. U.S. Food and Drug Administration. RECLAST (zoledronic acid) injection. Prescribing information. Novartis; SPL 2026-02-03. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=5a9b3737-9ce6-4a89-b76e-6aab79eba9cf
90. U.S. Food and Drug Administration. FOSAMAX PLUS D (alendronate sodium/cholecalciferol) tablets. Prescribing information. Organon; SPL 2026-02-03. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=10307e7e-9a84-4aa1-8c5c-4b209cffe4d1
91. U.S. Food and Drug Administration. PROLIA (denosumab) injection. Prescribing information. Amgen; SPL 2026-06-26. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=49e5afe9-a0c7-40c4-af9f-f287a80c5c88
92. U.S. Food and Drug Administration. FORTEO (teriparatide) injection. Prescribing information. Eli Lilly; SPL 2026-08-03. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=aae667c5-381f-4f92-93df-2ed6158d07b0
93. U.S. Food and Drug Administration. SYNTHROID (levothyroxine sodium) tablets. Prescribing information. AbbVie; SPL 2024-02-20. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=1e11ad30-1041-4520-10b0-8f9d30d30fcc
94. Kobak AC, Oder G, Kobak S, Argin M, Inal V. Ochronotic arthropathy: disappearance of alkaptonuria after liver transplantation for hepatitis B-related cirrhosis. J Clin Rheumatol. 2005;11(6):323-325. doi:10.1097/01.rhu.0000191157.25894.55
95. Ranganath LR, Milan AM, Bay-Jensen AC, Thudium CS. A case report of pregnancy in untreated alkaptonuria - Focus on urinary tissue remodelling markers. Mol Genet Metab Rep. 2021;27:100766. doi:10.1016/j.ymgmr.2021.100766
96. Wolff JA, Barshop B, Nyhan WL, Leslie J, Seegmiller JE, Gruber H, et al. Effects of ascorbic acid in alkaptonuria: alterations in benzoquinone acetic acid and an ontogenic effect in infancy. Pediatr Res. 1989;26(2):140-144. doi:10.1203/00006450-198908000-00015
97. U.S. Food and Drug Administration. Drugs@FDA: nitisinone (NDA 021232 Orfadin; NDA 206356; NDA 209449 Nityr/Harliku; ANDA 211041, 212390, 216201). Data via openFDA drugsfda. https://www.accessdata.fda.gov/scripts/cder/daf/. Accessed 6 November 2026.
98. Chandani HK, Hujjat SFZ, Chandani DK, Khan MS, Mahmoud A. Harliku (nitisinone): first FDA-approved disease-modifying therapy for alkaptonuria. Ann Med Surg (Lond). 2025;87(12):7877-7879. doi:10.1097/ms9.0000000000004033
99. Lequeue S, Norman BP, Del’haye GG, Neuckermans J, Colemonts-Vroninks H, Hughes JH, et al. Liver-directed AAV gene therapy metabolically corrects AKU in Hgd deficient mice. bioRxiv. 2026. doi:10.64898/2026.06.01.729174
100. Lequeue S, Allach H, Nulmans I, Desmet L, Salvi NS, Rombaut M, et al. A robust bacterial high-throughput screening assay to identify pharmacological chaperones targeting human homogentisate 1,2-dioxygenase missense variants in alkaptonuria. Eur J Pharmacol. 2025;1005:178048. doi:10.1016/j.ejphar.2025.178048
101. Bernardini G, Trezza A, Petricci E, Romagnoli G, Zambardino D, Manetti F, et al. A Comprehensive In Vitro and In Silico Approach for Targeting 4-Hydroxyphenyl Pyruvate Dioxygenase: Towards New Therapeutics for Alkaptonuria. Int J Mol Sci. 2025;26(7):3181. doi:10.3390/ijms26073181
102. Zatkova A. An update on molecular genetics of Alkaptonuria (AKU). J Inherit Metab Dis. 2011;34(6):1127-1136. doi:10.1007/s10545-011-9363-z
103. Moniz R, Araújo F, Viana G. Femoral head ochronotic pigmentation in a patient with alkaptonuria and secondary hip osteoarthritis. ARP Rheumatol. 2025;4(4):311-312. doi:10.63032/mymi3386
104. Ortiz Hernández IY, García Grullon C, Rijo Florimon K, Bueno Fernandez LA, Espaillat Solano L. A Rare Diagnosis in a Resource-Limited Setting: Alkaptonuria in a Young Dominican Child. Cureus. 2025;17(9):e92939. doi:10.7759/cureus.92939
105. León-Sanabria MC, Zarante-Bahamón AM. Alkaptonuria in two Colombian patients: identification of HGD variants including a novel finding. BMC Med Genomics. 2026;19(1):79. doi:10.1186/s12920-026-02344-3
106. Bernardini G, Braconi D, Zatkova A, Sireau N, Kujawa MJ, Introne WJ, et al. Alkaptonuria. Nat Rev Dis Primers. 2024;10(1):16. doi:10.1038/s41572-024-00498-x