GI Cancer Expert Is Searching for Big Leaps in Care

John L. Marshall, MD

In a research career that has spanned 3 decades, John L. Marshall, MD, has participated in all phases of clinical trials. He was at the forefront of research into immunotherapy and targeted therapy for gastrointestinal (GI) cancers and helped to establish standards of care for patients with metastatic colorectal cancer (CRC). He explored vaccines for CRC in the late 1990s.

But his focus sharpened in 2006 after his wife, Liza Marshall, was diagnosed with stage IIIA near-inflammatory triple-negative breast cancer. She’s healthy today, but he was frustrated by the unanswered questions in cancer care. “We should be in more of a hurry to cure cancer,” Marshall said in Off Our Chests: A Candid Tour Through the World of Cancer, a book he coauthored with his wife about her cancer journey.“I could not understand why there was no sense of panic among the research and advocacy communities.”

Since then Marshall has concentrated on early-phase trials, particularly genomic and biomarker studies, looking for giant leaps in GI cancer care rather than small steps. He has yet to find the major advance he’s looking for but believes that investigators are getting closer. In February, the American Society of Clinical Oncology (ASCO) named molecular profiling in GI cancers as the 2021 Advance of the Year in oncology.¹

“The HER2-positive story gets better and better. The IO [immuno-oncology] story gets better and better across different diseases,” he said in an interview with OncologyLive^®. “There was clear evidence of progress in each of the fronts but we’re still waiting on that big sea change. We don’t have it quite yet.”

Marshall has made it his mission to improve the lives of patients with GI cancers through innovative research, personalized medicine, and focused advocacy at Georgetown University’s Otto J Ruesch Center for the Cure of Gastrointestinal Cancer, which he helped establish in 2008. He serves as director of the center, holds the Frederick P. Smith Chair of Medicine, and is chief of the Division of Hematology/Oncology with Lombardi Comprehensive Cancer Center at Georgetown University Medical Center in Washington, DC.

He some of the most interesting trials being conducted at the moment involve treating patients in basket studies exploring agents targeting narrower and narrower genetic mutations. Investigators are zeroing in specifically on the KRAS G12C mutation, which is associated with a poor prognosis and lack of response to standard treatments. The mutation is found in approximately 14% of lung adenocarcinomas, 3% to 4% of CRCs, and 1% to 2% of other solid tumors.²

Although there are no approved therapies targeting the KRAS G12C alteration, several agents are advancing in clinical development, notably in non–small cell lung cancer and CRC. Adagrasib (MRTX849) monotherapy resulted in a 17% objective response rate (ORR) among 18 evaluable patients with advanced CRC harboring the mutation, according to findings from the phase 1/2 KRYSTAL-1 trial (NCT03785249).² Sotorasib (AMG 510) demonstrated a 7.1% ORR across dosing levels among 42 patients with heavily pretreated metastatic KRAS G12C­–mutated CRC, according to findings from the phase 1/2 CodeBreak 100 trial (NCT03600883).³

“Right molecular biomarker, right drug, best outcome,” Marshall said. “That’s the coolest thing that’s happening in oncology.”

SOGO^®: Getting the Latest on GI Cancer

Marshall will share his insights into emerging developments across the GI cancer field during the 6th Annual School of Gastrointestinal Oncology^® (SOGO^®) hosted by Physicians’ Education Resource®, LLC (PER^®) on March 20. Marshall is cochairing the meeting with Michael A. Choti, MD, MBA, division chief of surgery at Banner MD Anderson Cancer Center in Gilbert, Arizona, and an internationally recognized pancreatic-hepatobiliary surgeon and surgical oncologist.

The virtual multitrack symposium features general session presentations on important concepts and debates, educational tracks exploring locoregional treatment of GI cancers and management of advanced GI cancers, SOGO^® seminars, and select case-based discussions on multidisciplinary, real-world management of GI cancers.

“We have really an amazing faculty that emphasizes the latest data in GI cancer, the innovative new stuff that you’ve got to know for practice whether you’re an oncologist, a surgeon, or a radiation physician. We encourage all of those folks to attend,” Marshall said. “We also have a heavy emphasis on cases so that we can really walk through some of the challenging issues that we face and make sure everybody’s up to speed on how to deliver the best care for GI cancer.”

Marshall said the program always includes hot topics in cancer care. This year, organizers felt that the most important subject in cancer is racial inequalities in outcomes and access to care. Marcus S. Noel, MD, associate professor of medicine and codirector of the Clinical Research Management Office at Georgetown University Medical Center, is scheduled to deliver a keynote address exploring racial disparities in GI cancer.

“The second major volatile topic that’s out there is the cost of health care,” Marshall said. “We know that right now, there are ongoing discussions about balancing the rising cost of drugs and the need for innovation.”

