Investigators have achieved some success with immune checkpoint inhibitors and antiangiogenesis agents in patients with advanced gastroesophageal cancers, and exciting new agents targeting different tumor growth pathways are under investigation.
Yelena Y. Janjigian, MD
Progress is frustratingly slow in the development of therapies for advanced gastroesophageal cancers. However, investigators have achieved some success with immune checkpoint inhibitors and antiangiogenesis agents, and exciting new agents targeting different tumor growth pathways are under investigation, according to a panel of international experts.
During a recent OncLive Peer Exchange®, the panel discussed some of the trials that have shown mixed results or failed to improve outcomes for patients with gastric and gastroesophageal junction (GEJ) cancers. The panelists also shared their insights into the lessons learned from those trials, which are helping to guide research and improve patient selection. In addition, they discussed several agents being investigated in the second- and later-line settings that have led to survival gains as well as some promising novel strategies that are in various stages of development.
In the early 1900s, gastric cancer was one of the most commonly diagnosed cancers in the United States but is now no longer among the top 10. Despite that success, gastric cancer remains common worldwide, particularly in East Asia, where it is the third-most commonly diagnosed cancer.1-3 Most patients present with advanced unresectable or metastatic disease, which has few treatment options. Unlike the landscape for other solid tumors, treatment has not advanced far beyond chemotherapy, and survival gains have remained modest.Discovery of various mutations and other targets in patients with gastroesophageal cancers has led to the assessment of various novel agents. Some of the most actively investigated targets include HER2, PD-1, PD-L1, and angiogenesis.4 Although agents directed against these and other targets have shown clear benefit in other solid tumors, the results have been largely underwhelming in gastroesophageal cancers.
“Gastric cancer is a lukewarm tumor: It’s not a cold tumor, but it’s not a hot tumor,” Yelena Y. Janjigian, MD, explained. “The idea is that in gastric cancer, there is not a single pathognomonic driver because the epithelium has been exposed to so many insults over the patient’s years, resulting in DNA damage.” Subsequently, these tumors are likely to have multiple disease drivers, which may vary by geographic region and tumor subtype and change as the disease progresses, making these tumors elusive targets.
HER2: JACOB and Other Studies
HER2 positivity has been identified in approximately 15% to 20% of patients with gastric cancer.4 Although the addition of the HER2- targeted therapy trastuzumab (Herceptin) to chemotherapy in the first-line setting has shown a survival benefit for patients with HER2-positive advanced gastric or GEJ cancer, with a median overall survival (OS) of 13.8 months versus 11.1 months with chemotherapy alone,5 other strategies exploiting HER2 have not been as successful. This indicates significant differences in HER2 positivity between breast cancer and gastric cancer.
“[In the JACOB trial], pertuzumab [Perjeta] plus trastuzumab led to a small benefit, but this was not significant,” Eric Van Cutsem, MD, PhD, said. The JACOB trial was conducted based on the results of the CLEOPATRA trial in breast cancer, whose results showed a significantly reduced risk of progression or death when dual HER2 blockade (pertuzumab and trastuzumab) was added to chemotherapy (HR, 0.62).6 Dual HER2 blockade in JACOB did increase OS by 3.3 months, but this finding did not reach statistical significance.7 “Because of the differences in the role of HER2 in gastric versus breast cancer, unfortunately, the combination with pertuzumab and trastuzumab failed,” Van Cutsem said.
