MATTERHORN is a large phase III study that tested the hypothesis that anti-PD-1 therapy in the neoadjuvant perioperative setting will improve outcomes in patients with non-metastatic adenocarcinoma of the stomach and gastroesophageal junction. As you know, we use immunotherapy anti-PD-1 agents such as nivolumab and pembrolizumab in stage 4 disease routinely with FOLFOX typically in the first line metastatic setting and we wanted to bring these drugs to early stage disease. So we randomised patients with non-metastatic adenocarcinoma, irrespective of PD-L1 status, to receive three-drug chemotherapy with FLOT or FLOT with durvalumab. So it’s a placebo controlled randomised phase III study with PD-L1 as a stratification factor, regional lymph node status and also Asia versus non-Asia.

Patients received two cycles of durvalumab given once a month with FLOT given on day 1 and day 14, also with two cycles. So essentially it’s four cycles of FLOT, four treatments of FLOT before and after surgery. Then patients received durvalumab for up to a year after completion of everything. So, of course, at ASCO in the plenary session we saw event free survival, disease free survival and pathologic complete response rate and all of those were improved with the addition of durvalumab to FLOT. We demonstrated that FLOT is a global feasible standard; patients all over every continent, in South Korea, in Japan, in the United States, in South America, in Europe, were able to receive FLOT. We demonstrated that durvalumab did not compromise the delivery of FLOT and the delivery of FLOT was actually better than what we saw in other randomised studies. Most patients were able to complete all perioperative FLOT and a substantial proportion of patients, more than 50%, finished all cycles of adjuvant therapy.

So that was a big win, it was an ASCO plenary, that was a New England Journal of Medicine publication and at ESMO Congress this year in Berlin we presented the final overall survival. That was the only missing piece, it was trending positive but we couldn’t make it a statistically determined positive endpoint. I’m excited to report that ESMO was a transformative year for gastroesophageal cancer – for the first time we were able to show in the perioperative setting that a targeted agent such as durvalumab improved survival. We saw 18-month survival of 81%, 24-month survival of 75% and 36-month survival of 68.6%. So at all of these landmark analyses the survival delta is approximately 4-6% in favour of durvalumab. The medians were not reached because this patient population with curative disease, particularly in modern studies, do well, but we also showed that pathologic complete response and nodal negative status does correlate with survival which is helpful for the future studies to have a surrogate endpoint of pathCR.

What’s interesting is that in every subgroup we looked at, including PD-L1 negative patients, including diffuse histology and others, there was a benefit for use of anti-PD-1 therapy, anti-PD-L1 durvalumab with FLOT compared to placebo. So irrespective of whether or not pathCR or the degree of response or nodal response, the systemic disease control with FLOT + durvalumab is superior and these patients live longer and their cancers are less likely to recur.

The grade 3/4 events are mostly related to chemotherapy but as the community of oncologists are more comfortable giving these drugs this regimen is more and more accepted. It will be a recommended part of the guidelines, it’s already part of NCCN guidelines and it’s in Europe and even in countries like Japan where historically FLOT was not accepted. So it’s an exciting opportunity for us to unite approaches to this disease.

What could be the implications of these findings?

The implications for the MATTERHORN study is now we can move forward and do biomarker-selected subpopulations. Of course in MATTERHORN patients benefitted irrespective of PD-L1 but what if we use tumour targeting together with immunotherapy? We’re exploring that for HER2 positive disease with a combination of anti-HER2, trastuzumab, plus anti-PD1 and FLOT. How long do we need to keep patients on therapy? Can we use other surrogates of efficacy such as pathologic complete response or ctDNA clearance or some sort of composite endpoint to de-escalate, to introduce non-operative management, to introduce risk-adjusted stratification to help patients live longer but also tolerate the therapy well?

Is there anything else you would like to add?

The important part is in the MATTERHORN study we allowed patients and their oncologists to guide their own perioperative dose intensity or even adjuvant dose intensity, the timing and when they started. We gave ranges but in general what we have to remember is three drug combinations are important but you don’t have to go with parent FLOT, you can start at the lower doses, especially in patients where you’re concerned for additive, perhaps older patients or patients where you’re concerned. Using growth factor support and using hydration and also dose modification helps us get through perioperative FLOT much better.

In the MATTERHORN there was a range of age. We have treated patients in their 80s with FLOT and in the subgroup analysis older patients benefitted just as much. So as you’re selecting patients for FLOT, getting more comfortable with dose modifications and supporting the patients with antiemetics, hydration and growth factor support will make this regimen more deliverable to your average patient in the community.