ESMO Commentary: Immune therapy brings new treatment options for lower GI cancers

VIENNA, Austria – Immunotherapy is a relatively new therapeutic option for patients with lower gastrointestinal (GI) cancer. Although in colon cancer results are still poor compared to melanoma and lung cancer, immunotherapy seems to be effective in selective subgroups of patients. Professor Eric Van Cutsem, Head of Clinical Digestive Oncology at University Hospitals Leuven, Belgium, comments as new research is presented at ECC 2015 in Vienna. 

What is the rationale for immunotherapy in lower GI cancer?

Van Cutsem: Immune therapy is potentially important in GI cancer. We know that cytotoxic immune responses, if they are up-regulated, can damage the host. We also know that there is often an up-regulation of immune mechanisms in the tissues surrounding the tumour and that that plays a meaningful role. So there is a good rationale to block some of the pathways that are involved in up-regulation and in the cytotoxic immune response. 

These pathways do play a role in many cancers and we know for instance that giving programmed death (PD-1) or programmed death ligand (PD-L1) antibodies plays a significant role in lung cancer, melanoma, renal cancer, bladder cancer and Hodgkin lymphoma. So there is already a range of tumours where there is activity shown for this class of agents, which are also called checkpoint inhibitors. And we know that these immune reactions also play a role in lower GI cancer including colon cancer and anal cancer. It’s logical then to look at the role of these antibodies in lower GI cancer.

What is the most recent evidence of efficacy from the research?

Van Cutsem: Preliminary results of KEYNOTE-028 have been presented at ECC 2015 in Vienna. This is an important study investigating the safety and efficacy of pembrolizumab.1 In the cohort with colorectal cancer there was unfortunately very limited activity (abstract 502). The reason for that is not that the drug is not active. It is because the selection of patients is extremely difficult. We have recently seen a study published in the New England Journal of Medicine of PD-1 blockade with pembrolizumab.2 This study checked patients for microsatellite instability (MSI), in other words mismatch repair deficiency. There was clear activity in patients with MSI tumours but not in those with microsatellite stable (MSS) tumours. 

Unfortunately, only 5% of patients with metastatic colorectal cancer have MSI tumours. These patients are hypermutated, meaning that they harbour a lot of mutations, and they seem to be good candidates for checkpoint inhibition with anti PD-1 antibodies. That’s also shown in KEYNOTE-028 where one patient was known to have MSI and experienced a partial response. On the other hand, MSS tumours are not hypermutated and patients with these tumours (also called mismatch repair proficient) are not good candidates for checkpoint inhibition.

What can we conclude from these preliminary results? 

Van Cutsem: The main message of the KEYNOTE-028 study is that the selection of patients is key to getting a response to the drug in colorectal cancer. The investigators included one patient with an MSI tumour and this patient responded but the other patients probably had an MSS tumour. It seems that these MSS patients do not respond very well to immune therapy with checkpoint inhibitors. If we bring these findings together with the NEJM paper we now know that pembrolizumab has an activity in colon cancer but probably only in patients with an MSI tumour. 

Another relevant point raised by this study stems from the fact that the researchers examined PD-L1 expression. It seems logical that patients with a high PD-L1 expressing tumour would have a high likelihood of responding to pembrolizumab. But this was not the case in this study, probably because of methodological issues. As yet there is no validated test to assess PD-L1 expression so the selection of these patients is not adequate.

Squamous cell carcinoma of the anal canal is a completely different entity to adenocarcinoma of the colon and it seems that there is a bit more activity with pembrolizumab. This is an unmet need population with few treatment options. In this relatively small study of 47 patients there was not a spectacular response, but there was a very interesting response rate and activity of pembrolizumab in these squamous cell tumours. 

How do the results of immunotherapy in lower GI cancers compare to other types of cancers?

Van Cutsem: In colon cancer, if you look at all comers, the results are poor compared to melanoma and lung cancer for instance. But the art of oncology is of course to select patients. If you look at the subgroup of patients with an MSI tumour, there the activity in colon cancer compares well. In anal cancer there is some activity which is better than for the other targeted agents in anal cancer. There is a bit less activity compared to melanoma if you just consider the blunt response rate. But there will again be a subgroup of patients in whom this will be further confirmed as a new treatment option. So yes, compared to other tumours there is a place for immunotherapy in lower GI cancers. But in colon cancer it’s only a subgroup (5% of the tumours) which is much less than in melanoma.

So can we conclude that we are approaching the immunotherapy era in lower GI cancers?

Van Cutsem: Probably yes, but only in specific subgroups. Immune therapy is coming in colorectal cancer for MSI patients. 

Squamous cell cancers of the anus are rare, much rarer than adenocarcinoma of the colon. For this specific, but relatively infrequent tumour, PD-1 blockade with pembrolizumab may bring a new treatment option because there are not many other drugs showing activity.

In the future, I expect the results we have seen with pembrolizumab in MSI colorectal cancer to be further confirmed. And we may also see combination treatments of other immune therapies, for instance combinations with CTLA-4-blocking antibodies and/or with angiogenesis inhibitors. 


1Abstracts presented at ECC 2015, held 25–29 September in Vienna, Austria:

500: Pembrolizumab (MK-3475) for PD-L1–positive squamous cell carcinoma (SCC) of the anal canal: Preliminary safety and efficacy results from KEYNOTE-028. P.A. Ott, USA. Sunday 27th September 2015 – 17:05-18:55 Proffered Paper Session HALL D1

502: Pembrolizumab (MK-3475) for patients (pts) with advanced colorectal carcinoma (CRC): Preliminary results from KEYNOTE-028. B.H. O’Neil, USA. Sunday 27th September 2015 – 17:05-18:55 Proffered Paper Session HALL D1

2 Le DT Uram JN Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372(26):2509–2520. doi: 10.1056/NEJMoa1500596. Epub 2015 May 30.

Information contained in this commentary was provided by the interviewee.