CE / CME
Physician Assistants/Physician Associates: 1.50 AAPA Category 1 CME credits
Nurses: 1.50 Nursing contact hours
Physicians: maximum of 1.50 AMA PRA Category 1 Credits™
Pharmacists: 1.50 contact hours (0.15 CEUs)
Released: December 22, 2023
Expiration: December 21, 2024
MATTERHORN: Durvalumab vs Placebo + FLOT for Resectable Gastric or GEJ Cancer
John Marshall, MD:
At ESMO 2023, interim results were presented from the randomized phase III MATTERHORN study evaluating the addition of durvalumab to perioperative fluorouracil plus leucovorin/oxaliplatin/docetaxel (FLOT) in patients with stage II-IV gastric and gastroesophageal junction (GEJ) adenocarcinoma.[Janjigian 2023a]
We already knew that FLOT given before surgery is beneficial. In a phase II/III trial of patients with resectable gastric and GEJ adenocarcinoma, perioperative FLOT improved overall survival (OS) compared with epirubicin/cisplatin plus fluorouracil or capecitabine-based therapy (50 vs 35 months; HR: 0.77; P = .012.) More patients in the FLOT group achieved margin-free resection.[Al-Batran 2019] The MATTERHORN study evaluated whether addition of durvalumab, an anti‒PD-L1 antibody, to FLOT could have additional benefit. In the study, 948 patients with previously untreated stage II, III, and IVA gastric and GEJ adenocarcinoma were randomly assigned to receive preoperative durvalumab or placebo, each with FLOT, followed by surgery and postoperative durvalumab or placebo, each with FLOT. Patients were stratified by geographic region (Asia vs non-Asia), lymph node status, and PD-L1 status (tumor area positivity <1% vs ≥1%).
The primary endpoint was event-free survival (EFS). Secondary endpoints included central review of pathologic complete response (pCR) by the modified Ryan criteria, OS, and safety.
MATTERHORN: Baseline Characteristics
John Marshall, MD:
Baseline characteristics were well balanced between arms. This was a worldwide trial, with only 19% enrollment in Asia.[Janjigian 2023a] Gastric cancer was more common than GEJ cancer, occurring in 67.5% of patients. Most (90%) patients had some PD-L1 expression, and some had high levels of PD-L1. Arms were well balanced based on histology type, with both intestinal and diffuse types represented.
MATTERHORN: Efficacy Outcomes
John Marshall, MD:
In this preliminary analysis, pCR rates by central review were significantly higher with durvalumab plus FLOT vs placebo plus FLOT (19% vs 7%; odds ratio: 3.08; P <.00001).[Janjigian 2023a] There was an absolute difference of 12% between arms. Having a 1 in 5 chance of a pCR is impressive.
Other clinical endpoints including OS are still pending. A strategy in these analyses is to define an earlier endpoint, such as pCR, that might aid in speeding drug approval, as was done for other tumor types (eg, breast cancer).
MATTERHORN: Surgical Outcomes
John Marshall, MD:
Although all patients had planned to undergo surgery, not everyone received surgery for various reasons.[Janjigian 2023a] There were no significant differences in rates of surgery performed or rates of R0 resection between the 2 arms.
MATTERHORN: Safety
John Marshall, MD:
The addition of durvalumab added some toxicity, but not dramatically so; the addition of durvalumab was well tolerated.[Janjigian 2023a] Some toxicity is observed with the FLOT regimen in a population of patients who already are pretty ill. Grade 3/4 adverse events (AEs) possibly related to any study treatment occurred in 58% of patients receiving durvalumab plus FLOT vs 56% of patients receiving placebo plus FLOT; serious AEs occurred in 20% vs 16%. The most common possibly treatment-related grade 3/4 AEs observed with durvalumab plus FLOT vs placebo plus FLOT were neutropenia (20% vs 21%), neutrophil count decrease (19% vs 22%), diarrhea (5% vs 4%), and white blood cell decrease (5% vs 6%). A similar number of patients in each arm (63%) received 4 postoperative doses of FLOT.
MATTERHORN: Clinical Implications
John Marshall, MD:
In this preplanned interim analysis of pCR, the implications are clear. Addition of immunotherapy to FLOT can translate into an improved pCR rate for this population, with an absolute difference of 12% between arms. Whether these findings translate into improved OS and EFS remains to be seen, but my prediction is that they will. It will be interesting to see whether the standard of care changes in the near future.
