2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Syma Iqbal, MD, shares the main highlights from an IPC meeting, which centered on the evolving landscapes and importance of implementing best practices for patients with hepatobiliary cancers, locally advanced and advanced pancreatic cancer, advanced gastric/gastroesophageal junction cancer, and metastatic colorectal cancer.
Following years of negative clinical trials and little advancement, the incorporation of frontline immunotherapy regimens has reshaped the treatment landscape of several gastrointestinal (GI) malignancies, including upper GI cancers and hepatocellular carcinoma (HCC), said Syma Iqbal, MD, who added that the next hurdle to overcome is regarding optimal sequencing with the novel agents.
“The biggest [advance] in GI oncology has been the addition of immunotherapy,” Iqbal said in an interview with OncLive® during an Institutional Perspectives in Cancer (IPC) webinar on GI malignancies. “We have had VEGF inhibitor therapy in some format across tumors in GI oncology, including gastric and colon cancers, as well as HCC with the TKIs [tyrosine kinase inhibitors]. Now, by incorporating immunotherapy, we have made advances across these disease types in GI oncology.”
In the interview, Iqbal, cochair of the IPC event, shared the main highlights from the meeting, which centered on the evolving landscapes and importance of implementing best practices for patients with hepatobiliary cancers, locally advanced and advanced pancreatic cancer, advanced gastric/gastroesophageal junction (GEJ) cancer, and metastatic colorectal cancer (mCRC). At the Keck School of Medicine of the University of Southern California (USC), Iqbal is an assistant professor of medicine, assistant program director of the Oncology Fellowship Program, educational officer, and codirector of the Liver Cancer Program in the Division of Oncology.
Iqbal: This has been a big year for upper GI tumors. We’ve had several positive studies after a very long time of multiple negative trials. We’ve been able to incorporate checkpoint inhibitors into the frontline setting in HER2-negative and HER2-positive disease. That has been the biggest change we have had this year.
Also, in the adjuvant setting, checkpoint inhibitors have been approved for patients with GEJ and esophageal adenocarcinoma, having shown a doubling in disease-free survival. That was a huge step forward for upper GI tumors.
Although it has been a good year, [the advances] have thrown a wrench into ongoing studies and studies that were designed [prior to these changes]. Oncologists who are working on those studies need to take a step back to figure out how we will deal with this new standard to still investigate new [strategies] for the future.
Many of the protocols are being amended to incorporate immunotherapy, at least into the standard-of-care arms. The trials that don’t include immunotherapy as standard of care are going to suffer. We are already seeing that because patients are coming in and asking for immunotherapy. Patients are aware of it; they see commercials on television. Now that there are data to support it, it is a hard discussion with patients to offer them something that doesn’t include [immunotherapy].
More is likely to come on that as these approvals get reviewed, but that is correct. The National Comprehensive Cancer Network guidelines came out prior to the FDA approvals. When the guidelines for [nivolumab (Opdivo)] in HER2-negative patients were proposed, they were based on CheckMate649 [NCT02872116], where the primary end point was a combined positive score [CPS] of 5 or greater. The guidelines reflected that and are recommended for patients who have a CPS of 5 or greater. The FDA gave a broad approval [to nivolumab plus chemotherapy] for all-comers as it appeared that all-comers benefitted. However, if we look at the subgroups, the benefit in the CPS less than 5 population is quite small, if at all.
This will continue to evolve because there may be some modifications around the approval. For now, [immunotherapy] is something that can be offered to all patients. At most academic centers in the upper GI oncology world, it seems that patients with a CPS less than 5 are receiving other therapeutics or nonimmunotherapy combinations.
We would like to hear more about trastuzumab deruxtecan in the second-line setting. Although [the approval was based on] phase 2 data, we also have data in refractory disease. We haven’t seen a positive trial for HER2 suppression after firstline HER2-directed therapy plus chemotherapy. That was also a promising step forward for second-line therapy for patients with HER2-positive disease that offers continued benefit.
The first thing that he emphasized was profiling these tumors to know exactly how best to treat specific populations. We’ve learned a lot more about the BRAF-mutant population with the use of encorafenib [Braftovi]. The combination [of encorafenib/ binimetinib (Mektovi)/cetuximab (Erbitux)] showed more toxicity but not necessarily an improvement in outcome.
