Elizabeth Smyth, MD
Invited discussant, Elizabeth Smyth, MD, a consultant in gastrointestinal oncology at the Oxford University Hospitals NHS Foundation Trust in the United Kingdom, commented on the notable impact in pathologic complete response rates with the addition of checkpoint inhibitors in MATTERHORN and KEYNOTE-585. “There is no doubt that adding immune checkpoint inhibitors to perioperative chemotherapy downstages the tumor better, but the studies’ inclusion of patients with locally advanced disease probably compromised these outcomes to a degree,” she said.
Dr. Smyth emphasized that although encouraging, the current findings do not yet impact clinical practice, as the ultimate improvement on clinical outcomes is not known. In MATTERHORN, event-free survival data were not reported, and in KEYNOTE-585, an interim analysis showed a numerical—but not a statistically significant—benefit with the addition of pembrolizumab to chemotherapy.
“On the face of it, a 20-month improvement seems fairly impressive, but I think we need to consider the control group received a chemotherapy regimen that is not standard (mostly, cisplatin plus fluorouracil),” Dr. Smyth noted. She added that the results were probably driven by the subset with a combined positive score ≥ 10 and by the additional 9 months of adjuvant pembrolizumab, which gave a “head start” to that arm.
All patients in the MATTERHORN trial received standard-of-care FLOT (fluorouracil, leucovorin, oxaliplatin, docetaxel), which may explain why the pathologic complete response rate was higher (19%) than in KEYNOTE-585 (13%), she said. However, in both studies, pathologic complete response rates were less than seen in historical controls—for reasons that are not clear, she added. Dr. Smyth noted that adding PD-1/L1 to neoadjuvant chemotherapy did not impact the ability to have surgery, and R0 resections were comparable to historical controls.
More to Learn
“As perioperative chemoimmunotherapy in operable gastroesophageal adenocarcinoma, do MATTERHORN and KEYNOTE-985 meet the goal for a successful neoadjuvant therapy?” Dr. Smyth asked. As mentioned, immune checkpoint inhibition did not compromise the ability to undergo surgery. The regimens did downstage tumors and produced an “unequivocal” improvement in pathologic complete response, though the control arms underperformed, maybe because tumors were locally advanced, and the treatment toxicity was “not trivial,” she added.
Did the regimens eliminate micrometastases and develop or reinvigorate T-cell responses against the tumors? “Maybe, but probably only for immunologically hot tumors,” she maintained.
Dr. Smyth said she remains “reasonably hopeful” for a positive event-free survival endpoint with durvalumab in MATTERHORN, based on the study’s statistical plan, use of FLOT, and use of adjuvant chemotherapy. If the endpoint is not met, the next step in evaluating this approach will be biomarker-selected trials and possibly more dose-intense chemotherapy or total neoadjuvant therapy, she explained.
DISCLOSURE: Dr. Smyth reported financial relationships with Amal Therapeutics, Aptitude Health, Amgen, Astellas, AstraZeneca, BeiGene, BMS, Celgene, Daiichi Sankyo, Elsevier, Everest Clinical Research, First Word Group, Five Prime Therapeutics, Gritstone Oncology, Imedex, Merck, My Personal Therapeutics, Novartis, Pfizer, Roche, Sai-Med, Servier, Touch Oncology, Turning Point Therapeutics, Viracta Therapeutics, and Zymeworks.