KEYNOTE-811: Pembrolizumab vs Placebo + Trastuzumab and Chemotherapy in Metastatic HER2-Positive Gastric or GEJ Cancer
John Marshall, MD:
Another study evaluating the addition of immunotherapy to treatment of gastric cancer was the randomized phase III KEYNOTE 811 trial. In this study, pembrolizumab or placebo was added to trastuzumab and chemotherapy in the first-line treatment of advanced HER2-positive gastric or GEJ cancer. The combination of pembrolizumab, trastuzumab, fluoropyrimidine, and platinum-containing chemotherapy received accelerated approval by the FDA in 2021 based on a protocol-specified interim analysis of KEYNOTE-811.[Pembrolizumab PI; Janjigian 2021] The addition of pembrolizumab improved the objective response rate (ORR) by 22.7% (P = .00006.).[Janjigian 2021] At ESMO 2023, OS and PFS data were updated from KEYNOTE 811.[Janjigian 2023b; Janjigian 2023c]
In this update, after a median follow-up of approximately 39 months, treatment with pembrolizumab, trastuzumab, fluoropyrimidine, and platinum-containing chemotherapy was associated with a significant improvement in median PFS vs placebo, trastuzumab, and chemotherapy (10.0 vs 8.1 months; HR: 0.73; 95% CI: 0.61-0.87; P = .0002). Median OS was 20.0 months vs 16.8 months, respectively, but this did not meet the criteria for significance (HR: 0.84; 95% CI: 0.70-1.01). The benefit with pembrolizumab was greater in patients with a PD-L1 combined positive score (CPS) ≥1 for PFS (HR: 0.71; 95% CI: 0.59-0.86) and OS (HR: 0.81; 95% CI: 0.67-0.98).
Treatment-related AEs (TEAEs) were as expected, with immune-mediated AEs more frequent in patients receiving pembrolizumab. Grade ≥3 TEAEs occurred in 59% of patients receiving pembrolizumab vs 51% of patients receiving vs placebo. The most common TEAEs in both arms were diarrhea, nausea, and anemia. Pembrolizumab could be added to this regimen without substantial additional toxicity.
In summary, these findings provide reassurance that pembrolizumab combined with HER2 targeted therapy is the standard of care for HER2 positive gastric cancer. Based on this analysis, the approval for pembrolizumab was amended to restrict the indication in gastric cancer for use in combination with trastuzumab, fluoropyrimidine, and platinum-containing chemotherapy for the first-line treatment of adults with locally advanced unresectable or metastatic HER2-positive gastric or GEJ adenocarcinoma whose tumors express PD-L1 CPS ≥1%.[FDA; Pembrolizumab PI]
CABINET: Cabozantinib vs Placebo for Advanced Neuroendocrine Tumors
John Marshall, MD:
Neuroendocrine tumors (NETs) arise from neuroendocrine cells, and tumors are highly vascularized. Several routes—including the VEGF/VEGFR pathway, FGF-dependent signaling, and the PDGF/PDGFR axis—have been demonstrated to regulate angiogenesis.[Lauricella 2022] These pathways have been areas of focus for drug research in NETs. Cabozantinib is a multikinase inhibitor of VEGFR, RET, c-MET, and AXL. In a phase II study of patients with progressive disease of pancreatic neuroendocrine tumor (pNETs) (n = 20) or extrapancreatic neuroendocrine tumor (epNETs) (n = 41), cabozantinib demonstrated a median PFS of 21.8 and 31.4 months in each cohort, respectively.[Chan 2017; Lauricella 2022]
There had been few advances in the treatment of NETs in recent years until the promising phase II results with cabozantinib. The phase III Alliance CABINET study compared cabozantinib with placebo in patients with advanced NETs who had received ≥1 prior therapy. In the study, patients were enrolled to the pNET cohort (n = 93) or the epNET cohort (n = 197) and randomly assigned to receive cabozantinib 60 mg/day or placebo. The primary endpoint was PFS by blinded independent central review (BICR); secondary endpoints were OS, ORR, safety, and tolerability. Efficacy and safety findings in each cohort were presented.[Chan 2023]
CABINET: Baseline Characteristics in epNET Cohort
John Marshall, MD:
Baseline characteristics were well balanced between arms. The median age of enrolled patients was 66 years (range: 28-86 years).[Chan 2023] The most common primary tumor site in the epNET cohort was gastrointestinal (50% in the cabozantinib arm and 66% in the placebo arm). A primary tumor site could not be determined for 14% to 15% of patients. Concurrent somatostatin analogues were used in 62% of patients receiving cabozantinib and 65% of patients receiving placebo. Patients had received a median of 2 previous therapies; the most common were everolimus, Lu-177 dotatate, and temozolomide with or without capecitabine.
