Welcome to the CCO Site

Thank you for your interest in CCO content. As a guest, please complete the following information fields. These data help ensure our continued delivery of impactful education. 

Become a member (or login)? Member benefits include accreditation certificates, downloadable slides, and decision support tools.


CCO Independent Conference Highlights of the Virtual 2021 Gastrointestinal (GI) Cancers Symposium

Axel Grothey Headshot
Axel Grothey, MD
Manish A. Shah, MD
Released: March 10, 2021

Gastric/Gastroesophageal Cancers

FIGHT: First-line Bemarituzumab Plus mFOLFOX6 vs Placebo Plus mFOLFOX6 in Advanced Gastric/GEJ Cancer

Manish A. Shah, MD:
The FIGHT trial is a global, double-blind, placebo-controlled, randomized phase II trial evaluating the first-in-class humanized IgG1 anti-FGFR2b monoclonal antibody bemarituzumab plus mFOLFOX6 vs mFOLFOX6 alone in patients with gastric/gastroesophageal junction (GEJ) cancer (N = 155).[21,22] All patients had FGFR2b overexpression by IHC and/or FGFR2 gene amplification by ctDNA. This was a large study for this rare patient population—the frequency of this amplification is somewhere between 5% and 10% of patients, depending on geographic region. Enrolled patients had not received previous therapy and had unresectable locally advanced or metastatic gastric/GEJ adenocarcinoma. Patients had an ECOG PS of 0 or 1 and were not HER2 positive. Stratification was based on geographic region, whether the patient had received a single dose of mFOLFOX6 during screening and whether they had previous adjuvant or neoadjuvant therapy.

Patients in the bemarituzumab group received 15 mg/kg of bemarituzumab every 2 weeks, plus a single dose of 7.5 mg/kg on Day 8 of cycle 1 plus mFOLFOX6 (n = 77). Patients in the placebo group were on the same dosing schedule (n = 78). Treatment continued until progressive disease, unacceptable toxicity, or death. The primary endpoint was investigator-assessed PFS, and key secondary endpoints were OS and ORR. The results of FIGHT were presented at ASCO GI 2021 by Wainberg and colleagues.[21]

FIGHT: Patient Characteristics

Manish A. Shah, MD:
FIGHT had well-balanced arms. As expected, more than two thirds of patients were male, and a little more than half were Asian. About 50% of patients received a single dose of mFOLFOX6 during screening, and about 80% had measurable disease at baseline. This was a unique study that allowed any measure of overexpression or amplification to be eligible. About 95% of patients had IHC reflecting their FGR2b status, and an additional 15% had FGFR2 gene amplification through ctDNA.


Manish A. Shah, MD:
Median PFS was 9.5 months for patients who received bemarituzumab plus mFOLFOX6 and 7.4 months for patients who received mFOLFOX6 alone (HR: 0.68; 95% CI: 0.44-1.04; P = .0727). The 9-month PFS rate was 52.5% in patients who received combination treatment compared with 33.8% in patients who received mFOLFOX6 monotherapy.

Median OS had not been reached for the combination group and was 12.9 months for the group receiving mFOLFOX6 alone (HR: 0.58; 95% CI: 0.35-0.95; P = .0268). The 12-month OS rate was 65.3% in the combination arm and 56.9% in the mFOLFOX6 monotherapy arm. These data are encouraging for phase III of the study.[23]

FIGHT: PFS Based on FGFR2b Overexpression

Manish A. Shah, MD:
Patients with FGFR2b overexpression of ≥ 5% (IHC 2+/3+; n = 118) had a 10.2-month median PFS with bemarituzumab (n = 58) vs 7.3 months with placebo (n = 60; HR: 0.54; 95% CI: 0.33-0.87). The 9-month PFS rate was 56.3% and 28.6%, respectively. Patients with FGFR2b overexpression of ≥ 10% (n = 96) had a median PFS of 14.1 months with bemarituzumab (n = 44) vs 7.3 months with placebo (n = 52; HR: 0.44; 95% CI: 0.25-0.77). The 9-month PFS rate was 57.0% for bemarituzumab and 26.4% for placebo.

Overall, bemarituzumab provided a significant survival benefit, particularly in patients with greater FGFR2b overexpression. Of interest, PFS in the placebo arm was reduced in patients with higher FGFR2b overexpression, suggesting that the degree of overexpression is an adverse prognostic feature for gastric/GEJ tumors and an important target in this patient population.

