Keynote speech on the JAVELIN not going far enough to improve survival

The treatment landscape for metastatic renal-cell carcinoma has changed dramatically with the introduction of immunotherapies. Unfortunately though, we are promoting combinations over single agents without having much idea of added benefit of each drug. This is an important issue because when we combine two drugs, the only thing we are certain of are the added toxicities. PD-1 inhibitor nivolumab had improved OS when given in second line, however nivolumab was tested in combination with ipilimumab (not as a nivolumab monotherapy) in the first line trial. Now, pembrolizumab and avelumab have followed suit, although their combination partner was axitinib - a VEGF inhibitor. The control arm was sunitinib for both of the trials of pembrolizumab plus axitinib (KEYNOTE 426) and avelumab plus axitinib (Javelin 101). This is a little surprising because we are testing A B versus C, where both A and B haven’t been approved for the given setting - axitinib was approved for RCC in second line. Both these combinations improved PFS versus sunitinib but only the pembrolizumab combination has shown improved OS. However, I have doubts about the contribution of axitinib to these results. What would the outcome be if pembrolizumab alone is followed by sunitinib in second line? It is important to note that only one third of patients who discontinued sunitinib received PD-1 inhibitor subsequently in the KEYNOTE 426 trial. The important question for patients and clinicians would be to consider a survival difference had most of these patients received a PD-1 inhibitor subsequently. As for avelumab, the JAVELIN trial hasn’t reached as far as pembrolizumab and nivolumab have reached: The OS benchmark - so let’s reserve this combination until we see that benefit.

We should remember that this combo-mania with PD-1/PD-L1 inhibitors may also backfire. Previously, the RCTs of nivolumab and pembrolizumab combos were halted in multiple myeloma for higher deaths in the combo arms. Another RCT IMblaze 370 also reports that atezolizumab, alone or in combination with cobimetinib, failed to improve survival versus regorafenib in patients with metastatic colorectal cancer.  This time again A B failed versus C although C in itself is a drug with very marginal benefits in this setting. Also, I don’t understand testing A plus B combo when both A and B are unapproved for the disease.

Have we successfully landed on the COMET?

SABR-COMET was an open label phase 2 RCT which tested if the use of stereotactic ablative radiation therapy (SABR) in the treatment of oligometastases (up to 5) improved survival. Among 99 patients randomised 2:1 to SABR plus standard care or standard care alone, the median OS was found to have improved dramatically by 13 months in the SABR group (41 v 28m, HR 0.57, p = 0.090). The authors conclude appropriately that “Phase 3 trials are needed to conclusively show an overall survival benefit.” I have no criticism against this study. However, I have criticism against some top experts claiming that this trial is already practice changing. No, this is a phase 2 trial whose objective was to determine if this intervention was worth testing in a phase 3.  The trial is positive, so the appropriate conclusion is “let’s test this in a phase 3 trial”, not “let’s change practice today”. It is also worth noting that the intervention had a 4.5 percent treatment-related mortality rate! We should never jump to conclusions, and kudos to the authors for concluding appropriately in the report of this public-funded RCT.

You may want to imagine how a similar RCT may have been published had it been an industry-sponsored drug trial. Well, you don’t need to imagine hard because we already have an example. In this partially industry-funded phase 2 RCT of regorafenib in 42 patients with relapsed metastatic osteosarcoma, the conclusion reads: “Regorafenib should be considered a treatment option for patients with relapsed metastatic osteosarcoma”. Guess what the results were to enable such a conclusion? The gain in PFS (primary endpoint) was less than 2 months. And the OS is in fact...wait for it…less in the regorafenib arm than the placebo arm by more than 2 months with a hazard ratio of 1.26 (95% CI, 0.51-3.13)! Can you now see how bias creeps in even in high-profile journals? SABR-COMET, a public-funded randomised phase 2 of 99 patients with both an OS endpoint and an OS difference of more than a year concludes (appropriately) that we should test the intervention in a phase 3 trial, while an industry-funded phase 2 RCT of a drug in only 42 patients with a PFS primary endpoint concludes that the drug should be considered a treatment option, despite the benefit in PFS being less than 2 months - but the detriment in OS (although not significant) being more than 2 months. I’m puzzled by the need to sugarcoat these results, especially on the background of olaratumab’s recent failure to improve OS in phase 3 despite showing improved OS in phase 2 in patients with soft tissue sarcoma. Speaking of conclusions based on phase 2 trials, a trial reports that “administration of nab-paclitaxel plus gemcitabine-cisplatin may prolong survival vs administration of gemcitabine-cisplatin alone for the treatment of advanced biliary cancers”. You, like me, may be wondering why test nab-paclitaxel instead of the much cheaper paclitaxel. Well, this trial was funded by Celgene. So it may not come as a surprise to you at all that this was a single arm trial and that “versus” in the previous statement was comparison to historical controls. Thankfully though, they are testing this in a phase 3 RCT.

The time to study metastatic cancer survivorship was yesterday

In one of the most poignant pieces on cancer survivorship written by two cancer survivors, we realise how the oncology community has forgotten the needs to address quality of life issues of metastatic cancer survivors in our enthusiastic quest of chasing marginal gains in oncology. The authors poignantly describe the various issues including hope, financial toxicity, psychosocial needs, caregiver issues etc. that trouble metastatic cancer survivors. Although we can be proud that we now have millions of cancer survivors living long term, we can’t continue to ignore their issues in research and clinical practice. The best time to study metastatic cancer survivorship was yesterday, the next best time is today.

Let me take a selfie

Is it ok to provide full approval to cancer drugs on the basis of non-inferiority trials especially if these drugs don’t offer any other benefits such as ease of administration, lower cost etc to justify the non-inferiority design? I discuss these issues on the background of recent approval of lenvatinib in hepatocellular cancer based on non-inferiority trial in a JAMA Oncology viewpoint written with Dr. Aaron Kesselheim.

Dr. Gyawali is a medical oncologist with work experience in Nepal, Japan, US and Canada. He is also an Assistant Professor of Public Health Sciences and Scientist in the Division of Cancer Care and Epidemiology at the Queen’s University Cancer Research Institute, Canada and an affiliated faculty at Program On Regulation, Therapeutics And Law (PORTAL) at Brigham and Women’s Hospital, US. The opinions expressed herein are his own. You can read his previous blogs here.