The importance of Peer-Review

When I was new to prostate cancer research in 2011, I attended my first PCRI Conference. One of the presenters was Eugene Kwon. He presented some early data on an experimental new tyrosine kinase inhibitor called cabozantinib. Tyrosine kinase is an enzyme that stimulates metastatic growth. Kwon showed the following "gee whiz" images that had the whole room abuzz. It seems to show the complete resolution of numerous bone metastases on a bone scan in 3 patients in 6 or 12 weeks:

Kwon presented "Cabo" as a game-changer in PCa treatment to an auditorium full of patients and on YouTube. Was it too good to be true?

Several Phase 3 trials found that Cabo only disguised bone metastases so that they no longer showed up on a bone scan. Smith et al. (in the COMET-1 trial) showed that Cabo did not improve PSA or survival, and it had unacceptable toxicity. Basch et al. (in the COMET-2 trial) found that pain palliation was no better than mitoxantrone (an early kind of chemo.)

In spite of these negative findings, research went ahead into using Cabo with a variety of combinations. Neeraj Agarwal presented the results of the combination of Cabo+atelolizumab in the CONTACT 2 trial at an ASCO presentation. In theory, cabo makes bone metastases more permeable to immunotherapies like atelolizumab (A). And in an early Phase 1 trial it seemed to help progressed patients. 507 patients with soft-tissue mets were randomized to receive either Cabo+A or the control drug Xtandi/Zytiga (whichever one they hadn't had already). After 12 months, the results were as follows:

• Radiographic progression-free survival (rPFS) was 6.3 months for Cabo+A vs 4.2 months for the control (statistically significant)
• rPFS among men with liver metastases was 6.0 months for Cabo+A vs 2.0 months for the control (statistically significant)
• rPFS among men who had had docetaxel while hormone sensitive was 8.8 months for Cabo+A vs 4.1 months for the control (statistically significant)
• rPFS among men who had bone metastases  was 6.3 months for Cabo+A vs 4.1 months for the control (statistically significant)
• Objective Response Rate was 13.6% for Cabo+A vs 4.2% for the control (statistically significant)
• Overall survival was 16.7 months for Cabo+A vs 14.6 months for the control (not statistically significant, but data immature)
• Treatment-related adverse events (serious or life-threatening) were 33% for Cabo+A vs 8% for the control.

Agarwal concluded that Cabo+A showed a clinically meaningful improvement in PFS vs second advanced hormonal therapy. This supports Cabo+A as a potential new treatment option for mCRPC patients who have progressed on one second-line hormonal therapy.

But...

Kim Chi provided the peer review. He pointed out that:

• The differences in PFS are modest and not clinically meaningful.
• There was no difference in time to pain progression and the same deterioration in QOL.
• As in the trials of Cabo alone, there is so far no difference in survival. In the COMET 1 trial (above) there was never any difference in survival. Cabo masked the deterioration, such that it only appeared that there was an improvement in rPFS.
• The difference in rPFS is the same as with Cabo alone (about 6 months). The immunotherapy didn't seem to improve outcomes.
• The control group (switch to the untried second-line hormonal) is not what most patients would try next. They would try docetaxel (rPFS= 8-9 months), cabazitaxel (rPFS= 8 months), or Pluvicto (rPFS= 8.7 months). All provide better rPFS than Cabo+A.
• Due to toxicity, there were dose delays/holds/reductions in 40-60% of patients
• Patients would experience less toxicity and greater benefit with taxane chemotherapy.
• Low rate of subsequent therapy (25%), suggests that these patients lost an opportunity to get another life-prolonging therapy.

Patients are sometimes tempted to learn about therapies from YouTube videos and other social media. Patients should be alerted that without good peer-review they may be dangerously misled.