Randomised comparison of fluorouracil, epidoxorubicin and methotrexate (FEMTX) plus supportive care with supportive care alone in patients with non-resectable gastric cancer: S Pyrhönen, T Kuitunen, P Nyandoto & M Kouri

Abstract

We discuss Pyrhönen’s paper as a catalyst for defining systemic therapy approaches in gastric cancer and how the field has evolved in the last two decades in sequential, targeted and more recently immune-directed therapies and how it may develop to transform survival in this cancer of high unmet need.

https://www.nature.com/articles/bjc1995114.

Pyrhönen et al.’s seminal paper [1] helped lay the foundations for modern-day combination systemic therapy in gastric/gastroesophageal junction (GEJ) cancer. It was one of three pivotal European randomised studies in the 1990s demonstrating a survival benefit for multi-drug chemotherapy with anti-metabolites including 5-fluorouracil (5FU), plus anthracycline or a topoisomerase inhibitor over best supportive care (survival 6–8 versus 3 months). The UK REAL1 study demonstrated similar efficacy for mitomycin C (alkylating agent) versus epirubicin (anthracycline) when added to cisplatin and infusional 5FU but improved quality of life (QoL) favouring anthracycline [2]. REAL2 was one of two studies to establish oxaliplatin and oral capecitabine as alternatives to cisplatin and infusional 5FU respectively (with epirubicin in REAL2) with comparable efficacy and certain toxicity/ease of administration benefits [3]. In early-stage gastric cancer, the newly developed ECX regimen given peri-operatively significantly improved survival in the pivotal MAGIC study, later superseded by peri-operative FLOT (taxane-inclusive) [4, 5]. Numerous advanced disease studies globally evaluated other doublet or triplet cytotoxic combinations including taxanes and anthracyclines, often on a fluoropyrimidine/platinum (FP) backbone. However, meta-analyses demonstrated a marginal survival benefit for treatment intensification. Median overall survival (mOS) was 9–11 months and no regimen surpassed the 12-month survival barrier in gastric cancer (excluding Asian gastric cancer studies). Based on this, toxicity profiles, global generalisability and combination potential with novel drugs, first-line FP is considered a global chemotherapy standard (CF, CX, FOLFOX, CAPOX).

In REAL2 only 20% of patients received second-line chemotherapy for gastric cancer, often as a platinum re-challenge. Three randomised studies established an evidence base for taxane or irinotecan following first-line FP-based chemotherapy improving mOS from around 3 to 5 months and the COUGAR study (docetaxel) showed a QoL advantage [6]. Nowadays, approximately 50% of patients can receive second-line chemotherapy which, for the fittest, is with paclitaxel plus the anti-VEGFR2 antibody ramucirumab based on the RAINBOW study associated with mOS of approximately 9 months (7 months if restricted to western patients) [7]. Surprisingly, VEGF(R) inhibition appears ineffective in first-line therapy and with conflicting data in later lines. Approximately 20% of patients receive third-line chemotherapy where the highest level of evidence is for trifluridine-tipiracil associated with a mOS of approximately 6 months [8]. Hence, for some patients, a sequential treatment pathway akin to colorectal cancer has been forged with survival gains.

The greatest step-change required for survival impact in advanced gastric cancer is in first-line treatment, challenging the limitations of chemotherapy and that 50% of patients only have one line of therapy. Hence, the next decade focussed on targeted therapy trials in advanced gastric cancer. The pivotal TOGA study validated HER2 targeting with trastuzumab combined with FP for the 20% of patients with HER2 overexpression by immunohistochemistry (3+ or 2+/FISH positive) with mOS improving from approximately 11 to 13 months, higher in the HER2/FISH positive and HER2 3+ sub-groups [9]. Sadly, additional HER2 targeting with pertuzumab/trastuzumab/FP in the first-line and with TDM1 in the second line was unsuccessful. However, interest in this target has been re-invigorated with the advent of the antibody-drug conjugate trastuzumab-deruxtecan (T-DXd) showing response rates of approximately 50% in heavily pre-treated patients (versus 15% for chemotherapy) [10] garnering 2021 FDA approval and with pivotal studies ongoing including the second-line DESTINY Gastric-04 (T-DXd versus paclitaxel/ramucirumab) and first-line new combinations (DESTINY Gastric-03).