FIGHT: OS Based on FGFR2b Overexpression

Manish A. Shah, MD:
The data for OS were similar to PFS in that patients with greater FGFR2b overexpression had increased 12 month OS rates when receiving bemarituzumab, but there was an inverse correlation for patients receiving placebo. In the intent-to-treat group, median OS was not reached for the bemarituzumab group and was 12.9 months for the placebo group (HR: 0.58; 95% CI: 0.35-0.95; P = .0268), with a 12-month OS rate of 65.3% and 56.9%, respectively. In patients with FGFR2b overexpression of ≥ 5%, median OS was not reached for the bemarituzumab group and was 12.5 months for the placebo group (HR: 0.52; 95% CI: 0.30-0.91), with a 12-month OS rate of 67.9% and 55.5%, respectively. In patients with FGFR2b overexpression of ≥ 10%, median OS was not reached for the bemarituzumab group and was 11.1 months for the placebo group (HR: 0.41; 95% CI: 0.22-0.79), with a 12-month OS rate was 70.2% and 49.5%, respectively.


Manish A. Shah, MD:
The ORR was also better with bemarituzumab vs placebo (47% vs 33%). The median time to response was about 1.8 months in both groups, but median duration of response (DoR) was slightly higher in the bemarituzumab arm vs placebo (12.2 months vs 7.1 months).

FIGHT: Select Treatment-Emergent AEs

Manish A. Shah, MD:
Toxicity was similar in the groups, except for the frequency of stomatitis and dry eye, which are more common with FGFR2 inhibitors. Stomatitis of any grade was present in 31.6% of patients receiving bemarituzumab and 13.0% of patients receiving placebo, and dry eye affected 26.3% and 6.5%, respectively. Stomatitis of grade ≥ 3 occurred in 9.2% of patients receiving bemarituzumab and 1.3% of patients receiving placebo. Dry eye syndrome of grade ≥ 3 occurred in 2.6% of patients receiving bemarituzumab and 0% of patients receiving placebo. Liver function test abnormalities slightly increased but were mostly grade 1/2.

FIGHT: Corneal-Related AEs

Manish A. Shah, MD:
Eye-related AEs were strongly linked with bemarituzumab therapy. Corneal related AEs occurred in 67.1% of patients receiving bemarituzumab vs 10.4% of patients receiving placebo, with a median onset of 16.1 weeks. Most patients continued bemarituzumab, but 20 patients (26.3%) discontinued therapy. Of those, 12 patients (60%) resolved by cutoff, with a median time to resolution of 27 weeks.

FIGHT: Clinical Implications

Manish A. Shah, MD:
First-line treatment with bemarituzumab—a first-in-class humanized IgG1 monoclonal antibody targeting FGFR2b—plus mFOLFOX6 showed statistically significant PFS, ORR, and OS in patients with gastric/GEJ cancers with FGFR2b overexpression or FGFR2 gene amplification vs mFOLFOX6 plus placebo. We need to wait for the phase III FIGHT data, but these are encouraging results that suggest bemarituzumab is advantageous in patients overexpressing FGFR2. It is also encouraging that there was a dose effect with FGFR2 overexpression and a benefit from bemarituzumab. This therapy was also tolerable, which bodes well for its potential continued use, keeping in mind the eye-related issues with this class of agents.

Axel Grothey, MD:
I like the data, and I think they have the potential to be a game changer for these patients if they hold up in phase III. I also think FGFR2 expression level could be the next biomarker beyond HER2, which could lead to highly targeted therapies for upper GI malignancies.

Manish A. Shah, MD:
Yes, maybe the next biomarker beyond HER2 and PD-L1.

Axel Grothey, MD:
PD-L1 is an interesting biomarker. When you look at ATTRACTION-2,[24,25] nivolumab worked in almost every patient. Still, combined positive score (CPS) scoring is interesting.

Manish A. Shah, MD:
I agree and only mention it because it may be practice changing. The CheckMate 649 trial was practice changing based on CPS.[26,27]

Axel Grothey, MD:
And MSI-H, too.

Manish A. Shah, MD:
Exactly. FGFR2 is another biomarker likely to change practice and is also something that highlights the importance of collecting tissue from gastric/GEJ tumors to determine microsatellite instability status, TMB, HER2 mutation, PD-L1 CPS, and FGFR2 expression. The increasing focus on studying biomarkers to personalize therapy for patients with gastrointestinal malignancies was also evident from a number of posters at this conference describing ongoing trials.

