Imagine a world where a simple test score could decide whether a cancer patient gets access to potentially life-saving treatment—but what if that test isn't as reliable as we think? That's the gripping reality we're diving into today as we explore the role of PD-L1 Combined Positive Score (CPS) in choosing frontline immunotherapy for gastric cancer. You might be surprised to learn how experts grapple with this in real-world practice, and trust me, there are twists that could challenge everything you believe about personalized medicine. Let's unpack this together, step by step, so even if you're new to oncology, you'll feel like you're in the room with the specialists.
But here's where it gets controversial: Is the PD-L1 CPS truly a game-changer, or is it more of a gamble? During an engaging live event, Yelena Y. Janjigian, MD, led a fascinating discussion on leveraging PD-L1 CPS status for immunotherapy decisions in gastric cancer cases. For those just starting out, PD-L1 refers to a protein on cancer cells that helps them evade the immune system—think of it as a sneaky shield. Immunotherapy, or IO, uses drugs to block this shield, allowing the body's defenses to attack the tumor. The goal? Better outcomes without always relying on harsh chemotherapy.
The National Comprehensive Cancer Network (NCCN) guidelines are pretty clear: they suggest incorporating immune checkpoint inhibitors for advanced esophageal and gastric cancers when the PD-L1 CPS is at least 1, but they only give it a top-tier (category 1) recommendation if the score hits 5 or higher. Imagine CPS like a scale from 0 to 100, where higher numbers indicate more PD-L1 activity—and potentially more response to IO. In a virtual Case-Based Roundtable, Dr. Janjigian posed a critical question to participants: How do you handle treatment for patients with CPS scores between 1 and 5? As the chief of gastrointestinal oncology and holder of the Carroll and Milton Petrie Chair at Memorial Sloan Kettering Cancer Center, she emphasized the overall trustworthiness of PD-L1 scores and stressed why seizing the chance for IO in the first-line setting is crucial. For beginners, first-line treatment means the initial round of therapy right after diagnosis, often the best shot at controlling the disease before it progresses.
And this is the part most people miss: The real-world application often defies textbook rules. Curious to join similar discussions? Sign up today for a Case-Based Roundtable near you (https://join.targetedonc.com/). This piece is part 2 of a 2-part series from that event (https://www.targetedonc.com/view/dosing-schedule-factors-into-chemo-ici-choice-in-upper-gi-cancer).
To spark your thoughts, here are some key discussion questions from the experts:
- How does the CPS score shape your choices in treatment?
- What cutoff point, like CPS ≥ 5 versus 1–4, guides your decision to include IO?
- Have you noticed real-world benefits for patients with low CPS scores (1–4)?
- What are the drawbacks of relying on CPS for first-line therapy, and how does that impact your clinical decisions?
Dr. Janjigian kicked off the dialogue by asking: When do you think about adding an immune checkpoint inhibitor alongside first-line chemotherapy? What specific cutoff do you employ?
Dr. Haiyun Wang replied: I start at a CPS of 1.
Dr. Fayez Estefan elaborated: I begin discussing IO at CPS 1. If Claudin18.2—a biomarker on certain stomach cells—is negative, I engage in a thorough conversation with the patient about the possible advantages and side effects, emphasizing shared decision-making. Remember, if you skip IO upfront, it might be the only window to use it, and they could be among those who benefit. I go over the evidence with patients whose scores range from 1 to 5 and collaborate on the choice.
Dr. Janjigian followed up: Do you hold off until you have the PD-L1 CPS results before deciding?
Dr. Estefan explained: Our pathology team runs HER2, mismatch repair (MMR), and PD-L1 tests for all gastric and gastroesophageal junction adenocarcinomas upfront. By the time I meet the patient, we usually have the data.
Dr. Wang agreed: It's much the same in our practice. We handle PD-L1, HER2, and MMR internally, so results are typically ready when I see the patient. If not, we wait—just a week tops. I base IO decisions on CPS, setting 1 as the threshold. If it's 1 or more, I'll incorporate it. Have you ever tried IO with less than 1% PD-L1 expression?
Here's where opinions diverge sharply: Some experts swear by retesting PD-L1 due to inconsistencies, while others trust the initial score. Could this variability be undermining patient care? Dr. Janjigian shared: Absolutely, I've encountered this frequently. Patients often arrive with pending PD-L1 tests from other places, and when we retest, scores can flip from negative to positive or vice versa. In practice, PD-L1 testing lacks perfect reliability—I've seen discrepancies even on the same sample. For instance, I once had a patient with HER2-positive, PD-L1-negative disease, which was unusual. Retesting revealed a CPS of 8 on a different sample from the same period. The initial test was on a lymph node overrun by tumor, lacking stroma (the supportive tissue around cells), which might affect results. This back-and-forth happens a lot in clinics. Yet, there's solid level 1 evidence from phase 3 trials in controlled settings with top-notch samples and consistent reviewers—showing that low PD-L1 tumors don't respond as well, though without harm. Survival perks are smaller. But in everyday practice, sample quality and testing vary wildly, so I lean toward anti-PD-1 therapy, especially since options like Claudin18.2 inhibitors come with their own challenges.