CABINET: Efficacy in epNET Cohort
John Marshall, MD:
After a median follow-up of 13.9 months in the epNET cohort, cabozantinib demonstrated fairly impressive efficacy with a PFS by investigator of 8.3 months vs 3.2 months with placebo (HR: 0.45; 95% CI: 0.30-0.66; P <.0001).[Chan 2023] For the primary endpoint of median PFS by BICR, the HR was 0.50 (95% CI: 0.32-0.79). There were minimal differences in ORR, but there was more stable disease with cabozantinib. There was no significant difference in OS between arms, with a median OS of 21.9 months for cabozantinib vs 22.4 months for placebo (HR: 0.90; 95% CI: 0.56-1.46; P = .34). Regarding the OS data, patients in both arms received subsequent therapies, which may have influenced OS. However, the PFS outcome was very encouraging.
CABINET: Baseline Characteristics in pNET Cohort
John Marshall, MD:
In the pNET cohort, most patients had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 or 1 and grade 1 or 2 disease (86% of patients receiving cabozantinib and 80% of patients receiving placebo).[Chan 2023] Tumors were well differentiated in 90% of patients receiving cabozantinib and 97% of patients receiving placebo. Patients receiving cabozantinib had been treated with a median of 3 prior lines of therapy (range: 1-9), most commonly everolimus (82%), Lu-177 dotatate (55%), and temozolomide with or without capecitabine (68%). Patients receiving placebo had been treated with a median of 2 prior lines of therapy (range, 0-7), most commonly everolimus (77%), Lu-177 dotatate (58%), and temozolomide with or without capecitabine (52%). There were many heavily pretreated patients.
CABINET: Efficacy in pNET Cohort
John Marshall, MD:
In the pNET cohort, the PFS benefit with cabozantinib was even more dramatic, with a median PFS by investigator of 11.4 months vs 3.0 months with placebo (HR: 0.27; 95% CI: 0.14-0.49).[Chan 2023] For the primary endpoint of median PFS by BICR, the HR was 0.25 (95% CI: 0.12-0.54; P <.0001). The ORR was 18% with cabozantinib vs 6% with placebo; the stable disease rate was 58% with cabozantinib vs 39% with placebo. Median follow-up was 16.7 months for the pNET cohort. In terms of OS, the median OS with cabozantinib was 43.5 months vs 31.0 months with placebo (HR: 077; 95% CI: 0.34-1.73). For a patient with a pNET, with cabozantinib we have a drug that might hold a patient for 1 year or more. In summary, there was clear efficacy in both the pNET and epNET cohorts.
CABINET: Adverse Events
John Marshall, MD:
The safety profile is as expected. The most common grade ≥3 AEs in the epNET cohort were hypertension (27.4% with cabozantinib vs 4.8% with placebo), fatigue (13.7% vs 7.9%), and diarrhea (10% vs 3%), and in the pNET cohort the most common grade ≥3 AEs were hypertension (26.7% vs 20.0%), fatigue (13.3% vs 3.3%), hyperglycemia (8.3% vs 10.0%), thromboembolic events (11.7% vs 0%), and hand‒foot syndrome (10.0% vs 0%).[Chan 2023] These agents need to be dosed wisely, and the dose must be managed with modifications if needed. These are chronic therapies that people will receive for some time, so it is important to find the optimal dose and schedule for patients once this treatment has been started.
CABINET: Clinical Implications
John Marshall, MD:
In this phase III analysis of cabozantinib for previously treated NETs, improved PFS was observed vs placebo in both epNETs and pNETs. No new safety signals were noted. The clinical implications of these results are clear. Cabozantinib will be a new therapy that we will be recommending, and it will almost certainly receive guideline approval, if not more formal approvals for its use in these patients with ≥1 prior treatment.
NICHE-3: Neoadjuvant Nivolumab and Relatlimab in dMMR Colon Cancer
John Marshall, MD:
Many of us are familiar with the recent findings observed with the anti–PD-1 antibody dostarlimab for patients with mismatch repair deficient (dMMR) rectal cancer. A phase II study of neoadjuvant dostarlimab in 12 patients with dMMR stage II/III rectal cancer demonstrated a 100% clinical complete response (CR) rate after ≥6 months of follow-up without the need for surgery.[Cercek 2022] For patients with nonmetastatic dMMR colon cancer (n = 112), neoadjuvant nivolumab plus ipilimumab demonstrated a major pathologic response rate of 95% and a pCR rate of 67% in the NICHE-2 study.[Chalabi 2022]
The multicenter phase II NICHE 3 trial was undertaken to evaluate neoadjuvant nivolumab plus the LAG-3 inhibitor relatlimab in patients with resectable dMMR colon cancer.[Verschoor 2023] Part 1 of the trial enrolled 19 patients with previously untreated resectable locally advanced dMMR colon adenocarcinoma. Patients received nivolumab plus relatlimab at 8 weeks and 4 weeks prior to surgery. The primary endpoint was pathologic response rate.
