COMPASS: Genomic Profiling Could Guide Treatment in Advanced Pancreatic Cancer

February 2, 2018

In this interview we discuss preliminary results of the COMPASS trial, which used whole-genome sequencing and RNA sequencing to analyze the tumors of patients with advanced pancreatic ductal adenocarcinoma to identify possible first-line treatments.

As part of our coverage of the 2018 Gastrointestinal Cancers Symposium, held January 18–20 in San Francisco, we are speaking with Kyaw L. Aung, MBBS, PhD, a clinical fellow at the Princess Margaret Cancer Centre in Toronto. Dr. Aung presented updated data from the COMPASS trial (abstract 211), which used whole-genome sequencing and RNA sequencing to analyze the tumors of patients with advanced pancreatic ductal adenocarcinoma to identify possible first-line treatments.

-Interviewed by Anna Azvolinsky

Cancer Network: First, can you tell us about the design of the COMPASS study? What is unique about the study, either as a pancreatic cancer study or as a cancer trial in general?

Dr. Aung: The COMPASS study was designed to acquire a fresh tumor biopsy from patients with advanced pancreatic cancer, and both locally advanced and metastatic patients were included in this trial. Before starting first-line chemotherapy with either modified FOLFIRINOX or gemcitabine and nab-paclitaxel, we performed whole-genome sequencing and RNA sequencing, prospectively. The study’s aim was mainly to acquire the whole-genome sequencing data 8 weeks from the time of tumor biopsy, so that that information can be used to influence patient management. One of the main secondary endpoints was to identify subsets of patients with unique genomic features who may benefit from personalized- or precision-medicine approaches.

Our study is unique in a few ways. The first thing is that most of the genomic sequencing studies in pancreatic cancer were done in early-stage, resectable cancer. Currently, data is very much limited for advanced-stage pancreatic cancer, mainly because it is very difficult to do molecular profiling studies in advanced disease. It is not easy to get good quality tumor material for whole-genome, RNA sequencing, or other comprehensive molecular profiling types of studies. So this was the challenge we really wanted to overcome. As you can see from other studies, most of the clinical data with large-scale genomic sequencing are retrospective and ours is a prospective trial-and with full clinical annotation-which is very powerful and vey much unique to COMPASS. We really are proud to achieve this in this study.

Cancer Network: What were the key results you presented from this update of the trial at the meeting?

Dr. Aung: So the primary objective of the study was to prove the feasibility-to show that it is possible to do this comprehensive molecular profiling prospectively, and in real time, in the pancreatic cancer clinic. We demonstrated this and the results are impressive. In 98% of the patients that we biopsied, whole-genomic sequencing was successful; in 95% of the patients, RNA sequencing was successful; and I think we achieved this result by team science. One of the features of the study was that we performed laser capture microdissection for tumor content enrichment, and that made a significant difference in getting good quality genomic results.

The main interesting result is the RNA sequencing identified two patient subtypes in pancreatic cancer, classical and basal-like, and we found that the classical subtype responded better to chemotherapy-this is the first prospective evidence that showed difference in response to chemotherapy in pancreatic cancer based on RNA subtype. Also, whole-genome sequencing identified a small subset of patients with unstable genomes, and they responded very well to modified FOLFIRINOX. The interesting thing is that these three patients did not have any germline mutations, so if we had not done whole-genome sequencing we might not have identified this subset, and we are quite excited to see these results.

A third important finding is that about 25% of the patients have genomic alterations that could be targeted with currently available drugs. A cautionary note on this though, whether targeting these genomic aberrations by using currently available drugs will make any difference in terms of clinical efficacy in pancreatic cancer remains very much unknown right now-we still need to do a lot more studies.

Cancer Network: What’s next for this trial? Are you recruiting more patients or doing more analyses and follow-up on these current patients?

Dr. Aung: As you know, these are our preliminary results that were presented at the meeting. Currently, the COMPASS study is still ongoing and we would like to recruit 180 patients as an exploratory dataset. We already recruited more than 100 patients, and we are hoping we will meet the target of 180 patients early next year or the middle of next year at the latest. We really want to update the data and consolidate our findings.

An important thing is that we are now trying to dig deeper to understand the biological mechanism behind these pancreatic ductal adenocarcinoma subtypes and by doing so, we really want to find subtype-specific therapies. I think rather than just subtyping superficially, we really need to understand the underlying biological mechanisms at play in these subtypes. We are working very hard on this and hopefully, with data from all 180 patients, we will be able to present a lot more than what is in these preliminary results.

Cancer Network: You mentioned that the trial met your expectations, at least in part, as you were able to obtain whole-genome sequencing data and RNA sequencing data from patients’ biopsies. Were there other expectations going into this trial for you and your colleagues, and how did that compare with the execution of the patient-sample analyses and findings? And what are the implications for doing these types of trials that rely on patients’ tumor-sequencing information?

Dr. Aung: Our expectation initially was that we would get the data in 80% of the patients that we biopsied, and we exceeded that expectation. Almost all patients could get their genomic data. In terms of the enrollment and patient-sample analysis, we are very pleased with the results. In terms of the findings of possible new therapies for pancreatic cancer based on the genomic results, I have to admit that we still need to do a lot more. We are still just scratching the surface and trying to find the biological mechanisms so that we could target them and also develop new therapies. I think our results definitely established that there are patients with unique subtypes that may benefit from personalized treatment and strategies. And these results need to be validated, which we will continue to do.

I think the implications of this study, in terms of using the genomic sequencing information, is very significant. As you probably know, the findings in most preclinical studies cannot really be translated into the clinic, so what we are focusing on is to translate genomic information into the pancreatic cancer clinic to have an impact on patient management. Obviously this is challenging, and we are at the very beginning, but we really believe that we are on the right track. The major impact of these kinds of studies and trials will be to change the treatment paradigm in advanced pancreatic cancer and finally to improve patient survival outcomes. We really believe that we can achieve that by using patients’ tumor sequencing information.

Cancer Network: Thank you so much for joining us today, Dr. Aung.

Dr. Aung: Thank you!