Decision Driven Factors for Allopurinol Dosage in Tumor Lysis Syndrome Prophylaxis: The European Experience of the Florence Pivotal Study

Allopurinol for tumor lysis syndrome (TLS) prophylaxis is mostly given at 300 mg daily, but the exact dose is variable and depends on physician’s choice. In order to assess which baseline factors mostly drive the choice, we performed an analysis of the FLORENCE study database. FLORENCE is a double-blinded randomized trial, the largest one in TLS prophylaxis, conducted in 11 European countries (Croatia, Czech Republic, Germany, Hungary, Italy, Poland, Romania, Russia, Serbia, Spain, Ukraine) and Brazil, with this latter country accounting only for 1 patient. Patients with hematologic malignancies (HM) were stratified according to TLS risk (intermediate or high; Cairo et al, British Journal of Haematology, 2010) and serum uric acid (sUA) level (≤ or > 7.5 mg/dl) and randomized at one-to-one ratio to receive TLS prophylaxis with febuxostat or allopurinol. At randomization, 341 patients were blindly assigned by the physician to receive standard or high daily dose level of study drug, containing either allopurinol 300/600 mg or fixed febuxostat 120 mg, respectively. A total of 286 (84%) patients were assigned to receive standard dose level, 253 (88%) and 33 (12%) of which were at intermediate and high TLS risk, respectively. By contrast, in a total of 55 (16%) patients the high dose level of treatment was chosen, 28 (51%) and 27 (49%) of which were at intermediate and high TLS risk, respectively. In order to identify which factors might have influenced the physician’s choice, an analysis of variance (ANOVA) model was used considering the dose level as dependent variable and each of the following variables as covariates (one at a time): TLS risk grade (intermediate or high), sUA level (≤ or > 7.5 mg/dL), type of HM (acute leukemia or chronic lymphocytic leukemia / non Hodgkin lymphoma), presence/absence of high disease burden (defined as bulky disease for lymphomas or white blood cells count ≥ 100 x 10⁹/L for leukemias), lactate dehydrogenase level (< or ≥ 2 x upper limit of normal) and country. Since the standard dose level was chosen mainly for patients at intermediate TLS risk while the high one was equally distributed between TLS risk groups, ANOVA models were performed in the two TLS risk subgroups with the dependent and independent variables described above. In the intermediate TLS risk subgroup (n=281), all the covariates were statistically significant excluding high disease burden. Conversely, in the high TLS risk subgroup (n= 60) only country reached statistical significance (p= 0.053) despite the small sample size. Our study confirms that 300 mg/day is the preferred dose of allopurinol for patients at intermediate TLS risk; in this patient group objective disease-related factors as well as country drove the investigator’s choice. On the other hand, the 300 mg and 600 mg daily dose levels were almost equally distributed in patients at high TLS risk and only country impacted on the physician’s choice for this subgroup. The availability of a fixed posology scheme, as febuxostat offers, may overcome the subjective nature of variability in the clinical practice of TLS prophylaxis, in particular for patients at high TLS risk.