Differences have also been observed between breast and gastric cancer when HER2 blockade is used in second and later lines, Van Cutsem noted. In a Japanese study comparing weekly paclitaxel alone with weekly paclitaxel plus trastuzumab in patients with HER2-positive advanced gastric or GEJ cancer progressing during trastuzumab-containing therapy, continuation of trastuzumab failed to improve progression-free survival (PFS) or OS.8 In contrast, continuous use of trastuzumab beyond progression has shown benefit in HER2-positive metastatic breast cancer. For example, a post hoc analysis of trastuzumab as third-line therapy showed better postprogression survival for patients with breast cancer than for those not receiving this targeted treatment.9 However, there may also be geographic and other differences (eg, tumor type and location) regarding the continuation of trastuzumab beyond progression in patients with advanced gastric cancer. Although results from the Japanese study did not show a benefit, results from a study not discussed during the OncLive Peer Exchange® program but presented at the 2017 American Society of Clinical Oncology Gastrointestinal Cancers Symposium by French investigators did show a benefit with trastuzumab continuation in patients with advanced gastric adenocarcinoma, highlighting the complexity of gastric tumors.10
When HER2 expression is very focal in gastric cancer and many biopsies are needed to find it, HER2 may not be the major cause of malignancy, Janjigian suggested. “In those patients, the benefit from HER2-directed therapy is minimal, and they are the patients who are bringing down the survival curves on these studies,” she said. Patients with focal amplifications that are not high enough to be picked up by next-generation sequencing generally respond to HER2 blockade similar to the way patients with HER2-negative disease respond. Therefore, more careful patient selection is needed when implementing HER2 blockade in patients with gastric tumors, even in those found to have some HER2 positivity, she added.
PD-L1: KEYNOTE-061 Immunotherapy against PD-L1 has also shown mixed results in advanced gastric cancer. In September 2017, based on the results of the KEYNOTE-059 study, the FDA approved the anti—PD-1 immunotherapy pembrolizumab (Keytruda) for PD-L1–positive recurrent or advanced gastric or GEJ adenocarcinomas previously treated with 2 or more lines of chemotherapy. However, the results of the randomized, open-label, phase III KEYNOTE-061 study, which compared pembrolizumab with paclitaxel in patients with advanced gastric or GEJ cancer that progressed on first-line therapy with platinum and fluoropyrimidine, were not as favorable, Kohei Shitara, MD, said.11 The median OS was 9.1 months with pembrolizumab versus 8.3 months with paclitaxel.
However, KEYNOTE-061 also included PD-L1—negative patients with advanced gastric or GEJ cancer, and a post hoc analysis showed benefit with increased PD-L1 expression. “A higher PD-L1 expression showed a higher benefit [with pembrolizumab], with a hazard ratio of 0.64 compared with paclitaxel [in patients with a PD-L1 combined positive score (CPS) of ≥10]. This is a very remarkable benefit, so it should be grounds for additional studies,” Shitara said. The treatment with pembrolizumab also improved OS in patients with a CPS ≥5, with an HR of 0.73.11
The panelists also discussed the phase III JAVELIN Gastric 300 study, which randomly assigned patients in the third-line setting with gastric or GEJ cancer to receive the anti—PD-L1 antibody avelumab (Bavencio) or physician’s choice of chemotherapy (paclitaxel or irinotecan) or best supportive care if chemotherapy was not feasible.12 “The [results of the] study did not show benefit,” Van Cutsem said. In the primary analysis, the median OS was 4.6 months for avelumab and 5.0 months for chemotherapy, and the median PFS was 1.4 months versus 2.7 months, respectively. Avelumab, however, was better tolerated than chemotherapy.
However, it is unclear whether increased PD-L1 expression would lead to improved benefit with avelumab. JAVELIN Gastric 300 did not select patients according to biomarker. “There was no enrichment for PD-L1 or for EBV [Epstein-Barr virus] or MSI [microsatellite instability], so it was just all-comers. That’s an important message. That’s probably not the way to go in gastric cancer with checkpoint inhibitors; you have to select patients,” Van Cutsem said.
