Phase II open-label study of recombinant circularly permuted TRAIL as a single-agent treatment for relapsed or refractory multiple myeloma

Multiple myeloma (MM) is a malignancy of mature plasma cells. In the United States, it is the second most common hematologic malignancy. Recently, new treatments, including autologous stem cell transplantation, proteasome inhibitors (e.g., bortezomib), and immunomodulatory drugs (IMiDs; e.g., thalidomide and lenalidomide), have substantially improved the response and survival rates of patients with MM [1]. However, relapse and drug resistance remain inevitable. Therefore, additional novel and more effective treatments are urgently needed for patients with relapsed or refractory MM (RRMM).

In 1995, apoptosis ligand 2/tumor necrosis factor-related apoptosis-inducing ligand (Apo2L/TRAIL) was discovered as a new member of the tumor necrosis factor superfamily [2]. It binds to death receptor 4 (DR4) and death receptor 5 (DR5) on the cell surface to form the death-inducing signaling complex, thereby activating the caspase cascade to induce apoptosis, a signaling event known as the extrinsic apoptotic pathway [3]. Apo2L/TRAIL selectively and quickly induces cell apoptosis in many solid tumors and hematologic malignancies, while exhibiting no toxicity to normal cells [4, 5].

Circularly permuted TRAIL (CPT) is a novel mutant form of recombinant human Apo2L/TRAIL that is currently in clinical development for the treatment of MM and other hematologic malignancies. The primary molecular feature of CPT is amino acid (aa) 121–135 at the N-terminus of wild-type Apo2L/TRAIL, which is connected to aa 135–281 at its C-terminus via a flexible linker. Consequently, compared with wild-type Apo2L/TRAIL, CPT has better stability, demonstrates more potent anti-tumor activity, and has a slightly longer in vivo half-life in mice, rats, and humans [6].

For pretreated patients with either hematologic malignancies (including lymphoma and MM) or advanced solid tumors (e.g., lung, colorectal, and renal cancers), a phase I study (unpublished data) evaluated the safety of CPT monotherapy by single-dose escalation (0.5–3.5 mg/kg) and by multiple-dose escalation (1.0–2.5 mg/kg). In the multiple-dose escalation, CPT was administered once daily for 14 consecutive days for each 21-day cycle. Unexpectedly, 2 of 3 patients with RRMM achieved a partial response after the first cycle of CPT treatment at the 2.5-mg/kg dose. This early evidence of activity of single-agent CPT in patients with RRMM caused the dose-escalation study to be terminated, while the maximum tolerated dose (MTD) was not reached. MTD was defined as the highest dose level at which 33% or fewer patients experienced a dose-limiting toxicity (DLT), according to the 3+3 design. DLT was predefined as any of the following that occurred during the first course of treatment and was determined to be possibly, probably, or definitely related to study treatment: (a) grade 3 or higher non-hematologic toxicities and (b) grade 4 hematologic toxicity. In this phase I trial of 28 patients, the most common treatment-related adverse events (TRAEs) were alanine aminotransferase (ALT) elevation (42.9%),

aspartate aminotransferase (AST) elevation (32.1%), fever (14.3%), fatigue (10.7%), nausea (10.7%), and vomiting (10.7%). Of these TRAEs, 14.3% were grade 3 or 4, all were associated with hepatotoxicity, and all were resolved by treating the symptoms or discontinuing CPT treatment (unpublished data).

Preliminary results suggested that single-agent CPT, given at a well-tolerated dose of 2.5 mg/kg, once daily for 14 consecutive days for each 21-day cycle, has promising anti-tumor activity in patients with RRMM. Accordingly, this dose and schedule was then chosen as the recommended dose for the current phase II trial that targets a cohort of patients with RRMM only. Therefore, our phase II, single-arm, open-label, multicenter clinical trial further evaluated the safety and efficacy of single-agent CPT, given at 2.5 mg/kg once daily for 14 days for each 21-day cycle, in patients with RRMM.

