A phase 2 study of SRT501 (resveratrol) with bortezomib for patients with relapsed and or refractory multiple myeloma
Resveratrol (a naturally occurring phytoalexin produced by plants) has described cancer prevention properties and is potentially responsible for the disease-preventing properties of red wine (Jang et al, 1997; Baur & Sinclair, 2006). It has poor water solubility and low bioavailability, hence SRT501 (a micronized oral formulation) was developed and subsequently demonstrated to activate Sirtuin Enzyme 1 (SIRT1) (Sirtris Pharmaceuticals, Inc, 2009). Over-expression of SIRT1 inhibited cancer cell growth (Michishita et al, 2005), reduced cell invasion, and regulated cell cycle and DNA replication. Resveratrol demonstrated cytotoxicity in multiple myeloma (MM) cell lines and inhibited nuclear factor kappa B, as well as AKT and Signal transducer and activator of transcription 3 (Jazirehi & Bonavida, 2004). The combination of resveratrol and bortezomib achieved synergistic cytotoxicity in vitro (Bhardwaj et al, 2007) providing rational for clinical development. Twenty four patients (median age 66·5 years) were enrolled into this phase 2 clinical trial of SRT501 with or without bortezomib in MM patients who had relapsed or were refractory to at least one prior therapy. Inclusion criteria included adequate renal/bone marrow function, and prior bortezomib was permitted irrespective of response. Patients received 5·0 g of SRT501 (Sirtris Pharmaceuticals Inc., Cambridge, MA, USA) following breakfast for 20 d in a 21-d cycle up to 12 cycles. After two cycles, those with ≥stable disease (SD) received two additional cycles, those with progressive disease (PD) had bortezomib added (1·3 mg/m2 on days 1,4,8, and 11). Patients achieving ≥SD after four cycles monotherapy continued SRT501 until PD whereupon bortezomib was added. Those with PD during SRT501 and bortezomib were withdrawn. 40 subjects were planned to result in 30 evaluable patients (defined as completing the first two cycles of SRT501monotherapy) with an estimated overall response rate (ORR) of 35% according to European Group for Blood and Marrow Transplant criteria (Blade et al, 1998). Patients had received a median of 4 (range 1–9) prior therapies, 19 had prior bortezomib of which 14 had responded. Prior treatment combinations were: thalidomide (71%), bortezomib (88%), lenalidomide (58%), autograft (75%), cyclosphosphamide (38%), melphalan (17%), others (21%). All 24 patients commenced SRT501 monotherapy, and nine had bortezomib added for PD. Two protocol violations occurred, resulting in exclusion from efficacy (but not safety) analysis. The mean duration on study was 92·8 d (range 1–351) and patients received a mean cumulative dose of 336 g SRT501 (range 5–1505 g). Eleven patients discontinued before first response assessment and, according to study protocol, were not evaluable. The modified intention-to-treat population (mITT) comprised 13 patients. The median number of cycles was two (monotherapy) and three additional cycles for SRT501 and bortezomib. 10 (42%) subjects continued with SRT501 monotherapy beyond cycle 2, and 4 received only SRT501 monotherapy (maximum eight cycles); however none receiving monotherapy achieved ≥SD. Nine (69%) of 13 evaluable subjects had SD following a minimum of two cycles, but as four crossed over to the SRT501 + bortezomib arm, durability could not be assessed. Nine patients received SRT501 and bortezomib combination, achieving an ORR (≥MR) by mITT of 22% or, by intention-to-treat analysis, an ORR of 8% (Table 1). Six of nine patients treated with the combination continued for more than two cycles, up to a maximum of eight cycles. The median time to progression was 2·8 months and overall survival was not reached. The most commonly reported adverse events [AEs, National Cancer Institute Common Toxicity Criteria v4·0 (http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4·03_2010-06-14_QuickReference_8·5x11.pdf)] were: nausea (79%), diarrhoea (71%), vomiting (54%), fatigue (46%) and anaemia (38%). 54% of patients reported ≥grade 3 AEs; most commonly 21% haematological (anaemia and thrombocytopenia), 21% renal failure, 13% nausea and 13% infections. 50% had a serious AE (SAE) and two deaths occurred on study (one possibly treatment related; one due to PD). 15 of 24 patients treated with SRT501 monotherapy were withdrawn from treatment (10 due to AEs, 1 due to AE leading to death and 4 following investigator's decision). All nine patients receiving SRT501 and bortezomib were withdrawn from study (5 due to AEs, 3 following investigator's decision and one patient withdrawal). The predominant study finding was unexpected renal toxicity, with five SAEs of renal failure leading to early study termination (Table 2). Three of these five patients had an elevated serum creatinine prior to first dose and 4 of the 5 were taking long-term sodium clodronate. Renal failure occurred within the first two cycles of SRT501 monotherapy, with a median time from first dose to renal impairment of 7 d (range 1–37); however one patient had an increase in creatinine before treatment. All five patients reported nausea and vomiting prior to hospitalization and two required temporary haemodialysis. Renal biopsy performed in two patients demonstrated cast and crystal nephropathy in one, and acute tubular damage without cast nephropathy in the other. Disease progression occurred in four of the five patients at the time of renal failure and two patients were discharged for palliation. Following three cases, a medical review meeting recommended further monitoring and recruitment was stopped following five SAEs. Two hundred and thirty seven patients in seven studies had previous SRT501 treatment, predominantly healthy volunteers (n = 92), but also type two diabetes (n = 136) and mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS) syndrome. The phase 2 dose (5·0 g) was safely assessed in these studies (Sirtris Pharmaceuticals, Inc, 2009). Furthermore, a Phase 1 study for metastatic colorectal cancer patients (Howells et al, 2011), did not report nephrotoxicity. As SRT501 is extensively metabolized, renal failure seemed specific to MM patients. Renal impairment can occur in up to 50% of MM patients (Bird et al, 2011) due to multiple causes, hence all are at risk. Three of these patients had an elevated serum creatinine at screening, and rose in another prior to treatment, suggesting pre-existing renal damage. Furthermore, four patients demonstrated progressive myeloma at SAE. Renal failure was not observed for SRT501 and bortezomib despite low efficacy (14 had prior bortezomib and five not previously responded). However disease stabilization by bortezomib may have prevented renal failure whereas low efficacy of SRT501 with nausea and vomiting may have resulted in disease progression and dehydration, leading to renal failure. This study demonstrated an unacceptable safety profile and minimal efficacy in patients with relapsed/refractory MM highlighting the risks of novel drug development in such populations. This study was registered at www.ClinicalTrials.gov:NCT00920556 and was approved by local ethics boards, compliant with declaration of Helsinki and informed consent obtained for all patients. All authors were involved in conducting the study, collecting trial data, writing and approving the final manuscript. JC and EJ designed the study, RP, EJ and JC analysed the data. PE and EJ are/ were employees of Sirtris, A GSK Company at time of study. TP, FD and RP have received honorarium from Janssen.