The coronavirus disease 2019 pandemic has taught the medical world that it is vital to mobilize science in the pharmaceutical industry, Marshall noted. A high-end pharmaceutical network is critical to our existence and our health, but it’s also expensive. Medical innovations have improved care while also raising cost, which makes those innovations less accessible.

“That’s true around the world, too, if you think about the global impact of cancer,” Marshall said. “We in the United States, for all of our fussing about calling insurance [for coverage], we have unlimited access compared to the rest of world. So how do we get some of these novel approaches to those patients out there, particularly these breakthrough therapies for these smaller and smaller subgroups of molecularly defined patients? How do we get those patients the drugs that they need, wherever they are on the planet?”

Marshall said the SOGO^® program will also address the most important clinical issues of the day including molecular profiling, multidisciplinary care, neoadjuvant therapies, reducing toxicity, and combined modality approaches.

“We’re trying to put each other out of business,” Marshall said. “We’re trying to put the surgeons out of business with chemoradiation. We’re trying to put the radiation oncologist out of business with better chemo and better surgery. We’re trying to make the same outcomes or better affect patients less, with fewer adverse effects, shorter treatments, and less toxicity.”

The program aims to provide a comprehensive overview of GI cancer in just a few hours. “At the end of that day, you’re going to know everything [you need] to know and be totally up to speed on the management of all the GI cancers,” Marshall said.

Molecular Profiling Is the Key

During the course of his career, Marshall has served as principal investigator for more than 100 clinical trials in CRC and other GI malignancies. His work on novel therapies includes studies of erlotinib (Tarceva), the EGFR inhibitor; obatoclax (GX15-070), a BH3 mimetic that binds to a broad spectrum of BCL2 regulator proteins; and dendritic cell vaccines for CRC. He also has been a leading investigator into chemotherapy regimens.

In 2003, Marshall was part of a team that established FOLFOX4 (leucovorin, 5-fluorouracil [5-FU], and oxaliplatin) as the standard of care for second-line treatment of metastatic CRC. The combination improved response rate, time to tumor progression, and tumor-related symptoms compared with 5-FU/leucovorin. At the time, there was no effective second-line therapy for patients with progressive metastatic CRC.⁴

Three years later, he coauthored the findings demonstrating a survival benefit for FOLFIRI (leucovorin, 5FU, and irinotecan) and FOLFIRI plus bevacizumab (Avastin) in patients with treatment-naïve metastatic CRC. In findings published in the Journal of Clinical Oncology, the median progression-free survival (PFS) was 7.6 months for patients assigned FOLFIRI compared with 5.9 months for those assigned to irinotecan plus bolus FU/leucovorin (mIFL) and 5.8 months for those assigned to irinotecan plus oral capecitabine (CapeIRI). After adjusting the protocol to add bevacizumab, the median overall survival (OS) was not reached for the FOLFIRI/bevacizumab combination compared with 19.2 months for mIFL/bevacizumab (P = .007).⁵

In 2015, he established and directed the Precision Oncology Alliance, a collaborative network of more than 40 cancer centers working with Caris Life Sciences, a Texas-based company, to study the impact of molecular profiling on cancer research, value, and outcomes.⁶

Marshall said he has watched enviously as his colleagues who treat patients with lung cancer leverage an expanding menu of molecularly targeted therapies. Recently, however, the GI field has made progress toward several new targets recently, including BRAF and HER2. The 2 advances specifically cited in the ASCO report were FDA approvals for fam-trastuzumab deruxtecan-nxki (Enhertu) for previously treated locally advanced or metastatic HER2-positive gastric or gastroesophageal adenocarcinoma, and for pembrolizumab (Keytruda) for the first-line treatment of patients with unresectable or metastatic CRC that is microsatellite instability–high (MSI-H) or mismatch repair–deficient.¹

Such developments highlight the impact of tumor profiling. “It is so important for us as oncologists to make sure we’re doing the best molecular testing,” Marshall said. “You need to do it right from the beginning….so that you know all of the chess pieces on the chessboard because they’re not common. And unless you look under every single rock, you’ll never find it.

“With so many new approvals in this space, [there are] so many options for all of these patients that deliver not just minor benefits, but major benefits,” he added. “You can’t miss these opportunities. They won’t come every day. But when you find them, it’s pretty cool.”

Meanwhile, Marshall said much work remains in delineating early intervention strategies throughout the GI cancer field. “In the adjuvant and neoadjuvant world, I think we are all struggling with how to incorporate things like microsatellite instability into neoadjuvant approaches,” he said. “On the one hand, we have data that says chemotherapy is harmful in the space. On the other hand, we don’t know any better—we don’t have IO approval in that space for neoadjuvant approaches. So those studies are going to be critical to how we approach those patients from the beginning, not just in the metastatic setting.”

