Ask AI
Back Back1

Expert Perspectives on Questions From Healthcare Professionals on Recurrent and Metastatic Head and Neck Cancer Management

Start Activity
Clinical Thought
Clinical Thought Start Activity
See Activity Information

Released: August 20, 2026

See Activity Collection

Step 1 of 2

1 ClinicalThought 2 Complete Activity

This expert commentary examines key clinical decision points in recurrent or metastatic head and neck cancer (HNC) based on questions submitted by healthcare professionals, including management of locoregional recurrence after prior definitive therapy. Topics include first-line treatment selection based on PD-L1 CPS, disease burden, and symptoms; interpretation and potential repeat testing of PD-L1 CPS; management after salvage surgery in patients with high-risk pathologic features but no measurable disease; and treatment decisions following progression on pembrolizumab. The expert Q&A also addresses the limitations of PD-L1 as a predictive biomarker, the role of multidisciplinary evaluation, and the importance of distinguishing limited or uncertain progression from clear treatment failure.

Robert Haddad
Robert Haddad, MD
Victoria M. Villaflor
Victoria M. Villaflor, MD
Expert Perspectives on HNC

Continue
Key Takeaways
  • Pembrolizumab monotherapy can be considered for a patient with low PD-L1 CPS and low tumor burden with reassessment of response after 2 months.
  • Multidisciplinary evaluation is recommended before proceeding with a definitive salvage approach for patients who develop recurrence after prior chemoradiation.

Head and neck cancer (HNC) encompasses a heterogeneous group of malignancies in which treatment decisions are often influenced by prior therapy, disease burden, biomarker expression, and the extent and pace of recurrence. In recurrent or metastatic disease, healthcare professionals must weigh the role of PD-L1 combined positive score (CPS) in selecting immunotherapy, determine when repeat biomarker testing may be informative, consider options following salvage surgery, and distinguish limited or uncertain progression from clear treatment failure on pembrolizumab. These decisions require an individualized, multidisciplinary approach that balances efficacy, toxicity, and future treatment options. Experts Robert Haddad, MD, and Victoria M. Villaflor, MD, addressed audience questions on key clinical considerations in HNC gathered from multiple meetings at cancer centers across the United States held April through July 2026.

Would you favor chemoimmunotherapy in a patient with HNC, PD-L1 CPS9, and a low disease burden?

Robert Haddad, MD:
For a patient with a low disease burden, I would start with pembrolizumab alone rather than pembrolizumab plus chemotherapy. Then, I would reassess after approximately 2 months of treatment. If the patient were responding, I would continue pembrolizumab alone to avoid the added toxicity of chemotherapy.

Victoria M. Villaflor, MD:
In an asymptomatic patient with a low disease burden and PD-L1 expression, I would favor first-line pembrolizumab monotherapy. I would reassess after approximately 3 cycles, or 8-9 weeks. If the patient remained clinically stable with stable or responding disease, I would continue single-agent pembrolizumab. This approach minimizes treatment-related toxicity and aligns with our goals of improving both quality and quantity of life by avoiding the additional toxicity associated with cytotoxic chemotherapy.

When do you repeat or question a CPS test result?

Robert Haddad, MD:
The answer largely depends on tissue availability. If a patient with metastatic disease has had a biopsy of a metastatic site, I would use the CPS result from that specimen and generally favor testing the most recent biopsy available. If the patient were previously treated for locally advanced disease and only the original diagnostic biopsy is available, it is reasonable to use the CPS result from that specimen to guide treatment.

In HNC, a CPS of 0 is relatively uncommon. In KEYNOTE-048, approximately 15% of patients had a CPS <1, but based on my personal experience, the number is much lower. If a CPS result is 0 and you question its accuracy, consider retesting using another available tissue sample. If multiple specimens are available, I would test the most recent one and use that result to guide therapy. If only the initial diagnostic biopsy is available, its CPS result can be used to inform treatment.