Conversely, 2010–2020 witnessed several negative phase III targeted-drug studies including EGFR, MET, FGFR, PARP, STAT inhibitors. Some were biomarker unselected which likely hampered success whereas some failed despite identifying putative tissue-based predictive biomarkers. Notably, a recent FGFR-targeted phase II study has indicated a signal leading to phase III evaluation and targeting the gap junction protein claudin 18.2 (~ 40% of gastric cancer) led to survival gains in the recently reported Spotlight study. Why did the earlier studies fail and does this invalidate precision medicine in gastric cancer? Gastric cancer heterogeneity remains a major challenge, both spatial (intra- and inter-tumoral) and temporal, some recent studies suggesting up to a 30% biomarker discordance between primary and metastatic sites. Gene level events (i.e., amplifications) may increase the predictive potential for some targeted therapies and sequencing (NGS) of plasma circulating tumour DNA (ctDNA) may circumvent heterogeneity (i.e., reflecting clonal drivers) and temporal (i.e., resistance mechanisms) issues for some patients. However, this equates to small patient sub-groups (1–5%), a reality for precision medicine. Next-generation studies such as the Japanese GOZILA study facilitate screening through plasma ctDNA, matching druggable targets with best-in-class drugs with efficiencies in time to result, speed of enrolment and likelihood of response for liquid versus tissue-based NGS [11]. For the future, there is not a “one-size-fits-all” approach to precision testing; the most appropriate methodology (tissue/liquid/both) should be deployed depending on the target, the drug and the patient.

The most recent frontier in gastric cancer has been immunotherapy. The first-line 1-year mOS barrier in HER2 negative disease has finally been surpassed through adding the PD1-targeted immune checkpoint inhibitor (ICI) nivolumab to standard first-line FP chemotherapy (checkmate 649 study) leading to FDA and EMA approvals in 2021; in patients who are PDL1 CPS 5 or greater, the mOS was approximately 14 months [12]. This finding has been replicated with other ICIs added to FP (tislelizumab, sintilimab), in contrast to one negative study in gastric cancer (pembrolizumab/FP). Challenges include harmonisation of PDL1 testing, PDL1 threshold selection levels, additional predictive biomarkers, developing other rational first-line combination immunotherapies and intrinsic/acquired resistance to ICIs. In tandem, cellular therapies are being developed with an early signal for claudin 18.2-directed approaches, noting these treatments can be complex to manufacture/administer and cause serious toxicities (cytokine release syndrome). Although immunotherapy is evidence-based in the third-line setting and beyond, first-line developments supersede these. For MSI-H/mismatch repair deficient gastric cancer (4%), pembrolizumab alone is transformative (2017 FDA approval; tumour agnostic) and increasingly deployed in first-line given good response rates, durability and limited toxicities. For HER2-positive gastric cancer final results are awaited for the Keynote 811 study, with potential for a new first-line standard with FP/trastuzumab/pembrolizumab given highly promising interim results [13].

Since Pyrhönen’s paper, we have witnessed some major shifts in our understanding and treatment of gastric/GEJ adenocarcinoma, accelerating in the last 2–3 years. Through advanced approaches encompassing digital pathology, novel diagnostics, integrative data science and application to earlier stage disease, the hope is further rapid transformation of outlook in this challenging cancer.

Author contributions

NS and DC co-wrote this article.

Competing interests

Within the last 5 years, COI for both authors is detailed: NS and DC are supported by the NIHR Biomedical Research Centre at The Royal Marsden and Institute for Cancer Research. NS is supported by the following grants: NIHR NIHR128529 and SBRI SBRIC01P3008. DC is supported by NIHR128529. NS: Honoraria: Merck Serono, Novartis, MSD Oncology, Eli Lilly, Pierre Fabre, Amgen, GSK, Servier, Pierre Fabre. Advisory role; Pfizer, Servier, AstraZeneca, MSD Oncology Novartis, Bayer, GSK, Guardant Health, Gilead, Seagen; Research support: Pfizer, AstraZeneca, Guardant Health Travel/accommodation support: Roche, BMS, MSD DC: Research funding: MedImmune/AZ, Clovis, Eli Lilly, 4SC, Bayer, Celgene, Leap, Advisory: OVIBIO.