Not all patients with GI cancers are deriving benefit from frontline therapy with immune checkpoint inhibitors. For some, chemotherapy induces better results, at least for a short time. Marshall said he is looking forward to results from trials studying combinations of checkpoint inhibitors with chemotherapy or other immunotherapy or agents in earlier lines of therapy.

“We have to remember, first, that it was only 10 years ago when no one on the planet except for a few people thought immunotherapy was ever going to have an impact at all,” he said. “We thought that was crazy. Why bother? And it was really these checkpoint inhibitors that were a breakthrough. Now we’re saying, ‘How come they’re not working everywhere?’ We should get them to work in every cancer and that’s not happening so far.”

One of the few GI cancers where immunotherapy has shown signs of making a difference is hepatocellular carcinoma (HCC), traditionally one of the most difficult cancers to treat. Since 2017, the PD-1 inhibitors pembrolizumab and nivolumab (Opdivo) have gained approvals as monotherapy in second-line HCC settings; nivolumab also is indicated in combination with ipilimumab (Yervoy) for previously treated patients with HCC.⁷

The FDA is now taking another look at 2 of those approvals, which were granted under the agency’s accelerated pathway. The monotherapy approvals for pembrolizumab and nivolumab in HCC are among the immunotherapy indications that will be reevaluated as part of an industry-wide review of accelerated approvals in oncology “in which confirmatory trials did not confirm clinical benefit,” according to the FDA. Those indications are among 6 that will be discussed during an Oncologic Drugs Advisory Committee public hearing scheduled from April 27 to 29. An indication for pembrolizumab in patients with recurrent locally advanced or metastatic gastric or gastroesophageal junction adenocarcinoma that express PD-L1 and has progressed after 2 or more prior lines of therapy also will be reviewed.⁸ Other indications for these agents are not affected.

Meanwhile, immunotherapy has been gaining ground in earlier therapeutic settings in HCC. In May 2020, the FDA approved atezolizumab (Tecentriq), a PD-L1 inhibitor, in combination with bevacizumab for patients with unresectable or metastatic HCC who have not received prior systemic therapy. The approval was based on findings from the phase 3 IMbrave150 trial (NCT03434379), in which the atezolizumab regimen showed an OS benefit compared with sorafenib (Nexavar).

The median OS was not reached for participants who received with atezolizumab followed by bevacizumab vs 13.2 months (95% CI, 10.4-not estimable) in those treated with sorafenib monotherapy (HR, 0.58; 95% CI, 0.42-0.79; P = .0006). Estimated median PFS was 6.8 months (95% CI, 5.8- 8.3) vs 4.3 months (95% CI, 4.0-5.6), respectively (HR,0.59; 95% CI, 0.47-0.76; P < .0001).⁹

Findings from the phase 2 KEYNOTE-224 trial (NCT02702414) presented at the 2021 Gastrointestinal Cancers Symposium showed that pembrolizumab induced durable antitumor activity and demonstrated a survival benefit for patients with treatment-naïve HCC.

In 51 patients treated with pembrolizumab, the ORR was 16% (95% CI, 7%-29%), which was comprised exclusively of partial responses. The median duration of response was not reached (range, 3-20+ months), with 70% of patients estimated as having a response duration of 12 months or more. The 12-month PFS was 24%, with a median of 4 months (range, 2-6). The median OS was 17 months, with a 12-month rate of 58%.¹⁰

“What we are seeing in liver cancers and bile duct cancers and some others [is that] combinations of therapies are resulting in improved outcomes from immune therapies,” he said. “The Holy Grail right now is in microsatellite-stable colon cancers and pancreas cancers. These are tough nuts to crack with regard to immunotherapy.”

Machine Learning Is the Future of Cancer Care

Marshall said artificial intelligence (AI) is, surprisingly, one of the biggest developments driving improvements in cancer care. Automated algorithms can extract meaningful patterns that provide practical knowledge and change the way in which treatments are developed, patients are classified, and diseases are studied.

Using computers to analyze genomic data from cancer patients, investigators can identify up to 100,000 genomic mutations for each tumor sample, although the clinical significance is unclear.^11,12 AI applications in imaging, genomics, and personalized medicine are expected to expand in multiple oncology settings.¹² With more and more assays collecting more and more information, AI will be vital to interpretating and applying the data.

Marshall was part of a team that in February 2021 published findings from a study evaluating a machine-learning approach called FOLFOXai applied to clinical and next-generation sequencing data from a real-world evidence (RWE) dataset.

Investigators collected RWE outcomes data from the Caris Life Sciences Precision Oncology Alliance registry and insurance claims data from more than 10,000 physicians. They then conducted a blinded retrospective-prospective analysis of samples from patients enrolled in the phase 3 TRIBE2 study (NCT02339116). Next-generation sequencing analysis was performed for further clinical validation.

