Regardless of the treatment strategy, MM is always classified as a low-risk disease for developing TLS despite an increasing number of novel therapies that can achieve antitumor responses, such as proteasome inhibitors, immunomodulatory drugs, and anti-CD38 antibodies [3]. The type of therapy is a known TLS risk factor [1–6]. Actually, the TLS risk for chronic lymphocytic leukemia is classified as low; however, the risk increases to an intermediate level when targeted and biological therapies (fludarabine or rituximab) are used [3]. MM is a relatively rare cancer, with age-adjusted morbidity in Japan of 5.2 and 4.8 per 100,000 men and women, respectively [10], but the risk factors for TLS have not been fully studied. In the present study, we found that Bor-CT in particular is likely to increase TLS risk in MM, particularly among men.
In this study, we defined the duration for the diagnosis of TLS as the period of each treatment cycle during primary treatment, although this is different from the criterion established by expert consensus, which was defined as the duration within 3 days before or up to 7 days after the initiation of therapy [3]. Our changes are based on two reasons for the evaluation of treatment-related TLS. The first is to avoid missing TLS cases that develop > 8 days after the initiation of therapy, because some prior case reports have described Bor-induced TLS that developed beyond 8 days [12, 14, 15]. In fact, in our study, we did identify such patients. The second reason is to exclude spontaneous TLS that develops before initiating cancer therapy. The baseline blood levels of uric acid, potassium, and phosphorus can spontaneously increase in patients with renal insufficiency; thus, such patients could be diagnosed with TLS. Therefore, we excluded patients who already had laboratory abnormalities that met the criteria for LTLS at the initiation of MM treatment and also defined the aforementioned diagnosis period for TLS, which excluded the timeframe before treatment initiation.
Two prior retrospective studies with small sample sizes investigated the development of TLS in MM patients [17, 18]. These reports described the associations between TLS and the following factors: the presence of renal dysfunction or high uric acid levels at baseline, rapid progression of anemia before MM treatment, and treatment with Bor-CT. However, both prior reports did not use multivariate analysis to adjust for the effects of potential confounding factors. Moreover, in previous studies, the incidence of CTLS was found to be more than approximately four- to five-fold higher (12.7% [17] and 17.2% [18], respectively) than that recorded in the present study (3.3%). These discrepancies might be due to the abovementioned differences in the TLS diagnosis period, or they could have been caused by differences in the characteristics of the study populations. All or part of the study populations in these prior studies included recurrent/refractory MM patients, whereas our study included only newly diagnosed, untreated patients. The evaluation of treatment-induced TLS might be difficult in cohorts that include both untreated and recurrent/refractory MM because tumor responses generally would be different between untreated and recurrent/refractory cases or among recurrent/refractory MM patients who had varying pretreatment histories [11, 21].
Interestingly, our multivariate analysis revealed that the risk of TLS is increased by Bor-CT only in males. Treatment with Bor-CT was found to be one of the factors associated with an increased risk of LTLS, and this could be attributable to the enhanced antitumor response it induces, which would be consistent with our initial hypothesis. Moreover, similar to this result, a prior retrospective study investigating the risk factors for TLS in patients with acute myelogenous leukemia receiving induction chemotherapy indicated a significant association between LTLS and the male sex [22]. That report hypothesized that the sex difference in uric acid levels before the commencement of cancer therapy might account for the increase in TLS risk in men, because it has been previously reported that uric acid levels are generally higher in men than in women [23]. However, our multivariate analysis revealed no significant association between LTLS and high pretreatment uric acid levels. Thus, the increase in TLS risk in men cannot be fully explained by only baseline uric acid levels. In contrast, the previous report also suggested that sex might be a surrogate marker for underlying unknown biological differences in the development of TLS. In this regard, a study on Japanese males revealed a strong association between the onset of gout and genetic dysfunction related to urate excretion [24]. A male-specific increase in the risk of developing LTLS in patients receiving Bor-CT might also be associated with genetic mechanisms related to urate excretion, given that in the present study, we found that 16 of the 17 patients who developed LTLS were diagnosed based on elevated levels of uric acid (data not shown). Hence, TLS risk factors should be evaluated further, as some unknown risk factors potentially exist.
The present study has several limitations, mainly with respect to the study design. Clinical data for the evaluation of TLS, such as phosphorus levels, were missing for some patients. Accordingly, the incidence of TLS associated with phosphorus could have been underestimated. Moreover, the detection power of our study might not have been sufficient based on the sample size. In addition, selection bias could not be avoided; hence, we analyzed all patients who met the criteria for enrollment in this study to reduce bias as much as possible.