Anthony El-Khoueiry, MD
By 2025, Anthony El-Khoueiry, MD, expects to see a larger menu of approved drugs and drug combinations for the treatment of advanced hepatocellular carcinoma (HCC), better use of molecular and immune signatures to guide treatment decisions, and, possibly, the chance to cure intermediate- and advanced-stage disease.
El-Khoueiry, assistant professor of clinical medicine at the Keck School of Medicine at the University of Southern California and director of clinical translation at the Southern California Clinical and Translational Science Institute, closed out the 2020 HCC-TAG Conference by predicting how the HCC treatment paradigm might change over the next 5 years.
At present, physicians treating patients with HCC have 2 standard first-line options (sorafenib [Nexavar] and lenvatinib [Lenvima]); another potential first-line option in atezolizumab (Tecentriq) plus bevacizumab (Avastin); and multiple second-line therapies (regorafenib [Stivarga], cabozantinib [Cabometyx], ramucirumab [Cyramza], pembrolizumab [Keytruda], and nivolumab [Opdivo]), some of which can be used in the third line (nivolumab and cabozantinib). El-Khoueiry expects that armamentarium to grow over time.
“In my opinion, immune checkpoint–based combinations in the first line will become the standard,” he said. “However, is this needed for all patients? There are toxicity concerns with these combinations, both from a safety perspective but also financial. And can biomarkers help us select which patients may benefit from single agent therapy in first line?”
El-Khoueiry foresees a “huge focus” on enhancing immune checkpoint activity with new targets and novel combinations. Physicians and investigators will use biology to address mechanisms of resistance to primary and secondary immune checkpoint inhibition.
He added that post-immune checkpoint therapy, what to do after administering atezolizumab/bevacizumab or pembrolizumab plus lenvatinib, will become an increasingly important question. “It's going to be a lot of arbitrary usage of existing drugs, but I think we should try to work toward some rational sequences,” El-Khoueiry said. “There will be a lot of drug development opportunities in that space.”
El-Khoueiry said drug development should do a better job of addressing molecular heterogeneity. Much of the research in HCC has focused on the VEGF pathway, the MAP kinase pathway, and FGFRs. Investigators need to broaden their scope and start leveraging biologic subtypes to drive drug development and expand clinical trials.
“We need to start improving and enriching the patient population for studies,” he said.
Clinical evidence shows that patients who respond to treatment “do fantastic.” El-Khoueiry cited a subanalsysis of the CheckMate-040 trial presented at the 2018 Gastrointestinal Cancers Symposium which showed that the best responders to nivolumab also had the best survival. The median overall survival (OS) was not reached for the 22 patients who had a complete response and all were alive at 18 months. In contrast, the median OS for nonresponders (n = 59) was 8.9 months and 33 had died by 18 months.1
“If we can shift more patients to become those deep, long-lasting responders, whether with single agents or combinations, we’re going to get more cures,” El-Khoueiry said. “One way to do this is to have biomarkers to enrich that population.”
The goal of identifying a predictive biomarker has eluded investigators so far. PD-1 and PD-L1 do not make ideal biomarkers because investigators have not been able to connect the presence or absence of the proteins to response, but El-Khoueiry said PD-1/PD-L1 may have selection value as part of a group of biomarkers.
He noted that results from studies in other tumor types have shown that patients whose tumors with higher CD3 and CD8 infiltration have better outcomes, and investigators observed in CheckMate-040 that CD3 and CD8 infiltration correlated with response. El-Khoueiry added that investigators noticed an association between the presence of an inflammatory gene signature and response and OS.
Even as investigators struggle to determine who will respond to single agent anti–PD-1 therapy, they are getting closer to understanding which patients will not respond. Research into other tumor types has showed that the presence of WNT/β-catenin has an inverse relationship with tumor inflammation. Data published in 2019 showed that, for patients treated with immune checkpoint inhibitors, activating alteration WNT/β-catenin signaling was associated with lower disease control, shorter progression-free survival, and shorter median OS.2
El-Khoueiry acknowledged that there is more work to be done, but said the data show it is possible to use biomarkers to guide treatment in patients with HCC. To achieve the goal of personalized immunotherapy care, investigators will need to figure out, upfront, which tumors are inflamed, understand the challenges of treating the noninflamed tumors, and administer smart combinations based on that knowledge, he said.
“I think we have to start imagining a model like what has been proposed in renal cell carcinoma, where we think about categorizing patients into different groups depending on their molecular and immune makeup, then tailor therapy accordingly,” El-Khoueiry concluded.
- El-Khoueiry AB, Melero I, Yau TC, et al. Impact of antitumor activity on survival outcomes, and nonconventional benefit, with nivolumab (NIVO) in patients with advanced hepatocellular carcinoma (aHCC): Subanalyses of CheckMate-040. J Clin Oncol. 36, 2018 (suppl 4S; abstr 475). doi: 10.1200/JCO.2018.36.4_suppl.475.
- Harding JJ, Nandakumar S, Armenia J, et al. Prospective genotyping of hepatocellular carcinoma: clinical implications of next-generation sequencing for matching patients to targeted and immune therapies. Clin Cancer Res. 2019;25(7):2116-2126. doi: 10.1158/1078-0432.CCR-18-2293.