Three Trials From ASCO GI Renew Hope for Successful Treatment of Difficult GI Cancers

Mark A. Lewis, MD


March 08, 2022

This transcript has been edited for clarity.

This is Mark Lewis, MD, director of gastrointestinal (GI) oncology at Intermountain Healthcare in Salt Lake City, Utah. I want to summarize some takeaway points from the 2022 American Society of Clinical Oncology (ASCO) GI Cancers Symposium.

Many oncologists will tell you that the litmus test of a successful conference is how much it affects your practice, and I believe that data were presented at this meeting that will certainly affect practice immediately and give us a sneak preview of how our treatments may change in the future.

Three studies have particularly catchy trial acronyms, but I assure you that my interest in them is not just clever wordplay. This was meaningful research.

I'll start with KRYSTAL-1 — that's "crystal" with a "K" as a hint to what we're looking at. Can we make KRAS mutations "druggable?" KRAS mutations have famously, or infamously, been considered nonactionable mutations for many, many years, especially in the field of GI oncology. But recent advances in other areas, such as thoracic oncology, have raised our interest in seeing whether particular KRAS mutations can be exploited in otherwise refractory GI cancers.

KRYSTAL-1 looked at pancreatic ductal adenocarcinoma and other noncolorectal GI cancers that had previously been treated and specifically harbored the KRAS G12C mutation. Of the 42 patients in this phase 2 study, 12 had pancreatic ductal adenocarcinoma and 10 of those were evaluable. What's so remarkable in this fairly heavily pretreated population — with a median of two lines of therapy (ranging from one to five) — is that the disease control rate (DCR) with adagrasib treatment was 100%. The DCR is the sum of the response rate and stable disease, and that was 50%/50% exactly. The median progression-free survival was between 6 and 7 months.

I bring this up because for years it's been easy, even for GI oncologists, to adopt an attitude of therapeutic nihilism when it comes to pancreatic ductal adenocarcinoma; so often when we do sequencing on these tumors — 90% of the time, in fact — we find a KRAS mutation. But you're looking for the diamond in the rough among those KRAS mutations, the small minority that have G12C. And I think it'll be important to look out for those mutations and to consider adagrasib.

I was also quite taken by the TOPAZ-1 study. I believe TOPAZ is the first study to substantively improve on the first-line standard of care — gemcitabine and cisplatin — in advanced biliary tract cancer. Of course, ABC-02 established that doublet as the backbone of treatment more than a decade ago, and since then we've been trying to see whether the addition of another agent can meaningfully change the benefits we derive from gemcitabine/cisplatin. I think the answer here is a cautious yes.

This trial showed an improvement in median overall survival of 12.8 months in the durvalumab plus gemcitabine/cisplatin arm vs 11.5 months in the gemcitabine/cisplatin arm. The hazard ratio was 0.8, so a 20% reduction in risk of death with the addition of durvalumab on top of gemcitabine/cisplatin.

The discussant did a particularly fantastic job in pointing out the importance of a subset analysis. The real signal that needs to be further explored is that this combination may benefit Asian patients more than non-Asian patients. Regardless, this is exciting, and hopefully we can find not just which patients this fits, but which biomarkers can best guide us in treatment selection. We're all waiting with bated breath for results of the S1815 trial, to see if triplet chemotherapy will also unseat doublet in the treatment of advanced biliary cancers. That is a combination of gemcitabine/cisplatin and nanoparticle-bound paclitaxel in the SWOG trial.

Finally, I'd be remiss if I don't bring up the HIMALAYA trial. HIMALAYA looked at first-line dual immunotherapy in the setting of advanced hepatocellular carcinoma. Participants in this study were randomly assigned to STRIDE (single-dose tremelimumab and then regular-interval durvalumab), durvalumab alone, or sorafenib. The primary endpoint was overall survival (OS) and that was met. The median OS in the STRIDE arm was 16.4 months vs 16.6 months in the durvalumab arm vs 13.8 months in sorafenib control arm. Some critics will say that that's not the appropriate control arm in the aftermath of IMbrave150. However, at the time the trial was designed, I think sorafenib was a reasonable control.

This dual-immunotherapy approach is potentially giving us a therapeutic alternative for patients who would otherwise be excluded from exposure to atezolizumab/bevacizumab on the basis of varices, high risk for bleeding, or incompletion of upper endoscopy to aggressively exclude varices. So again, this widens the candidacy for immunotherapy.

Finally, and on a personal note and as I mentioned in a prior column, I enrolled a patient in HIMALAYA. He has passed away but asked me to posthumously convey the results of this study, whatever they were, to his widow. And the fact that this was a positive study when it matured was actually a reassurance to that very gracious woman. Her husband had said that if he could help even one person with his contribution to this study, then his participation would have been worthwhile. So I end wistfully, wishing this combination could have helped him more, but I'm glad that it may help other patients with advanced hepatocellular carcinoma in the here and now and in the future.

As always, ASCO GI gave me renewed hope. I'm grateful for all the researchers out there who dedicate themselves to some very difficult-to-treat diseases indeed. I believe that we can see immediate and potential impacts in this area.

This is Mark Lewis from Medscape.

Mark A. Lewis, MD, is director of gastrointestinal oncology at Intermountain Healthcare in Salt Lake City, Utah. He has an interest in neuroendocrine tumors, hereditary cancer syndromes, and patient-physician communication.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.