Allopurinol Desensitization: Is There Still a Place in Clinical Practice?

Allopurinol was the first xanthine oxidase inhibitor (XOI) approved by the US Food and Drug Administration in 1966, and has since become a cornerstone of urate‐lowering therapy (ULT) for primary and secondary hyperuricemia worldwide. Allopurinol inhibits xanthine oxidase, thus reducing the conversion of hypoxanthine and xanthine to uric acid, which thereby lowers serum uric acid (SUA) and reduces urinary uric acid excretion. Its active metabolite, oxypurinol, also inhibits xanthine oxidase and has similar effects. Allopurinol is also used to prevent or treat hyperuricemia associated with tumor lysis syndrome in patients receiving cytotoxic therapy, and in those with nephrolithiasis and hyperuricosuria [1]. Beyond urate lowering, allopurinol has been reported to reduce oxidative stress and improve endothelial dysfunction, suggesting possible roles in slowing chronic kidney disease progression and reducing cardiovascular events. Recent meta‐analyses have shown mixed results, suggesting potential reductions in cardiovascular morbidity but no consistent benefit for renal outcomes [2, 3]. However, these findings should be interpreted with caution, due to variability in study quality and substantial heterogeneity among the included studies.

Although allopurinol is often recommended as first‐line ULT for gout by several rheumatology organizations, including the American College of Rheumatology (ACR) [4], European Alliance of Associations for Rheumatology (EULAR) [5], and Asia Pacific League of Associations for Rheumatology (APLAR) [6]; others also recommended a uricosuric agent as first‐line ULT treatment, including the Japanese Society of Gout and Nucleic Acid Metabolism [7], and Chinese Society of Endocrinology [8], which alone, suggested guiding its use by renal urate excretion. These guidelines adopt a treat‐to‐target approach, which aims to maintain SUA below its saturation point (< 6.0 mg/dL) in order to prevent flares and promote dissolution of monosodium urate (MSU) crystal deposits.

In clinical practice, selection of ULT for the treatment of gout should be individualized and guided by current clinical guidelines. Assessment of urinary uric acid excretion and identification of underexcretion or overexcretion phenotypes may be an additional consideration in selected patients, particularly when uricosuric therapy is being considered. Allopurinol has several practical advantages, as it can be used in patients with either under‐ or over‐excretion of uric acid, and those with renal impairment and a history of renal calculi. Its use is limited by the development of allopurinol hypersensitivity reactions, ranging from mild gastrointestinal symptoms and mild‐to‐moderate cutaneous eruptions (e.g., maculopapular rash, angioedema or urticaria, and fixed drug eruption) to rare but potentially severe and life‐threatening reactions; namely allopurinol hypersensitivity syndrome (AHS) or severe cutaneous adverse reactions (SCARs) that include Stevens‐Johnson syndrome (SJS), toxic epidermal necrolysis (TEN) and drug reaction with eosinophilia and systemic symptoms (DRESS). AHS and SCARs are strongly associated with HLA‐B*58:01, which is more prevalent in East and Southeast Asian populations than in people of European ancestry [9]. Despite several guidelines recommending allopurinol as the first‐line ULT, only a few of them recommend HLA‐B*58:01 screening before allopurinol initiation. The ACR conditionally recommends testing for patients of Southeast Asian descent (e.g., Han Chinese, Korean, Thai) and for African American patients, and those who test positive are advised to use alternative XOIs (e.g., febuxostat or topiroxostat) or consider uricosuric therapy [4]. The APLAR weakly recommends HLA‐B*58:01 screening, even in populations with ≥ 5% of this gene frequency, and genetic testing should be performed only when it is cost‐effective in that population, and when inexpensive alternative medications are available for patients who tested positive [6]. The Hong Kong Society of Rheumatology recommends that HLA‐B*58:01 screening should be considered for patients of Asian descent (e.g., Han, Chinese, Korean, Thai), African American patients, and patients with risk factors for AHS, including those ≥ 60 years or with renal insufficiency (CKD stage ≥ 3), and allopurinol should be avoided in patients who tested positive [10].

The rationale for HLA‐B*58:01 screening before allopurinol initiation should probably be determined through cost‐effectiveness analysis. Interestingly, cost‐effectiveness analyses have shown conflicting results of HLA‐B*58:01 screening before allopurinol initiation to prevent severe AHS in East and Southeast Asia, a region with a relatively high prevalence of this allele (Table S1). Such differences might have been related to several factors including the socioeconomic status and prevalence of HLA‐B*58:01 in its populations, as well as the costs of genotyping and medical care, availability and cost of alternative ULTs, analytic model used and chosen time horizon. However, the decision to perform routine HLA‐B*58:01 testing should depend not only on cost‐effectiveness analysis, but also ethical issues because this adverse event is a serious medical condition, and a balance between payers, patients, and physicians should also be considered.

Recent reports indicate that the global prevalence of gout is increasing, particularly in Asia, Latin America and Africa, which are regions that include many low‐ and middle‐income countries (LMICs) and other resource‐limited settings (RLS) [11]. Implementing a screening‐first approach in these settings is challenging because HLA‐B*58:01 testing may be unavailable or costly, particularly outside tertiary centers. In these LMICs and RLS, allopurinol may be the only available XOI, while other XOIs (e.g., febuxostat or topiroxostat) and uricase‐based therapy (pegloticase) may be unavailable, inaccessible or unaffordable. Uricosuric agents such as probenecid, sulfinpyrazone, and benzbromarone may be available, but their use is often constrained because they generally require adequate renal function, are most appropriate for patients with uric acid under‐excretion, and are typically avoided in patients with a history of renal calculi. Consequently, as gout prevalence rises and access to HLA‐B*58:01 testing remains limited, allopurinol use is expected to increase, and cases of allopurinol hypersensitivity may be encountered more frequently. A common clinical scenario is a patient with gout who has moderate renal impairment and/or history of renal calculi, who develops a cutaneous reaction soon after starting allopurinol in a setting where HLA‐B*58:01 testing is unavailable or unaffordable, yet they still require long‐term XOI therapy. This raises a practical clinical question: should the patient be referred to a higher‐level facility for alternative therapies, or, when referral is not feasible, does allopurinol desensitization still have a role to play?