In this phase II study, single-agent CPT at a well tolerated dose (2.5 mg/kg per day) exhibited a response in patients with RRMM. In the past decade, novel treatments for patients newly diagnosed with MM have substantially improved outcomes; however, nearly all patients who respond to initial therapies eventually relapse and become refractory to current treatments, including new drugs [1]. The prognosis of these patients with RRMM remains dismal. Therefore, new and more effective therapies are urgently needed. In this phase II trial, CPT as a single-agent therapy showed promising activity: RRMM patients who had been aggressively treated previously (median, three prior therapies) had a 33.3% ORR and a 48.1% CBR. Interestingly, CPT was also effective for RRMM patients who had received prior regimens containing new anti-MM agents, including bortezomib (ORR, 51.9%) and bortezomib plus IMiDs (ORR, 33.3%). Thus, the present study showed activity of single-agent CPT in the treatment of MM.

In preclinical settings, pro-apoptotic receptor agonists (PARAs) have shown potent and selective anti-tumor activity; however, few clinical studies of PARAs have reported activity in the treatment of patients with cancer, including MM. For example, Ashkenazi and Herbst [9] summarized that rhApo2L/TRAIL and other PARAs had synergistic effects when combined with cytotoxic agents, including irinotecan, camptothecin, 5-fluorouracil, carboplatin, paclitaxel, doxorubicin, and gemcitabine. Additionally, the combination of Apo2L/TRAIL and bortezomib led to enhanced activity in the induction of apoptosis in cell lines of a variety of solid tumors [10–12] and hematologic malignancies [13–15]. To this end, bortezomib up-regulates DR4/DR5 expression while reducing cellular FLICE (FADD-like IL-1β-converting enzyme)-inhibitory protein (c-FLIP) levels, thereby effectively overcoming Apo2L/TRAIL resistance in MM cells [16]. However, HGS-ETR1 (mapatumumab, a DR4 agonistic monoclonal antibody [17]), when administrated in combination with bortezomib and when compared with bortezomib alone, has failed to improve outcomes of patients with advanced MM.

In patients with MM, high serum β2-microglobulin levels have been reported to be associated with poor prognosis. In the study by Bergsagel [18], 66.7% of patients with higher β2-microglobulin levels (i.e., ≥3.5 mg/L) achieved an MR or better, whereas patients with lower β2-microglobulin levels (i.e., <3.5 mg/L) had a CBR of only 28.6%. These findings suggested that CPT might benefit MM patients who have higher baseline levels of serum β2-microglobulin, which is a well-established biomarker for poor prognosis of this disease [18]. In general, RRMM patients who have been aggressively pretreated usually respond poorly to current regimens, including the new drugs (e.g., proteasome inhibitors and IMiDs). However, in our study, we observed an interesting trend, namely that, after CPT treatment, both ORR and CBR were notably higher in patients who had received more than three prior treatments than in patients who had received three or fewer prior treatments (ORR, 46.2% vs. 21.4%; CBR, 61.5% vs. 35.7%). We observed similar results when we compared response rates between the subgroups of patients who had been previously treated with bortezomib (ORR, 42.9%; CBR, 57.1%) or bortezomib plus IMiDs (ORR, 33.3%; CBR, 55.5%) and those who had not received prior bortezomib treatment (ORR, 23.1; CBR, 38.5%). One possibility for not observing cross-resistance between CPT and current anti-MM agents is that CPT primarily targets DR4 and DR5 to induce apoptosis through activation of the extrinsic apoptotic pathway, whereas these agents (e.g., bortezomib and IMiDs) act mostly via the mitochondria-mediated intrinsic apoptotic pathway. Moreover, a potential explanation for the likelihood of better responses to CPT in patients who received prior bortezomib is that this agent might modulate the signaling molecules (e.g., DR4/DR5 [18]) related to the extrinsic apoptotic pathway, thereby sensitizing MM cells to CPT that targets this pathway. The differences in the response rates (both ORR and CBR) between those subgroups of patients with different baseline characteristics were not statistically significant, most likely because of the small sample size; thus, further studies are needed to validate these preliminary but interesting findings. It is too soon as well to draw any conclusions about which subsets of RRMM patients would likely benefit more from single-agent CPT treatment.