In HER2-positive patients, these trials include the phase II/III MOUNTAINEER-02 trial with tucatinib plus ramucirumab and paclitaxel ± trastuzumab in previously treated HER2-positive gastric/GEJ cancers, with primary endpoints of safety/toxicity in phase II and OS and PFS in phase III,[28] and the phase II MOUNTAINEER trial in previously treated (with no anti-HER2 agent) HER2-positive mCRC with open-label tucatinib ± trastuzumab in 3 cohorts with a primary endpoint of ORR,[29] as well as trastuzumab deruxtecan (T-DXd) in HER2-positive gastric cancer in the DESTINY-Gastric03 trial, with T-DXd in multiple combinations in the dose escalation phase Ib (patients who progressed on or after prior trastuzumab regimens, with primary endpoints of safety and recommended phase 2 dose), and T-DXd combinations vs T-DXd monotherapy or a non-T-DXd combination in the dose expansion phase II (patients who were treatment-naive for metastatic disease, with primary endpoint of ORR) .[30] Researchers are also investigating patritumab deruxtecan in an open-label phase II study of patients with previously treated advanced/metastatic HER3-high or HER3-low/negative CRC, with a primary endpoint of ORR.[31] Additional biomarkers are also being explored, such as claudin 18.2 for zolbetuximab.[32,33]

LEAP-005 Gastric Cohort: Pembrolizumab Plus Lenvatinib in Previously Treated Advanced Gastric Cancer

Manish A. Shah, MD:
In upper GI cancers, checkpoint inhibition has modest activity on its own, with response rates of 10% to 15% in patients in the third‑line setting whose tumors have a PD-L1 CPS ≥ 1. In patients with a PD-L1 CPS ≥ 5, activity is better. For example, CheckMate 649 showed significant activity for nivolumab combined with chemotherapy in this patient population.

LEAP‑005 is an ongoing multicenter, open-label, multicohort phase II trial studying the combination of lenvatinib and pembrolizumab in adults with advanced and previously treated cancers (planned N = 760).[34] Lenvatinib is an oral tyrosine kinase inhibitor targeting angiogenesis that can overcome the tumor microenvironment’s resistance to checkpoint blockade.

An interim analysis of data from the gastric cancer cohort (n = 31) was presented at ASCO GI 2021 by Chung and colleagues.[35] Patients in the gastric cohort were previously treated with ≥ 2 lines of therapy and had an ECOG PS of 0/1, measurable disease per RECIST v1.1, and tissue available for central PD-L1 assessment. Patients received 200 mg of IV pembrolizumab every 3 weeks and 20 mg of lenvatinib daily for up to 35 cycles. Primary endpoints are ORR and safety/tolerability. Key secondary endpoints are disease control rate, DoR, PFS, and OS. The gastric cohort will be expanded to enroll 100 patients based on observed efficacy in the interim analysis.

LEAP-005 Gastric Cohort: Baseline Patient Characteristics and Disposition

Manish A. Shah, MD:
Patients were a median of 62 years of age and generally fit. Of the 31 patients, 20 (65%) had a baseline ECOG PS of 1, and 22 (71%) had PD-L1 CPS ≥ 1. All but 1 patient had received ≥ 2 previous lines of therapy. Only 7 patients (23%) are continuing treatment; 16 (52%) discontinued due to progressive disease, 7 (23%) due to AEs, and 1 (3%) at physician decision.

LEAP-005 Gastric Cohort: ORR by BICR (Coprimary Endpoint)

Manish A. Shah, MD:
At a median cutoff of 7 months, the ORR was 10% (95% CI: 2-26), and the disease control rate was 48% (95% CI: 30-67). One patient had CR and 2 had PR—results similar to what is observed with pembrolizumab alone. The patient achieving CR had a PD-L1 CPS of 0 (PD-L1 negative), and both patients achieving PR had a PD-L1 CPS ≥ 1. The median DoR has not yet been reached.

LEAP-005 Gastric Cohort: Safety (Coprimary Endpoint)

Manish A. Shah, MD:
Overall, the combination of pembrolizumab and lenvatinib was well tolerated. Only 12 patients (39%) experienced grade 3 toxicities, with 2 (6%) leading to treatment discontinuation. Immune‑related AEs were minimal, affecting 8 patients (26%), but 1 was grade 3. Most immune-mediated AEs related to hypo- or hyperthyroidism. Clinically significant AEs in ≥ 5% of patients were hepatotoxicity (14 patients; 45%), hemorrhage (6 patients; 19%), and hypertension (6 patients; 19%), but these seemed related to lenvatinib and not pembrolizumab. No grade 4 AEs were observed, but there was 1 death from a lenvatinib-associated AE.