The other relevant information from his talk was regarding sequencing of the many therapeutics that we have. There has been hesitancy to use drugs like regorafenib [Stivarga] and TAS-102 [trifluridine/tipiracil; Lonsurf] based on the randomized data that we have because oncologists feel there isn’t much benefit to using these drugs. However, the regorafenib data that were presented [were encouraging], particularly with the ReDOS trial [NCT02368886] that showed that starting patients on a low dose of regorafenib [allows them to] tolerate the therapeutic better. This is likely because patients were able to stay on therapy.
We don’t pay enough attention to sequencing in CRC, and that is particularly so when going to the refractory setting to choose regorafenib or TAS-102. Dr Lenz’s presentation gave strong support for the use of appropriately dosed regorafenib up front in patients with good performance status and adequate albumin. Those patients tend to do better when started with regorafenib in the refractory setting vs starting with TAS-102.
For locally advanced and borderline resectable disease, we as a community have used modified FOLFIRINOX [mFOLFIRINOX], assuming it was better than gemcitabine/nab-paclitaxel. We learned from the SWOG-S1505 trial [NCT02562716] that in the perioperative setting, there was no difference in outcome whether patients got mFOLFIRINOX or gemcitabine/nab- paclitaxel up front.
Something about locally advanced disease is that we still must incorporate assessment of the whole patient when choosing a regimen. Otherwise, it seems that radiation remains controversial, and these patients are treated similarly to patients in the advanced-stage setting. We don’t have an optimal pathway for all-comers. These are unique, tailored approaches to individual patients in the locally advanced setting.
To be honest, most people are still more inclined to use mFOLFIRINOX. We are swayed by the metastatic data, even though the local randomized data look similar. As we are evaluating patients, we are still looking at other factors like performance status, comorbidities, and [patient preference], knowing that it won’t necessarily impact outcome if we use gemcitabine/nab-paclitaxel in that patient population. We always feared that [gemcitabine/nab-paclitaxel] was inferior based on the metastatic data, but they’re 2 separate trials.
For a long time, we have become comfortable with the use of FOLFIRINOX or mFOLFIRINOX and gemcitabine/nab- paclitaxel as standard regimens in this disease. The steps forward are related to profiling these tumors, specifically [to identify] BRCA mutations. This is the first time now in pancreas cancer where profiling has impacted outcome. The use of PARP inhibition with olaparib [Lynparza] as maintenance therapy in the BRCA-[mutated] population is relevant to care, so profiling should occur prior to starting patients on therapy. The protein profiling is also assessing for the very small number of patients who have microsatellite instability–high disease so we can try to incorporate immunotherapy.
It’s very disappointing that the study with the CPI compound that targeted the Krebs cycle was negative. People thought that was going to be positive, as we always do in the early phase when the data look so promising. That was disappointing, but it will be interesting to see the formal presentations to see if any subgroup benefited from that agent.
We’ve seen huge changes in hepatobiliary cancer. We have several positive trials now in the frontline setting combining immunotherapy with VEGF inhibitors, both using agents such as bevacizumab [Avastin] and TKIs. That has been a big change in how we approach clinical care in HCC. We now have options in the frontline setting for these patients, including the combination of atezolizumab [Tecentriq] plus bevacizumab and immunotherapy plus CTLA-4 inhibitors.
What this has done in HCC is confuse us because it has created an issue where we don’t know how to sequence these patients or pick the best first-line regimen. Why do we need another frontline regimen when we have atezolizumab/bevacizumab, which looks good? [It’s because] many of these patients may not be candidates for bevacizumab because of varices, history of recent bleeds, or platelet counts. Having an alternative frontline option opens the door for therapeutics to a population that wouldn’t have been candidates for [atezolizumab/bevacizumab].
The challenge now in HCC is [determining] how to sequence the TKIs and what to use in the second-, third-, and fourth-line settings. Many of these patients with Child-Pugh A cirrhosis can go through multiple lines of therapy. Any time we have options, it confuses us more, but it is a great place for us to be. HCC has had huge changes and there are more to come.
Given the heterogeneity of the patients who present with comorbidities related to varices, bleeding, or portal hypertension, having an alternative to a risky drug like bevacizumab [is important]. Now we just have to determine what do we do with everything else.
In cholangiocarcinoma, we have also seen significant changes. What is most relevant here is the importance of profiling these tumors. We know that FGFR mutations and IDH mutations are relevant to the treatment of this disease. This wasn’t a disease that most folks were profiling before, but now that we have interesting data looking at moving immunotherapy into the frontline setting for these patients, [profiling is important]. We are also awaiting results from a phase 2 trial [NCT02392637] evaluating the triplet of nab-paclitaxel plus standard gemcitabine/cisplatin. We will see whether there is a further change to our standard of care up front in cholangiocarcinoma.