Stem Cell Inhibition: BRIGHTER Study
Cancer stem cell (CSC) inhibition is an evolving area in cancer research. CSC theory postulates that there is a subpopulation of phenotypically distinct cancer cells that may play an important role in tumor pathogenesis. Unlike normal actively dividing cancer cells, these cells are quiescent and cannot be killed with a cell-cycle poison like chemotherapy. In patients with pretreated advanced gastric and GEJ adenocarcinoma, the CSC inhibitor napabucasin (BBI608) is being studied in the BRIGHTER study as a second-line treatment in combination with paclitaxel.13 “Unfortunately, based on the data that are available to us at this point, the top-line result was negative with this study, and there was no benefit to adding this agent to paclitaxel in the secondline setting,” Peter C. Enzinger, MD, said. In the study, the median OS was 6.93 months for napabucasin plus paclitaxel and 7.36 months for placebo plus paclitaxel, and the median PFS was 3.55 and 3.65 months, respectively.13
Angiogenesis Inhibition: RAINFALL Study
VEGF and its receptors, particularly VEGF receptor 2 (VEGFR-2), play a critical role in tumor-associated angiogenesis. In the RAINFALL study, the VEGFR-2 IgG1 human monoclonal antibody ramucirumab (Cyramza) has produced underwhelming results as a first-line therapy in combination with chemotherapy in patients with metastatic gastric or GEJ adenocarcinoma, despite having shown more favorable results in the second-line setting.14
“[RAINFALL] looked at a platinum 5-FU [fluorouracil] frontline regimen, with or without ramucirumab, given in a different way than in subsequent lines of therapy,” Enzinger said. “The idea was to give higher doses of ramucirumab, so it was given on day 1 and day 8 on a 3-week cycle. Although there was a small, statistically significant improvement in progression-free survival [25% reduced risk of progression], there was no improvement in overall survival in this trial. Unfortunately, for that reason, the company that makes this agent is not pursuing a frontline indication.”
When those data are considered alongside the AVATAR and AVAGAST studies,15,16 both of which assessed the antiangiogenic agent bevacizumab (Avastin) with chemotherapy in the frontline setting, there is no hard evidence that a frontline angiogenesis strategy works in advanced gastric cancer, Enzinger said. “I think we’re just going to have to go back to the drawing board, and perhaps we can find better agents or [learn whether these agents are] more of a second- or third-line drug,” he suggested.The most recent advances in advanced gastric cancer have been in the second and subsequent lines. Yet historically, just 20% of patients received second-line therapy as a result of the aggressiveness of gastric cancer, the subsequent rapid decline in performance status, and the lack of robust evidence to support salvage chemotherapy.4 However, in more recent phase III trials, up to 40% of patients with advanced gastric cancer in Europe and 75% of such patients in Japan have proceeded to second-line therapy, which has led to a growing body of evidence supporting the safety and efficacy of various salvage therapy approaches in even heavily pretreated patients.4
Angiogenesis Inhibition: RAINBOW and REGARD Studies
Although ramucirumab did not show benefit in the first line in the RAINFALL study, it did show benefit in the second line in both the RAINBOW and REGARD trials, the panel noted. In RAINBOW, ramucirumab was assessed in combination with paclitaxel in patients with advanced gastric or gastroesophageal cancer who had been pretreated with the doublet of fluoropyrimidine plus a platinum.17 “There was a clear survival benefit of more than 2 months,” Van Cutsem said. The OS was 9.6 months for ramucirumab plus paclitaxel versus 7.4 months for placebo plus paclitaxel.
Results of the REGARD study showed a survival benefit with ramucirumab monotherapy, but it was more modest than with the ramucirumab and paclitaxel combination in the RAINBOW study. In REGARD, patients with advanced gastric or GEJ adenocarcinoma and disease progression after first-line platinum- or fluoropyrimidine-containing chemotherapy were randomly assigned 2:1 to receive best supportive care plus ramucirumab or placebo.18 The median OS was 5.2 months in the ramucirumab arm and 3.8 months in the placebo arm.
The evidence for ramucirumab alone or ramucirumab plus paclitaxel would support putting these regimens in the guidelines, Van Cutsem said. “Ramucirumab plus paclitaxel is for fit patients and the standard option in the second-line treatment because mean survival was around 9 months for the combination.”
