The success of neoadjuvant immunotherapy trials like Niagara and those with EV-Pembro means most patients will receive immune therapy before surgery. This fundamentally shifts the clinical landscape, making the question of starting adjuvant immunotherapy less relevant as perioperative treatment becomes the standard.

The CheckMate 9LA regimen provides exceptional benefit to PD-L1 negative and squamous histology NSCLC patients. This is significant because these subgroups often respond poorly to other immunotherapy combinations, with Dr. Carbone noting some trials where the control arm outperformed pembrolizumab in these patients.

In frontline clinical trials for KRAS G12C NSCLC, combining olomorasib with pembrolizumab alone yielded a 90% response rate in patients with >50% PD-L1 expression. This surpassed the 78% rate seen when chemotherapy was added, suggesting a more targeted approach may be superior for this specific biomarker-defined subgroup.

The HARMONY-2 study showed Ivanesimab delivered a median progression-free survival of 11.3 months compared to 5.8 months for Pembrolizumab in PD-L1 positive NSCLC. Analysis confirmed Pembrolizumab performed as expected, suggesting the dual VEGF/PD-1 blockade provides a genuinely superior clinical benefit over PD-1 inhibition alone.

While neoadjuvant-only immunotherapy has a strong rationale, a patient-level cross-trial comparison of CheckMate 816 (neoadjuvant) and 770T (perioperative) suggests the addition of adjuvant therapy improves event-free survival, favoring a full perioperative approach.

The Matterhorn study found that adding the immunotherapy drug Durvalumab to FLOT chemotherapy significantly improved survival in localized gastric cancer. Surprisingly, this benefit extended to patients with low or negative PD-L1 expression, challenging the biomarker's predictive value for immunotherapy efficacy in this perioperative setting.

Despite major advances in immunotherapy, patient selection remains crude compared to targeted therapies. PD-L1 is still the primary, yet imperfect, biomarker used. Dr. Carbone highlights an urgent need to develop better predictive biomarkers to customize immunotherapy regimens, as is standard for targeted agents.

Unlike in lung cancer, PD-L1 expression levels do not guide treatment for nonmelanoma skin cancers. Patients with low or even negative PD-L1 levels still show significant response to anti-PD-1 therapy, making the test an unhelpful discriminator for treatment decisions.

Data from KEYNOTE-671 and other trials show that even patients achieving a pathological complete response (PCR) after neoadjuvant chemo-IO have better outcomes with the full perioperative regimen. This challenges the idea of de-escalating adjuvant therapy based on surgical pathology alone.