LEAP-005 Gastric Cohort: PFS and OS

Manish A. Shah, MD:
In the phase II setting, it is hard to interpret PFS and OS rates, but reported median PFS was 2.5 months (95% CI: 1.8-4.2), with a 22% 6-month rate, and median OS was 5.9 months (95% CI: 2.6-8.7), with a 46% 6-month rate.

LEAP-005 Gastric Cohort: Clinical Implications

Manish A. Shah, MD:
Third-line treatment with pembrolizumab plus lenvatinib looks promising in LEAP-005. However, a previous report shows higher response rates with this drug combination,[36] so I expected the activity to be higher. LEAP-005 had 10 patients with PD-L1–negative tumors (CPS of 0), which are known to have an inferior response to pembrolizumab[37,38] and may have impacted the response rates. We need randomized studies to see if the combination holds weight, but it is encouraging that it is tolerable.

Axel Grothey, MD:
I like this combination, by the way. I think it has potential.

Manish A. Shah, MD:
The activity was less than I would have expected.

Axel Grothey, MD:
Yes, it was a little lower than what has been shown at ESMO, with an ORR of 32% for third-line gastric cancer, [39] and even for other cancers such as previously treated mCRC, with an ORR of 22% (95% CI, 9-40).[40] LEAP-005 is still ongoing with more cohorts and patients, so we will see more data in the future.

TAGS: Trifluridine/Tipiracil vs Placebo in Previously Treated Metastatic Gastric/GEJ Adenocarcinoma

Manish A. Shah, MD:
We have already discussed trifluridine/tipiracil (FTD/TPI) in CRC, but the combination is also being explored in metastatic gastric/GEJ adenocarcinoma. TAGS was an international, double blind, randomized phase III study evaluating FTD/TPI plus best standard of care vs placebo plus best standard of care in adult patients with previously treated, unresectable, metastatic gastric/GEJ adenocarcinoma in the third-line setting (N = 507).[41-43] All patients had an ECOG PS of 0 or 1 and were stratified based on region (Japan vs rest of the world), ECOG PS, and whether they had received ramucirumab previously. Randomization was 2:1 for FTD/TPI plus best standard of care (n = 335) or placebo plus best standard of care (n = 168). FTD/TPI was received at 35 mg/m2 twice daily on Days 1-5 and Days 8-12 of 3-week cycles. Placebo was given on the same schedule. Patients continued until disease progression, intolerable toxicity, or patient withdrawal. The primary endpoint was OS, and key secondary endpoints were PFS, ORR, health-related QoL, and safety.

Although ATTRACTION‑2 was the first randomized phase III clinical trial for third-line gastric/GEJ cancer,[24,25] TAGS was the first to study a non–checkpoint inhibitor in the third-line setting. The primary and secondary endpoint data from TAGS have previously been published, so the data reported at ASCO GI 2021 by multiple groups were updates based on study subanalyses.

TAGS: OS by Previous Therapy With or Without Ramucirumab

Manish A. Shah, MD:
Shitara and colleagues[44] evaluated OS by previous therapy, particularly looking at the influence of ramucirumab on FTD/TPI efficacy. Across subgroups, the median OS for patients who received FTD/TPI was 5.7 months vs 3.6 months for placebo (HR: 0.69; 95% CI: 0.56-0.85). Patients who had previously received ramucirumab and went on to receive FTD/TPI had a median OS of 5.0 months vs 3.8 months for the 55 patients randomized to receive placebo (HR: 0.76; 95% CI: 0.53-1.09). Patients who did not receive ramucirumab before FTD/TPI had a median OS of 6.0 months vs 3.3 months for placebo (HR: 0.66; 95% CI: 0.51-0.85). There were no major differences based on previous therapy with irinotecan, paclitaxel alone, or paclitaxel plus ramucirumab.

TAGS: PFS by Previous Therapy With or Without Ramucirumab

Manish A. Shah, MD:
Similar to the OS data, a small but clear PFS benefit was seen across all previous treatment subgroups with FTD/TPI. The median PFS was 2.0 months for patients receiving FTD/TPI vs 1.8 months for patients receiving placebo (HR: 0.57; 95% CI: 0.47-0.70). In patients who previously received ramucirumab, median PFS was 1.9 months with FTD/TPI and 1.7 months with placebo (HR: 0.54; 95% CI: 0.38-0.77). In patients who had not received ramucirumab, median PFS was 2.2 months with FTD/TPI and 1.8 months with placebo (HR: 0.58; 95% CI: 0.46-0.75).