In this phase II study, in addition to elevated ALT and AST levels and fever, which were most frequently observed in the phase I study, the most common adverse events were leucopenia, neutropenia, thrombocytopenia, and rash. The reason for the common hematologic adverse events is most likely due to RRMM itself, which often causes hematologic abnormalities. While gastrointestinal adverse events and fever were very common in an earlier trial of rhApo2L/TRAIL in patients with solid tumors [19], elevated AST and ALT levels and fever were more common in the present cohort of RRMM patients who received single-agent CPT. One possible reason is that in our phase II study, patients received prolonged treatment with CPT (i.e., 14 days for each cycle), whereas in the phase I trial, patients received rhApo2L/TRAIL for only 5 days for each cycle. In the clinical development of PARAs, liver toxicity has been a major concern [20]. In our study, more than half of patients who received single-agent CPT experienced elevated AST and/or ALT levels. However, it is noteworthy that most of these hepatic adverse events were grade 1 or 2, and nearly all were manageable with symptomatic treatments and/or discontinuation of CPT. Interestingly, most patients with elevated AST levels had normal ALT levels, but this was often accompanied by elevated LDH levels. This phenomenon was usually observed during the first cycle of CPT treatment and very rarely afterwards. AST levels returned to normal or near-normal within 1 week; LDH levels declined more slowly, to normal or near-normal within 2 weeks. Interestingly, patients with transient elevation of AST levels and/or LDH levels but normal ALT levels were more likely to achieve an MR or better, suggesting that AST elevation with normal ALT levels might not be associated with liver toxicity but rather tumor lysis in response to CPT treatment [21]. In contrast to transient AST elevation that peaked at day 2 or day 3 of the first cycle and then declined quickly, ALT elevation was more sustained during the CPT dosing period but declined and eventually recovered after discontinuation of CPT. Most patients (80.0%) had grades 1–2 ALT elevation, with normal bilirubin level. Only two patients discontinued CPT treatment because of liver toxicity; of these, one patient had grade 4 AST elevation and grade 3 ALT elevation, which was reported as a serious adverse event, and another patient had grade 3 elevation of AST, ALT, and bilirubin, both of which occurred during the second cycle of CPT treatment. Of note, in both cases, these liver adverse events resolved after discontinuation of CPT. Therefore, although patients in our study commonly experienced elevated AST and ALT levels, the elevated levels were mostly mild or moderate and all manageable. Together, these results suggest that single-agent CPT is well tolerated by patients with RRMM.

Our study does, however, have limitations. First, therapeutic response (i.e., ORR and CBR), rather than the gold standard endpoint overall survival, was selected as the primary outcome to evaluate efficacy of single-agent CPT in patients with RRMM. Second, CPT was administered for a relatively short period of time (only two 21-day cycles). Third, the follow-up time, which ended after the second cycle of CPT treatment, was also short. Collectively, these limitations prevent this study from providing more definitive evidence for the efficacy of single-agent CPT on RRMM. For example, whether longer treatment (e.g., increasing the number of intervention cycles) with single-agent CPT would be more beneficial while not increasing toxicity is still unknown. More importantly, whether single-agent CPT would improve long-term outcomes (e.g., progression-free survival, duration of response, or overall survival) of patients with RRMM is also still unknown.

This phase II study provides the first evidence supporting the promising therapeutic activity of single-agent CPT in the treatment of RRMM patients, including those who were aggressively treated previously with current regimens containing novel agents, such as bortezomib or IMiDs. It also demonstrated that CPT at the dose of 2.5 mg/kg, given once daily for 14 days per cycle for two cycles, was well tolerated. Although mild and moderate hepatic adverse events were relatively common, nearly all were manageable. Therefore, the single-agent CPT therapy and the regimens that combine CPT with first-line anti-MM agents (e.g., bortezomib and IMiDs) warrant further clinical investigation. Accordingly, a phase III trial to further develop CPT as a single-agent therapy and as part of combination therapies for RRMM is currently ongoing.