On October 27, 2026, Nature Medicine published results from the MDT-BRIDGE phase 2 trial, a phase 2 study evaluating neoadjuvant durvalumab plus platinum-based chemotherapy in patients with resectable or borderline resectable non-small cell lung cancer. The trial’s clinical findings are meaningful on their own terms. Its operational architecture, specifically the structured multidisciplinary tumor board (MTB) reassessment protocol embedded in the trial design, presents a more durable contribution to how adaptive decision-making in surgical oncology can be formalized at the protocol level.

The design challenge MDT-BRIDGE addresses directly is one that FDA’s current perioperative NSCLC framework leaves underspecified. AstraZeneca received FDA approval for durvalumab in resectable NSCLC on August 15, 2024, covering neoadjuvant durvalumab plus platinum-containing chemotherapy followed by adjuvant monotherapy. That approval was built on the AEGEAN trial’s Phase 3 dataset, which enrolled surgically resectable Stage II–IIIA patients with pre-confirmed operability. MDT-BRIDGE enrolls a materially different population: at baseline, a substantial proportion of participants were classified as borderline resectable, a category where the surgical endpoint itself is contingent on treatment response and the MTB’s post-induction judgment.

FDA’s existing guidance on pathological endpoints in oncology, published by CDER and available at the agency’s guidance portal, acknowledges pathological complete response (pCR) as a surrogate endpoint in certain accelerated approval contexts. For NSCLC, however, the agency’s primary evidentiary standard for regular approval remains overall survival, with progression-free survival acceptable in unmet-need populations showing a large treatment effect. The available FDA guidance reviewed does not appear to provide a prescriptive standard for how resectability status should be reassessed after neoadjuvant treatment, who should perform that reassessment, or how protocol-mandated MTB deliberation should be documented for a regulatory submission.

What the MTB Architecture Actually Does

MDT-BRIDGE’s protocol embeds MTB reassessment as a decision gate rather than a clinical courtesy. Per the trial design, patients underwent structured multidisciplinary review after completing neoadjuvant treatment, with surgical operability confirmed or revised before proceeding to resection. The trial enrolled its first patient in February 2024 and, with an overall resection rate reported across the full cohort. That 74% figure carries different regulatory weight depending on how resectability was initially classified: for the patients who were resectable at baseline, resection represents confirmation of maintained operability; for the 35.2% who were borderline resectable, it represents a protocol-defined outcome of downstaging.

The distinction matters for sponsors designing future neoadjuvant trials with borderline resectable populations. When surgical resection is itself an endpoint, the MTB’s reassessment process becomes part of the evidentiary chain. Without pre-specified criteria for who participates in the MTB, what imaging and pathological inputs are required, what the decision rules are, and how the deliberation is documented, the resulting resection-rate data carries significant interpretability risk at review. A sponsor submitting a perioperative NDA with a mixed resectable/borderline-resectable population will face reviewer questions that MDT-BRIDGE’s protocol directly anticipates.

ICH E6(R3), establishes that critical process steps must be identified and controlled proportionate to risk. MTB reassessment in a trial where operability is a primary endpoint qualifies as a critical process step under that framework. MDT-BRIDGE’s design operationalizes this in practice. What remains absent from FDA’s perioperative NSCLC guidance is an explicit expectation that sponsors pre-specify MTB composition, quorum requirements, and documentation standards as protocol-level quality controls rather than leaving them to site clinical practice.

Implications for Sponsors Filing Perioperative INDs in the Next 12 Months

Sponsors with Phase 2 or Phase 3 perioperative NSCLC programs that include borderline resectable patients should treat MDT-BRIDGE’s operational design as a reference model, not a published dataset to cite for efficacy comparisons. The NCCN Clinical Practice Guidelines in Oncology for NSCLC, Version 5.2024 updated April 2024, expanded neoadjuvant immunotherapy recommendations to the full ICI class plus chemotherapy. That expansion means more sponsors will be designing perioperative trials with heterogeneous resectability populations. The evidentiary challenge those sponsors face is not generating a resection rate. It is documenting the reassessment process that produced the rate in a format a regulatory reviewer can evaluate.

For sponsors with INDs currently under FDA review for neoadjuvant indications, a Type B meeting request that specifically queries the agency’s expectations for MTB documentation standards in mixed-resectability populations would generate useful written responses before enrollment begins. FDA’s December 2019 guidance on adaptive designs for clinical trials of drugs and biologics addressed adaptive design principles, but the adaptive element in perioperative surgical trials is structurally different: it is a within-subject treatment decision governed by clinical reassessment rather than a statistical boundary. Sponsors who pre-specify that decision architecture in their protocols, with explicit MTB criteria, create a cleaner dataset for both interim analysis and final submission.

Site networks and CROs running perioperative oncology programs should audit current MTB documentation practices against what a regulatory reviewer would require. Specifically: is the MTB composition pre-specified per protocol or assembled ad hoc at each site? Are the imaging criteria for resectability reassessment standardized across sites or left to local radiological interpretation? If a site’s answer to either question is “ad hoc” or “local practice,” the trial’s resection-rate data carries site-level heterogeneity that will require explanation in the clinical study report.

The Broader Regulatory Pattern

MDT-BRIDGE sits inside a pattern of Phase 2 perioperative trials generating operational data that Phase 3 programs will need to incorporate. KEYNOTE-671 (pembrolizumab, Merck) and IMpower030 (atezolizumab, Roche/Genentech) were Phase 3 randomized controlled trials in resectable NSCLC that helped establish the perioperative immunotherapy paradigm. Those trial designs were not primarily structured to address the borderline-resectable reassessment problem. As label expansion discussions for existing approved perioperative agents move toward earlier or more surgically ambiguous patients, the operational gap MDT-BRIDGE addresses will become a Phase 3 design question, not just a Phase 2 exploratory one.

FDA has not, in the sources reviewed, published a guidance document specific to MTB standards in perioperative oncology trials. The agency’s broader real-world data framework and the ICH E20 adaptive designs draft guidance, both provide relevant principles, but neither addresses surgical reassessment decision gates explicitly. The MDT-BRIDGE protocol’s value to the field, beyond its efficacy data, is that it provides a prospectively executed, Nature Medicine-published model of what structured MTB documentation in a regulatory-grade trial can look like.

Sponsors planning perioperative NSCLC submissions in 2027 and beyond should review the MDT-BRIDGE protocol’s MTB specification language before finalizing their own. The next signal worth tracking is whether FDA issues a perioperative oncology trial design guidance that addresses resectability reassessment standards explicitly, or whether the agency addresses it first through advisory committee questions during a future perioperative NDA review.

References

  1. Nature Medicine, “Neoadjuvant durvalumab plus chemotherapy in resectable or borderline resectable non-small-cell lung cancer: the phase 2 MDT-BRIDGE trial”
  2. AstraZeneca Clinical Trials, MDT-BRIDGE trial (D9106C00002): enrollment, baseline resectability, and resection rate data
  3. Drozdogan.com, “Neoadjuvant/Adjuvant Durvalumab Receives FDA Approval for Resectable Lung Cancer,” August 15, 2024
  4. ARUP Consult, NCCN Guidelines for Non-Small Cell Lung Cancer, Version 5.2024 (updated April 2024)
  5. FDA CDER, Pathological Complete Response in Neoadjuvant Treatment of High-Risk Early-Stage Breast Cancer: Guidance for Industry
  6. CASRAI, FDA Guidance: Adaptive Designs for Clinical Trials of Drugs and Biologics (finalized December 2019)
Website | + posts

Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.