A report from the APLAR Gout Special Interest Group registry indicated that allopurinol, febuxostat, and uricosurics were used in 63%, 42%, and 3% of patients, respectively [12]. This was observed even though several participating countries are located in East and Southeast Asia, in which HLA‐B*58:01 prevalence is relatively high [9]. The very low use of uricosurics in the Asia‐Pacific region raises questions about how treatment aligns with gout pathophysiology, given that many gout patients have renal under‐excretion of uric acid. This pattern probably reflects physician familiarity, prescribing habits, and the positioning of XOIs as first‐line therapy in major guidelines [4, 5, 6].

The rationale for desensitization is gradual re‐exposure to allopurinol to promote immunological tolerance while minimizing the risk of recurrent hypersensitivity. The first allopurinol desensitization regimen was developed by Meyrier in 1976 [13], and was subsequently modified by several groups, using both oral and intravenous desensitization regimens. These modified regimens vary in starting dose, dose increments, and the interval between dose increases. Oral desensitization can be categorized as slow (~1.5–3 months), standard (~4 weeks), rapid (~2 weeks), and rush (~1 week) desensitization, based on the time required to reach the target dose, while intravenous desensitization can be accomplished within 1 day (Tables S2 and S3).

Among the reported regimens, Fam's regimen is one of the most widely used [14]. However, evidence supporting efficacy remains limited, because most reports are single cases or small case series, and most patients had mild‐to‐moderate allopurinol hypersensitivity reactions (e.g., maculopapular rash, angioedema or urticaria, and fixed drug eruption). Allopurinol desensitization was successfully performed in 78% of a series of 32 patients with mild‐to‐moderate allopurinol hypersensitivity reactions [15]. On this basis, desensitization should generally be considered for patients with mild‐to‐moderate allopurinol hypersensitivity reactions [10], after careful risk–benefit assessment and a thorough discussion with the patient. If a patient develops a minor reaction (e.g., limited cutaneous symptoms) during desensitization, allopurinol should be stopped and the reaction managed appropriately; once symptoms have been resolved, desensitization can be restarted at the next lower dose level [14, 16]. Intravenous desensitization should be undertaken with extreme caution and performed only in hospital settings with immediate access to resuscitation equipment and appropriately trained staff.

Severe AHS is generally considered a contraindication for allopurinol desensitization. However, successful allopurinol desensitization has been reported in cases with severe AHS [17]. There has also been a report of successful desensitization at the second attempt, performed approximately 4 years after an initial failure due to severe AHS during the first desensitization [18]. These cases are extraordinarily exceptional and should not be considered as routine efficacy; thus, allopurinol desensitization should be performed in a situation when allopurinol use is inevitable. The author experienced success in performing allopurinol desensitization in three patients with severe AHS, in circumstances when allopurinol was the only available option at the author's institution. Such cases highlight both the clinical pressure that can exist in RLS and the need for extreme caution when considering desensitization outside the usual indications.

Prevention is, of course, preferable to treatment. In addition to HLA‐B*58:01, older age, concomitant diuretic use, renal impairment and a high starting dose of allopurinol are associated with increased risk of AHS [19]. A starting dose of ≥ 1.5 mg/eGFR (mL/min/1.73 m²) has been associated with AHS; therefore, allopurinol should be initiated at < 1.5 mg/eGFR [20]. The dose should then be gradually increased (e.g., monthly) until the target SUA is attained, or the maximum tolerated dose is reached (“start‐low, go‐slow”). Monthly increments of 50 mg and 100 mg in patients with eGFR < 60 and ≥ 60 mL/min/1.73 m², respectively, have been reported to be safe without AHS [21].

Another emerging preventive approach is the use of non‐genetic clinical prediction models that combine patient characteristics with routine laboratory data to estimate the probability of severe AHS. The first model was developed in Thailand, potentially representing East and Southeast Asian populations, and it categorizes the risk of severe AHS as low (< 1%), moderate (1%–2%) and high (≥ 2%) [22]. The second model was developed using UK National Health Service data, in which most patients were White (with low HLA‐B*58:01 prevalence) and the numbers of Chinese and other Asian ethnicities were smaller [23]. External validation is needed to confirm model performance in regions with a high prevalence of HLA‐B*58:01.

In conclusion, allopurinol desensitization plays a limited role, but it remains clinically useful, particularly in LMICs and RLS where allopurinol may be the only available ULT. Desensitization should be considered only for patients with mild‐to‐moderate hypersensitivity reactions and it is contraindicated for patients with severe AHS. Allopurinol desensitization in severe AHS patients should be reserved for only exceptional circumstances and performed in hospital with close monitoring and immediate access to resuscitation equipment. Where desensitization is pursued, standard oral protocols are usually preferred, whereas rapid, rush, or intravenous approaches should be avoided unless there is an urgent clinical need and adequate facilities are available. Finally, the key question is not whether desensitization works, but when it is appropriate. It can be a pragmatic option when treatment choices are limited, but it should not replace prevention (risk stratification, an appropriate starting dose, and a “start‐low, go‐slow” strategy).