Dr. Bolanle Adepoju brought up a case: A patient switched care late in FOLFOX (a chemo combo of leucovorin, fluorouracil, and oxaliplatin) treatment. He was HER2-positive, on FOLFOX plus trastuzumab (Herceptin), but no PD-L1 therapy was initiated—PD-L1 CPS wasn't checked in his records. He's now on cycle 12, oxaliplatin stopped, moving to maintenance. Thoughts on adding PD-L1 therapy? I'll review scans soon. If stable, should I add anti-PD-L1?
Dr. Janjigian advised: Yes, I would. If oxaliplatin is discontinued, the tumor tests PD-L1 positive, there's remaining disease, and the patient is still strong, I'd introduce anti-PD-1 now. He's still in first-line technically, so it's approved. If progression occurs, might he tolerate trastuzumab deruxtecan (T-DXd) in second-line? Possibly not, and anti-PD-1 definitely boosts survival upfront.
Dr. Yanyan Lou asked: PD-L1 expression varies in lung adenocarcinoma—given potential benefits, do you rebiopsy if PD-L1 negative?
Dr. Janjigian responded: I usually retest or consult the pathologist on sample quality. Sometimes tests use tumor proportion score instead of CPS, or the sample's poor, leading to errors.
Dr. Sheila Donnelly inquired: Have you seen long-term survivors with nivolumab (Opdivo) plus chemotherapy in squamous or adenocarcinoma with liver metastases?
Dr. Janjigian noted: In our experience, unlike colon cancer, liver mets aren't a bad prognostic sign for these drugs. Peritoneal disease is the bigger issue—with ascites and such, outcomes suffer. But small to medium liver mets aren't a major worry; the battle's often in the abdomen.
Have you tried nivolumab plus ipilimumab (Yervoy) in first-line squamous cases?
Dr. Estefan said: I'd go with chemo plus single-agent IO.
Dr. Janjigian concurred: They often present quite ill, needing rapid response. I'm cautious about dropping chemo entirely.
What about after chemo/IO fails? With many patients not getting beyond one treatment, why prioritize the strongest first? What do you use second-line?
Dr. Estefan suggested: A taxane with ramucirumab (Cyramza).
Dr. Janjigian asked: And for squamous?
Dr. Ike Onwere replied: Typically the same.
Dr. Janjigian clarified: Like carboplatin/paclitaxel or paclitaxel alone. Ramucirumab isn't approved for squamous. For HER2-positive?
Dr. Muhammad Masab answered: T-DXd.
Dr. Janjigian probed: Experiences with T-DXd?
Dr. Masab shared: One patient didn't fare well; they were frail and declined after a few cycles.
Dr. Janjigian checked: Did they receive pembrolizumab (Keytruda) and trastuzumab upfront?
Dr. Masab confirmed: HER2-positive on rebiopsy, so chemo plus IO first-line.
Ready for more? Register for a Case-Based Roundtable (https://join.targetedonc.com/).
Disclosures: Dr. Janjigian has previously disclosed research funding from Bayer, Bristol Myers Squibb, Cycle for Survival, Department of Defense, Fred’s Team, Genentech/Roche Lilly, Merck, National Cancer Institute, and Rgenix; advisory roles for Basilea Pharmaceutical, Bayer, Bristol Myers Squibb, Daiichi Sankyo, Imugene, AstraZeneca, Lilly, Merck, Merck Serono, Michael J Hennessy Associates, Paradigm Medical Communications, Seattle Genetics, Pfizer, Rgenix, AmerisourceBergen, Arcus Biosciences, Geneos, GlaxoSmithKline, Imedex, Lynx Health, Peerview, Silverback Therapeutics, and Zymeworks; stock options from Rgenix; and nonfinancial ties with Clinical Care Options, Axis Medical Education, and Research to Practice.
Reference: 1. NCCN. Clinical Practice Guidelines in Oncology. Gastric cancer, version 3.2025. https://www.nccn.org/professionals/physician_gls/pdf/gastric.pdf
Stay informed: Keep up with the latest on practice-changing data in community settings.
Now, let's turn this over to you—what's your take on relying on PD-L1 CPS for treatment decisions? Do you agree that retesting is always necessary, or is it overkill? Share your thoughts in the comments; I'd love to hear if this sparks debate or aligns with your clinical experience!
