Baseline characteristics of patients and incidence of TLS
This study included 210 patients. The baseline characteristics are summarized in Table 1. The proportion of patients aged ≥ 65 years was over 60%, and the proportion of men was approximately 50%. As parameters of tumor burden, the mean proportion of plasma cells in the bone marrow was 32.2 ± 22.7%; the proportion of patients with detectable plasma cells in the peripheral blood was 11.9%; the proportion of patients with extramedullary disease confirmed with computed tomography scan was 7.4%; and the mean LDH level was above the upper limit of the normal range. The mean SCr level was 1.19 ± 0.99 mg/dL, and the mean eGFR (61.7 ± 30.3 mL/min/1.73 m2) was categorized as "mildly decreased" chronic kidney disease according to the Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guideline criteria .
Seventeen patients had developed TLS, from which ten (4.8% of the total) had developed LTLS, and seven (3.3% of the total) were diagnosed with CTLS (Table 1) due to elevated SCr levels. In seven patients with CTLS, four patients met the criteria of myeloma cast nephropathy (CN) at the initiation of primary therapy according to prior reports . No CTLS patients died or were referred for dialysis during the treatment period. In patients with TLS, 11 were in-hospital stay, while six were outpatients. No outpatients needed to be hospitalized for TLS treatment. Out of these six outpatients, five had LTLS with no clinical sequelae and were not worsened despite continuing MM treatment. One remaining patient was given CTLS diagnosis with elevated SCr levels, and was closely monitored as an outpatient receiving febuxostat against hyperuricemia. Renal function for this patient was improved without hospitalization. In the seven patients with CTLS, three and four patients developed CTLS during the first and second treatment cycles, respectively. Moreover, focusing on the timeframe during which CTLS developed in the treatment cycle, three patients developed CTLS within seven days of commencing the therapy cycle, whereas four patients developed CTLS beyond eight days.
Administration of anticancer agents and supportive therapy agents
Anticancer agents and supportive therapy agents for TLS prophylaxis that were administered prior to the commencement of the primary treatment for MM are shown in Table 2 (A). The median number of therapy cycles and duration of the primary treatment were similar between patients who developed and did not develop TLS. The proportion of patients who had their delivered dose reduced from their originally planned dose was also not different between the patients with and without TLS. Regarding the prophylactic measures against the development of TLS, the administration of antihyperuricemic agents, such as allopurinol / febuxostat or rasburicase, hydration, and urine alkalization by sodium bicarbonate agent were used in a nonuniform manner based on physicians’ discretion for each patient. The proportion of patients who were administered antihyperuricemic agents before the initiation of the primary treatment was 34.8%; the proportion of patients who received these agents in this manner was similar between patients who did and did not develop TLS. Out of the 210 patients, 42.4% received hydration which was defined as the average volume over 1,000 mL/day for the first eight days of commencing the primary MM treatment. Moreover, 62.4% received urine alkalization using sodium bicarbonate agent at the start of the primary treatment. The proportion of patients received either hydration or urine alkalization was similar between patients with or without TLS.
The number of patients treated with Bor-CT and other therapies that did not include Bor (therapy without Bor) was 130 (61.9%) and 80 (38.1%), respectively (Table 2 (B)). The incidence of TLS in patients treated with Bor-CT had a higher tendency than in those treated with therapy without Bor. Furthermore, six of the seven patients who developed CTLS were also among those who had been treated with Bor-CT. TLS incidence in patients who have received Bor-CT or therapy without Bor is shown in Supplementary Table S1. Treatment regimens in patients treated with Bor-CT were as follows: the Bor plus dexamethasone and melphalan plus prednisolone with Bor regimens were administered to 54.6% and 40.8% of patients, respectively. TLS incidence was not high in specific regimen, and in patients who received triplet regimen, TLS incidence was not higher than those who received doublet regimen (Supplementary Table S2). Regarding bortezomib administration route in the 130 patients who underwent Bor-CT, 29 patients were administered intravenously while 101 subcutaneously; the proportion of patients who developed TLS was similar between both administration routes (Supplementary Table S3). The treatment regimens in patients receiving therapy without Bor were as follows, vincristine plus doxorubicin with dexamethasone, melphalan plus prednisolone, and lenalidomide plus dexamethasone regimens were administered to 46.9%, 33.3%, and 9.9% of patients, respectively. No patient was treated with carfilzomib, ixazomib, daratumumab, or pomalidomide.
Analysis of risk factors for developing TLS
Univariate analysis using an unadjusted logistic regression model (crude model) revealed a significant association with high pretreatment SCr levels. This analysis also revealed associations at P < 0.100 between TLS and male sex, Bor-CT, ISS stage III, and high pretreatment uric acid levels (Table 3). A high pretreatment level of serum potassium or phosphorus, one of the criteria for the diagnosis of TLS, was not included as an explanatory variable in this analysis, as only six and three patients showed evidence of abnormal serum potassium and phosphorus, respectively, prior to treatment initiation. Additionally, the presence of extramedullary disease was not included in this analysis, since no TLS patients had any extramedullary disease prior to initiating MM primary treatment. The model that was adjusted for age and sex (Model 1) also exhibited similar tendencies as the crude model. The multivariate model (Model 2), which was adjusted for the variables having P-values < 0.100 in the crude model, revealed a trend between Bor-CT and the TLS [odds ratio (OR) = 3.40, P = 0.069] and that between male sex and TLS development (OR = 2.29, P = 0.153). Then, we performed subgroup analyses focusing on sex. In the subgroup analyses of male and female patients (Table 4), Bor-CT was significantly associated with the development of TLS only in males, after adjusting for multiple variables using Model 2 on Table 3 (OR = 8.51, P = 0.046).