CAR-T, Bispecifics, and Beyond: Dr Ku on the Evolving Paradigm in Gastric/GEJ Cancers

In a recent interview with Targeted Oncology, Geoffrey Y. Ku, MD, discussed CAR T-cell therapy, bispecific antibodies, perioperative advances, and other treatment strategies with the potential to transform treatment for patients with gastric and gastroesophageal junction (GEJ) cancers. Ku is a gastrointestinal medical oncologist and cellular therapist at MSK.

Targeted Oncology: Claudin18.2 (CLDN18.2)-directed CAR T-cell therapy is showing great promise in GI cancers. What is the current state of CAR T-cell therapy in GI cancers and what are the next steps?

Geoffrey Y. Ku, MD: There is a Chinese company called Carsgen that has developed a CLDN18.2-directed CAR T-cell therapy called satricabtagene autoleucel (satri-cel). At the 2025 ASCO Annual Meeting, they presented data from a study (NCT04581473) that randomized Chinese patients with advanced GI or GEJ cancer to satri-cel versus third-line choice of chemotherapy.¹ The study achieved its primary end point of improving progression-free survival (PFS) with satri-cel. I would note, however, that the results were quite modest. So, for the satri-cel arm, PFS was about 3 months and overall survival (OS) was about 8 months. These are relatively modest outcomes, but enough to win out against standard of care third-line chemotherapy.

And so, in June 2026 regulatory authorities in China approved satri-cel as third-line therapy for patients with CLDN18.2-positive, HER2-negative advanced gastric/GEJ cancer.² This was the first regulatory approval anywhere in the world for a CAR T-cell product for solid tumors as a whole, not just GI cancers. I think looking at the data, the question becomes, is chemotherapy the appropriate comparator arm for this CAR-T product, or is it potentially some of the emerging CLDN18.2 antibody-drug conjugates (ADCs), some of which are showing response rates in the [30%] and 40% range.

It remains to be seen how CAR-T stacks up against standard of care, but it is noteworthy that we have a CAR T-cell product in gastric cancer that is safe and does have some activity. Ultimately, I really consider this much more of a beginning than an end. Hopefully, we will eventually have more potent second-generation CAR T-cell products.

Moving to a different treatment type, what is the current state of bispecific antibodies in GI cancers?

A lot of these bispecifics have involved a tumor-associated antigen, such as CLDN18.2 and HER2, and then attached to that, the other part of the antibody is typically some effector component. I have been involved in studies where it’s HER2 or CLDN18.2, and the other molecule is 4-1BB, which is expressed on activated T cells. So it is an immune agonist molecule, and the idea then is that we're kind of bringing together physically a CLDN18.2 or HER2 or some other tumor-associated antigen positive cancer cell and an activated T cell, and allowing them to engage.

But at the same time, there are lots of other bispecifics. There are CD3 bispecifics called T-cell engagers that conceptually do the same thing. But we also see immune checkpoint inhibitor molecule bispecifics. For example, we see CTLA4, PD-1, PD-L1. We see 4-1BB and PD-1/PD-L1, which are then designed to stimulate the immune system in 2 different ways.

Mechanistically, these work in a variety of different ways but conceptually the idea is that we’re trying to move beyond the first-generation antibodies, such as trastuzumab, zolbetuximab, or zanidatamab. And then in terms of ICIs, we're trying to move beyond PD-1 and PD-L1. They really do have a variety of different mechanisms of action, and

Looking at the perioperative setting, what is the significance of the MATTERHORN trial results?

The MATTERHORN trial really is a game changer. It was a phase 3 study that demonstrated that adding durvalumab (Imfinzi), an anti–PD-L1 antibody, to perioperative chemotherapy with the FLOT regimen (fluorouracil, leucovorin, oxaliplatin, and docetaxel) improved pathologic complete response rates, PFS, and most importantly, OS, compared to perioperative FLOT alone in patients with gastric/GEJ cancers.³ Based on these findings, the FDA approved the durvalumab-based regimen in this setting in November 2025.⁴

The regimen is now the new standard of care. So in general, any patient with a resectable, locally advanced, gastric/GEJ adenocarcinoma should receive the durvalumab combination, unless there are obvious contraindications to immunotherapy.

Are there emerging combinations or treatment strategies in the perioperative setting that you are excited about?

The most promising next step is really looking at targeted populations, such as the HER2-positive population. There was a German study called PHERFLOT that was presented at the 2025 ESMO Congress last year called where they added pembrolizumab (Keytruda) and trastuzumab to FLOT in the perioperative setting for patients with HER2-positive disease.⁵ The data are relatively immature, but they reported a pathologic complete response rate of 50% [in resected patients], which is mind-boggling, and quite a contrast to the pathologic complete response rate of 19% for durvalumab/FLOT in all-comers in the MATTERHORN trial.