In TRIBE2, patients with unresected metastatic colorectal cancer (N = 679) were assigned to FOLFOX/bevacizumab (Avastin) followed by FOLFIRI/ bevacizumab after progression or FOLFOXIRI/bevacizumab followed by the reintroduction of the same regimen after progression.

Blinded analysis showed that FOLFOXai was predictive for overall survival in both oxaliplatin-containing treatment arms of TRIBE2 (FOLFOX; HR, 0.629; P = .04; FOLFOXIRI, HR, 0.483; P = .02). FOLFOXai also predicted treatment benefit from oxaliplatin-containing regimens in patients advanced esophageal/gastro-esophageal junction cancers and pancreatic ductal adenocarcinoma.¹³

“We now have big data sets, huge sets of molecular data and clinical outcomes. We’re not smart enough, but we have machines that are using AI machine learning to say what [clinical] profile predicts what treatment we should give,” Marshall said. “By using this profiling, I can tell you that you’re better with this one [drug] than this one. It’s going to make us smarter. So in this confusing world of oncology, where we have multiple choices and lines of therapy, my hope is that this AI technology will in fact make us smarter.”

References

1. Smith SM, Wachter K, Burris HA, et al. Clinical cancer advances 2021: ASCO’s report on progress against cancer. J Clin Oncol. Published online February 2, 2021. doi:10.1200/JCO.20.03420
2. Johnson ML, Ou SHI, Rybkin II, et al. KRYSTAL-1: activity and safety of adagrasib (MRTX849) in patients with colorectal cancer (CRC) and other solid tumors harboring a KRAS G12C mutation. Presented at: 32nd EORTC-NCI-AACR Symposium on Molecular Targets and Therapeutics; October 24-25, 2020; Virtual. Abstract LBA-04. https://bit.ly/3mmx8kb.
3. Fakih M, Desai J, Kuboki Y, et al. CodeBreak 100: Activity of AMG 510, a novel small molecule inhibitor of KRASG12C, in patients with advanced colorectal cancer. J Clin Oncol. 2020;38:(suppl; abstr 4018). doi:10.1200/JCO.2020.38.15_suppl.4018
4. Rothenberg ML, Oza AM, Bigelow RH, et al. Superiority of oxaliplatin and fluorouracil-leucovorin compared with either therapy alone in patients with progressive colorectal cancer after irinotecan and fluorouracil-leucovorin: interim results of a phase III trial. J Clin Oncol. 2003;21(11):2059-69. doi:10.1200/JCO.2003.11.126
5. Fuchs CS, Marshall J, Mitchell E, et al. Randomized, controlled trial of irinotecan plus infusional, bolus, or oral fluoropyrimidines in first-line treatment of metastatic colorectal cancer: results from the BICC-C Study. J Clin Oncol. 2007;25(30):4779-86. doi: 10.1200/JCO.2007.11.3357
6. Caris Life Sciences announces Vall d´Hebron Institute of Oncology joins its Precision Oncology Alliance. News release. Caris Life Sciences; December 1, 2020. Accessed March 4, 2021. https://bit.ly/3kLBKk8
7. Hematology/oncology (cancer) approvals & safety notifications. FDA. Updated March 4, 2021. Accessed March 4, 2021. https://bit.ly/3uUFJzs
8. FDA in brief: FDA Oncologic Drugs Advisory Committee to review status of six indications granted accelerated approval. FDA. March 11, 2021. Accessed March 11, 2021. https://bit.ly/3cgG1bm
9. FDA approves atezolizumab plus bevacizumab for unresectable hepatocellular carcinoma. FDA. Updated June 1, 2020. Accessed March 4, 2021. https://bit.ly/3c4ziBk
10. Van Laethem JL, Borbath I, Karwal M, et al. Pembrolizumab (pembro) monotherapy for previously untreated advanced hepatocellular carcinoma (HCC): Phase II KEYNOTE-224 study. J Clin Oncol. 2021;39(suppl 3):297. doi:10.1200/JCO.2021.39.3_suppl.297
11. Zehir A, Benayed R, Shah RH, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med. 2017;23:703‐713.
12. Shimizu H, Nakayama KI. Artificial intelligence in oncology. Cancer Sci. 2020;111(5):1452-1460. doi:10.1111/cas.14377
13. Abraham JP, Magee D, Cremolini C, et al. Clinical validation of a machine-learning-derived signature predictive of outcomes from first-line oxaliplatin-based chemotherapy in advanced colorectal cancer. Clin Cancer Res. 2021;27(4):1174-1183. doi: 10.1158/1078-0432.CCR-20-3286