Chemotherapy Compound: TAGS Trial
In the phase III TAGS trial, which was not discussed by the panel, trifluridine/tipiracil (TAS-102; FTD/TPI; Lonsurf) reduced the risk of death by about one-third compared with placebo in patients with heavily pretreated gastric or GEJ cancer. Findings were reported at the 2018 European Society for Medical Oncology Congress, after the OncLive Peer Exchange® took place.19
Median OS, the primary endpoint, was 5.7 months for patients assigned to FTD/ TPI compared with 3.6 months for patients randomized to placebo (SIDEBAR with full results, page 85). The 12-month OS rate for the FTD/TPI group was 21% versus 13% for the placebo group. In October, trifluridine/
Anti—PD-1: ONO-4538 Study
“What everybody is hot on now—aside from being able to use antiangiogenic agents for gastric cancer—is immunotherapy,” said Johanna C. Bendell, MD. One promising checkpoint inhibitor trial, ONO-4538, included Japanese, Korean, and Taiwanese patients and evaluated the safety and efficacy of nivolumab (Opdivo) as salvage treatment after second- or later-line chemotherapy failure.20 In the study, the median OS was 5.32 months with nivolumab versus 4.14 months with placebo. The OS rates at 6 months were 46.4% and 34.7% with nivolumab and placebo, respectively, and 26.6% and 10.9%, respectively, at 12 months. Nivolumab also resulted in an improved overall response rate (11.2% vs 0%) and PFS (1.61 vs 1.45 months) but more grade 3 or higher adverse events (11.5% vs 5.5%) compared with placebo. “[Based on these findings], nivolumab was approved last September and included as a standard thirdor later-line treatment in Japan,” Shitara said, noting that it can be used in Japan regardless of PD-L1 status.
Another checkpoint inhibitor trial showing promise is CheckMate 032, which is examining the safety and efficacy of nivolumab with or without ipilimumab (Yervoy) in Western patients with chemotherapy-refractory esophagogastric cancers.21 The study is assessing 3 different dosing strategies: nivolumab 3 mg/kg, nivolumab 1 mg/kg plus ipilimumab 3 mg/kg, and nivolumab 3 mg/ kg plus ipilimumab 1 mg/kg. After a median follow-up of 28, 24, and 22 months across these 3 groups, 12-month PFS rates were 8%, 17%, and 10%, respectively, and 12-month OS rates were 39%, 35%, and 24%, respectively. Treatment-related grade 3/4 adverse events were highest in the nivolumab 1 mg/kg plus ipilimumab 3 mg/kg arm, affecting 47% of patients versus 17% receiving nivolumab 3 mg/kg and 27% receiving nivolumab 3 mg/ kg and ipilimumab 1 mg/kg.
“In gastric [cancer], similar to melanoma and small cell lung, nivolumab 1 mg/kg plus ipilimumab 3 mg/kg is the preferred regimen. The grade 3/4 toxicity rate is quite high with that regimen, and ipilimumab really drives this toxicity profile. Patients generally develop liver function test abnormalities, diarrhea—toxicities that are reversible with high-dose steroids—but you really have to know what you’re observing and not minimize the adverse effects,” Janjigian said. “That being said, in patients who are fit, combination therapies—and this is the sort of the lesson we learned in gastric cancer—2 drugs (combination therapies) are usually better than 1, particularly for the immune checkpoint inhibitors.”Despite a lack of blockbuster approaches in advanced gastric and GEJ cancers, there is reason for optimism. The panel noted that many new and promising treatments are currently under investigation, including MMP-9, an agent aimed at improving the tumor microenvironment to render chemotherapy more effective; claudiximab (IMAB362), which targets claudin 18.2, a major structural component in gastric cancer cell junctions; next-generation fibroblast growth factor receptor inhibitors; and oncolytic viruses, among others.
“I think the message has to be clear that gastric cancer is treatable, and there is hope, and there are a lot of new options coming out every year, essentially,” Janjigian said. Enzinger expressed a similar sentiment: “We’ve entered a new era where we’re moving beyond cell cycle poisons.” However, the panel also called for improved collaboration among gastric cancer experts globally. “I’m sure that if we can put the brains and the work of everybody together, we can make further progress for our patients,” Van Cutsem said.
Bendell added, “Teamwork does make dreams work.”