TAGS: Median OS and PFS in ITT Population by Line of Therapy

Manish A. Shah, MD:
At ASCO GI 2021, Tabernero and colleagues[42] reported significant OS and PFS benefit from FTD/TPI with all lines of therapy. The median OS for third-line therapy with FTD/TPI was 6.8 months vs 3.2 months with placebo (HR: 0.67; 95% CI: 0.47-0.97; P = .0318). In patients receiving fourth-line therapy or above, median OS was 5.2 months with FTD/TPI and 3.7 months with placebo (HR: 0.72; 95% CI: 0.55-0.95; P = .0192).

Regarding PFS, patients receiving third-line therapy had a median PFS of 3.1 months with FTD/TPI vs 1.9 months with placebo (HR: 0.54; 95% CI: 0.38-0.77; P = .0004). Fourth-line therapy or above had a median PFS of 1.9 months vs 1.8 months, respectively (HR: 0.57; 95% CI: 0.44-0.74; P < .0001). Although patients receiving fourth-line therapy and above had somewhat diminished median OS and PFS vs patients in the third-line setting, the response in fourth-line therapy is still significant.

TAGS: Median Time to Deterioration in ITT Population to ECOG PS ≥ 2 by Line of Therapy

Manish A. Shah, MD:
Improvement in the median time to deterioration—defined as an increase to an ECOG PS of 2 or above—was also observed for FTD/TPI in all lines of therapy. In the overall population, the median time to deterioration was 4.3 months for patients receiving FTD/TPI vs 2.3 months for patients receiving placebo (HR: 0.69; 95% CI: 0.56-0.85; P = .0005). Patients receiving third-line therapy had a median time to deterioration of 4.8 months with FTD/TPI vs 2.0 months with placebo (HR: 0.60; 95% CI: 0.42-0.86; P = .0049). Median times for fourth-line therapy or above were 4.0 months and 2.5 months, respectively (HR: 0.75; 95% CI: 0.57-0.98; P = .0329).

The frequency of AEs was similar for both groups. FTD/TPI is clearly an active drug across lines of therapy but is more active as a third-line treatment. It is also active regardless of previous therapies. This highlights that patients benefit longer when they receive additional treatments over the course of their disease.

TAGS: OS by Body Weight Loss

Manish A. Shah, MD:
It was unclear whether FTD/TPI was beneficial in patients who were losing weight, something common in gastric/GEJ cancers, so Ghidini and colleagues[45] evaluated OS by patient body weight loss. Body weight loss was categorized as < 3% from baseline or ≥ 3% from baseline. In the pooled patient population, median OS was 6.3 months for patients with < 3% weight loss and 3.8 months for patients with ≥ 3% weight loss (HR: 0.58; 95% CI: 0.46-0.73). Within the FTD/TPI arm, median OS was 6.5 months for patients with < 3% weight loss and 4.9 months for patients with ≥ 3% weight loss (HR: 0.75; 95% CI: 0.55-1.02) vs median OS of 6.0 months and 2.5 months, respectively, in the placebo arm (HR: 0.32; 95% CI: 0.21-0.49).

TAGS: Clinical Implications

Manish A. Shah, MD:
TAGS revealed that FTP/TPI improved survival in patients with metastatic gastric/GEJ cancers with ≥ 2 previous lines of therapy regardless of previous treatment. However, superior survival was observed during third-line therapy compared with fourth-line therapy and above. Early body weight loss was associated with unfavorable survival outcomes regardless of treatment, but it is important to note that patients showed greater benefit from FTD/TPI treatment vs placebo even with significant weight loss, provided patients with weight loss have adequate ECOG PS. I think these results are important and practice‑changing. People are enamored with checkpoint inhibition for GI cancers, but the response rates are very low. Checkpoint inhibition can be transformative, but for most other patients, I think FTD/TPI should be considered.

Axel Grothey, MD:
I like what you said about having all these lines of therapy available. I remember when ramucirumab was tested against best supportive care in the second‑line setting. Now we are looking at third- and fourth‑line treatment options, and the more options we have, the better. I believe FTP/TPI rightfully found its way to FDA approval based on the data that it works

Provided by Clinical Care Options, LLC

Contact Clinical Care Options

For customer support please email: customersupport@cealliance.com

Mailing Address
Clinical Care Options, LLC
12001 Sunrise Valley Drive
Suite 300
Reston, VA 20191

Supported by educational grants from
Daiichi Sankyo, Inc.
Merck Sharp & Dohme Corp.
Seagen Inc.
Taiho Oncology Inc.

Leaving the CCO site

You are now leaving the CCO site. The new destination site may have different terms of use and privacy policy.


Cookie Settings