NICHE-3: Baseline Characteristics
John Marshall, MD:
The median age of enrolled patients was 56 years—a younger population, as would be expected. Patients tended to have a good PS (ECOG PS of 0 in 74%) and primarily had T3 (58%) or T4 (37%) disease.[Verschoor 2023] The primary tumor location was the right colon in most patients (84%), again as would be expected. Documented germline mutations, known as Lynch syndrome, were presented in 5 of 19 patients (26%).
NICHE-3: Safety
John Marshall, MD:
The safety profile was as expected.[Verschoor 2023] Grade 1-2 immune-related AEs occurred in 74% of patients, with grade 3 AEs in 5%. Four patients (21%) developed endocrinopathies that required supplementation.
NICHE-3: Pathologic Response, Surgical Outcomes, and Clinical Implications
John Marshall, MD:
The pathologic response rate was 100% and included major responses in 89% of patients, CRs in 79%, and partial responses in 11%.[Verschoor 2023] This differs from the 100% CR rate observed in patients with rectal cancer.[Cercek 2022] However, the patients with rectal cancer did not proceed to surgery. In this study, all patients underwent surgery, allowing confirmation of whether there was residual disease.
For patients with colon cancer, I think surgery will remain the standard approach for eligible patients for now. The role of neoadjuvant therapy remains a question, but it certainly appears to be an option in this dMMR group.
Prognostic Value of KRAS/BRAFV600E Mutations in MSS and MSI-H Stage III Colon Cancer
John Marshall, MD:
Several studies presented at ESMO 2023 provided valuable information regarding the characterization of tumor mutations in prognosis for individual patients.
One study was a database analysis assessing the prognostic value of KRAS and BRAF mutations in patients with microsatellite stable (MSS) (n = 7492) and microsatellite instability‒high (MSI-H) (n = 968) stage III colon cancer. The patient-level data came from the IDEA and ACCENT consortia database, which contains pooled clinical trial data from 7 randomized clinical trials of adjuvant chemotherapy for patients with stage III colon cancer.[Taieb 2023a; Taieb 2023b]
Outcomes appeared better in patients with MSI-H disease than in patients with MSS disease. For patients with MSI-H disease, presence of BRAF and RAS mutations had no significant effect on disease-free survival (DFS) or OS; 8-year OS rates were higher in the double wild-type population (76.9%) and patients with KRAS mutations (74.6%) than in patients with BRAF mutations (71.6%).
By contrast, for patients with MSS disease, KRAS/BRAF mutations were significantly associated with differences in both DFS (P <.0001) and OS (P <.0001). Prognosis was most favorable in the double wild-type population, followed by patients with KRAS mutations, followed by patients with BRAF mutations, with 8-year OS rates of 73.6%, 67.2%, and 58.9%, respectively.
Anecdotally, I also have observed that immune therapy does not appear to be as effective for colon cancer in patients with MSI-H disease with a BRAF V600E mutation. We tend to use BRAF targeted agents in those patient populations. I predict that in the future, we will measure BRAF and RAS in the adjuvant setting. We hope there will soon be data with BRAF inhibitors or even RAS inhibitors in the adjuvant setting.
PEGASUS: ctDNA Testing‒Guided Adjuvant Therapy in Colon Cancer
John Marshall, MD:
Another notable abstract from ESMO 2023 assessed the use of postsurgical liquid biopsy‒guided treatment for patients with stage III and high-risk stage II colon cancer, with the goal of tailoring adjuvant therapy.[Lonardi 2023] In the multicenter, prospective PEGASUS study, circulating tumor DNA (ctDNA) analysis was used in 135 patients with stage III and high-risk stage II colon cancer. Testing occurred at 2 time points—after surgery and after adjuvant therapy—to assess for minimal residual disease and guide adjuvant therapy decisions.
Relapses were more common in patients testing ctDNA positive (12 of 35 patients; 34%) vs patients testing ctDNA negative (10 of 100 patients; 10%). The most common site of relapse among patients testing ctDNA negative was the lung. Of note, relapses in the lung and peritoneum occurred only in patients testing ctDNA negative. This may have implications for selecting patients suitable for ctDNA analysis.
Overall, these key studies presented at ESMO 2023 highlighted exciting possibilities in the management of gastrointestinal malignancies. The future is very encouraging.
Back
