ASCO 2017: Emerging VEGF-I/O Combinations: Efficacy and Toxicity

Chicago, IL ( Dr. Hans Hammers provided an excellent summary of abstracts 4504-4506 (Abstract 4504 TK Choueiri et al, Abstract 4505 MB Atkin et al, Abstract 4506 S. Chowdhury), all of which address emerging VEGF and immunotherapy combinations. The rationale for combination therapy can be traced back to the CheckMate 025 trial1, in which nivolumab was tested against everolimus as a second-line therapy for advanced RCC.

Nivolumab had a higher objective response rate (ORR) (25% vs. 5%) and less Grade 3-4 adverse events (19% vs. 37%), without necessarily improving progression-free survival (PFS). Based on this, the rational exists for potential combination therapy. However, balancing adverse events for survival benefit is often the limiting factor.

He had an interesting commentary on the possibility that combined therapy may not necessarily be synergistic in the traditional sense (targeting complementing pathways), but that they may help push cancer cells towards cell death. The balance between cell death and cell growth is normal cells is significantly altered favoring cell growth in cancer. However, while immune therapies shift it slightly towards cell death (in some patients), and targeted therapies balance out cell death and cell growth (static therapy), the combination may shift the balance to cell death.

He began by reviewing Abstract 4506 by Chowdhury et al. In this study, he congratulation the authors for their determination. He, along with the authors, makes it clear that the combination of pembrolizumab and pazopanib should not be considered for further studies due to hepatoxicity and side effect profile. However, it does serve as a very important example of the significant toxicity from combined therapy. This study, in conjunction with CheckMate 016, which compared nivolumab with either sunitinib or pazopanib, demonstrated that non-specific TKI’s may not be appropriate for combination therapy. Not all TKIs are equivalent! There, there exists a case for more selective TKI’s as combination agents with immune checkpoint inhibitors.

The study by Choueiri et al (Abstract 4504) utilizing avelumab and axitinib (currently a 2nd line TKI) demonstrated very minimal hepatotoxicity. The safety profile of the combination appeared acceptable in this phase 1/2 study, but he did note that approximately 8% of patients still had Grade 3 hepatotoxicity, more than the expected 1% for axitinib alone. While there was one immune-mediate death (myocarditis), the overall safety profile appeared acceptable. Particularly taken in context of its early clinical outcomes, this seems to be a promising combination. With a confirmed ORR of 58.2%, this is significantly better than monotherapy alone can offer. However, as this is a preliminary study, much longer follow-up is needed to ensure durability of response. 

This study also highlights the success with a more selected TKI. Selective TKI’s, such as axitinib and tivozanib, or multiselective TKI’s, such as cabozantinib and levantinib, may offer better oncologic benefit with more acceptable adverse event profiles than non-selective TKIs such as pazopanib and sunitinib. 

Lastly, he reviewed Abstract 4505 by Atkin et al, in which the authors assessed the combination of atezolizumab with bevacizumab (VEGF inhibitor) vs. sunitinib alone. He congratulated them on novel study design that allowed for cross-over after monotherapy. He notes that the ORR of combination therapy compared to monotherapy alone wasn’t very different: 32% for combination therapy and 25% for atezo monotherapy. Compared to 58% axitinib/avelumab ORR and a 71% pembro/axitinib ORR, this does not appear to be as effective. 

However, an interesting component of this group’s study is their sequencing of patients and identification of a potential biomarker of response. Specifically, the identification that patients with high T-effector cell expression & high levels of myeloid inflammation have a better response to bevacizumab addition than patients with high T-effector cell expression & low levels of myeloid inflammation may help guide the addition of bevacizumab in the future. He points out that it was unclear how independent of PD-L1 expression this new biomarker was and needs further evaluation.

What I particularly liked was his assessment of how all these combinations will affect the current Kaplan-Meier curve for immunotherapy – currently, 70-80% of patients have an early failure rate, but there is a long-tail (durable response) for the 20-30% that do respond. Ideally, these therapies maintain a durable response, but raise the response rate and the overall survival. 

At this time, without longer follow-up, we don’t yet know if these new combinations merely increase the early ORR but have no effect on PFS or OS. However, novel combination therapies are promising, as long as the balance against the side effect profile of the treatments themselves.

Presented By: Hans Hammers, MD, PhD, University of Texas Southwestern Medical Center, Dallas, TX

Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto
Twitter: @tchandra_uromd

at the 2017 ASCO Annual Meeting - June 2 - 6, 2017 - Chicago, Illinois, USA

1. Motzer RJ, Escudier B, McDermott DF, George S, Hammers HJ, Srinivas S, Tykodi SS, Sosman JA, Procopio G, Plimack ER, Castellano D, Choueiri TK, Gurney H, Donskov F, Bono P, Wagstaff J, Gauler TC, Ueda T, Tomita Y, Schutz FA, Kollmannsberger C, Larkin J, Ravaud A, Simon JS, Xu LA, Waxman IM, Sharma P; CheckMate 025 Investigators. Nivolumab versus Everolimus in Advanced Renal-Cell Carcinoma. N Engl J Med. 2015 Nov 5;373(19):1803-13. doi: 10.1056/NEJMoa1510665. Epub 2015 Sep 25.