So really, I think the suggestion is that targeting HER2 can greatly increase pathologic complete response rates. At this point, survival data are immature, but in general, there is a good correlation between pathologic complete response to systemic therapies, and eventually, PFS and OS. This is an extremely promising approach and there are phase 3 studies that are being developed to target the HER2 population.

Now adjacent to that and comparatively rarer is the mismatch repair–deficient (dMMR) or microsatellite instability–high (MSI-H) population. We've known for many years now that dMMR/MSI-H tumors, which comprise about 7% to 8% of operable gastric/GEJ tumors, are highly responsive to immunotherapy; there are now published, well-established data that pre-operative immunotherapy with either combination immune checkpoint inhibitors (ICI; anti-CTLA4, anti-PD1, anti-PDL1) or even single agent ICI result in very high pathologic complete response rates of about 50% to 60%.

So that has essentially become the standard of care in this setting and also I think in the HER2-positive setting. The next generation of studies are also focusing on the idea of non-operative management. If we have a high likelihood of being able to eliminate the cancer, can we also safely avoid surgery in the patients that we think have a complete response?

What are the next breakthroughs you hope to see in the GI cancers space?

I think cellular therapy in solid tumors overall, but certainly in GI cancers and upper GI cancers, is really in its infancy. I would love if we had more potent constructs that led to more durable responses. If we could [replicate] even a little bit of the success [that cellular therapies have had] in hematologic malignancies, that would truly be a game-changer.

I also hope to also see strategies advancing in the perioperative setting. So, again, HER2-positive, MMR deficient, and even CLDN18.2-positive patients in the perioperative setting, we are beginning to look at adding anti-CLDN18.2 therapies to current standard of care.

And I would finish by saying that the lowest hanging fruit, meaning it’s the most broadly applicable, but what has also been the most challenging is to develop a next-generation ICI.Beyond PD-1 and PD-L1 inhibitors broadly in solid tumors, there really has not been a validated next-generation treatment. Now, we may be getting signals that that a PD-1/VEGF bispecific may be better than a PD-1/PD-L1 alone; I think that remains to be seen outside of lung cancer. But I think the next-generation ICI to replace the current era would be a critical [development].

And lastly, I do not think we really understand the intrinsic and acquired mechanisms of resistance to let's say, first line immunotherapy plus chemotherapy combinations. If we were able to understand that and we were able to find an effective salvage immunotherapy treatment in the second-line setting, the amount of benefit that it would bring to our patients really would be tremendous. So again, I think we really have made significant progress in the last 10 or 15 years, but I think it's really only highlighted the additional challenges ahead of us.

References

1. Qi C, Liu C, Peng Z, et al. Claudin18.2-specific CAR T cells (Satri-cel) versus treatment of physician's choice (TPC) for previously treated advanced gastric or gastroesophageal junction cancer (G/GEJC): primary results from a randomized, open-label, phase II trial (CT041-ST-01). J Clin Oncol. 2025;43(16 suppl):4003. doi:10.1200/JCO.2025.43.16_suppl.4003

2. CARsgen Announces Approval of Satri-cel, the World’s First CAR T-Cell Therapy Product for Solid Tumors. Published online Jun 22, 2026. Accessed September 4, 2026. https://tinyurl.com/38j2tu9v

3. Janjigian Y, Al-Batran S-E, Wainberg Z, et al. Event-free survival (EFS) in MATTERHORN: a randomized, phase 3 study of durvalumab plus 5-fluorouracil, leucovorin, oxaliplatin and docetaxel chemotherapy (FLOT) in resectable gastric/gastroesophageal junction cancer (GC/GEJC). J Clin Oncol. 2025;43(suppl 17):LBA5. doi:10.1200/JCO.2025.43.17_suppl.LBA5

4. FDA approves durvalumab for resectable gastric or gastroesophageal junction adenocarcinoma. News release. US FDA. Published online November 25, 2025. Accessed September 4, 2026. https://tinyurl.com/mvaftssb

5. Goekkurt E, Stein A, Al-Batran SE, et al. Pembrolizumab and trastuzumab in combination with FLOT in the perioperative treatment of HER2-positive localized esophagogastric adenocarcinoma: interim analysis of the phase II PHERFLOT/IKF-053 trial of the AIO study group (AIO STO 0321). Ann Oncol. 2025;36(suppl). doi:10.1016/j.annonc.2025.08.2717