Victoria M. Villaflor, MD:
I generally consider repeat PD-L1 CPS testing when adequate new tissue is available, particularly if the prior specimen is older or the disease has evolved. If the prior CPS result was 0, repeat testing on a more recent specimen may also be reasonable given the known spatial and temporal heterogeneity of PD-L1 expression and potential sampling and interpretive variability. Clinically, this can be particularly important when establishing a CPS ≥1 would support eligibility for first-line pembrolizumab monotherapy.

More broadly, PD-L1 remains an imperfect predictive biomarker. I typically place greater weight on the most recent available specimen, but even a high CPS does not reliably predict response, and a low or negative CPS does not necessarily exclude benefit from immunotherapy. In practice, an important reason for PD-L1 testing in recurrent/metastatic HNC is to establish eligibility and access to checkpoint inhibitor therapy. Unfortunately, we still lack a sufficiently reliable biomarker that can accurately identify which patients will derive meaningful benefit from immunotherapy.

For patients with recurrent HNC previously treated with chemoradiation who undergo salvage surgery and have high-risk pathologic features, such as positive margins or extranodal extension, but no measurable disease, do you favor observation or immediate adjuvant therapy?

Robert Haddad, MD:
For those patients, I would first consider whether reirradiation is feasible, particularly if the recurrence is localized, there is extranodal extension, and a sufficient interval has elapsed since prior radiation. Because reirradiation can be associated with substantial toxicity, this decision should involve an experienced radiation oncologist.

If there is no measurable disease after salvage surgery, I generally defer systemic therapy. In addition to the lack of measurable disease, initiating chemotherapy or pembrolizumab in this setting could limit future eligibility for first-line clinical trials if the patient later develops measurable recurrent or metastatic disease. I therefore reserve systemic therapy for patients with measurable disease and, in the postoperative setting, routinely assess whether a focused reirradiation approach is appropriate.

Victoria M. Villaflor, MD:
For patients who develop recurrence after prior chemoradiation, salvage surgery may certainly remain an appropriate option, particularly when the disease is resectable. However, before proceeding with a definitive salvage approach, whether surgery, reirradiation, systemic therapy, or a combination, I believe these patients should, when feasible, undergo multidisciplinary evaluation at a high-volume center, with a second opinion considered when appropriate.

The goal is not necessarily to avoid surgery but rather to ensure that the full range of options has been considered before treatment begins. This is particularly important for more locally advanced recurrences, shorter disease-free intervals, or cases in which reirradiation ultimately may be part of the treatment strategy. Evaluation at a center with expertise in recurrent HNC, reirradiation, complex salvage surgery, and clinical trials can help optimize treatment selection and preserve access to investigational approaches that may no longer be available once definitive therapy has been initiated.

For patients with HNC who experience disease progression on pembrolizumab, would you add chemotherapy or discontinue pembrolizumab?

Robert Haddad, MD:
If there is clear evidence of progressive disease, I would stop pembrolizumab and transition to chemotherapy. If it is stable disease with some progression, maybe 10% or 15%, or a poor partial response, adding chemotherapy to pembrolizumab could be considered. However, with unequivocal progression, such as new lesions or substantial growth of existing disease, I would stop pembrolizumab.

Victoria M. Villaflor, MD:
This is an excellent question and can be difficult to answer because the decision depends heavily on how the patient is doing clinically and whether there appears to be ongoing clinical benefit from pembrolizumab. Pseudoprogression, although uncommon, is an important consideration with immunotherapy, and the goal is to avoid discontinuing an effective treatment in a patient whose apparent radiographic progression does not represent true disease progression.

If the patient remains asymptomatic and the degree of progression is limited or uncertain, it can be reasonable to continue pembrolizumab for a short period and then reassess. Additional information may help clarify the findings, including trends in human papillomavirus–associated circulating tumor DNA when available or, when feasible, biopsy of an enlarging lesion for pathologic confirmation.

On the other hand, if the patient were clearly experiencing significant clinical and radiographic progression, particularly with increasing symptoms or substantial disease growth, I would generally discontinue pembrolizumab and transition to enrollment in an appropriate clinical trial, a cytotoxic chemotherapy regimen, or an EGFR inhibitor rather than continuing an ineffective checkpoint inhibitor.

Your Thoughts
What factors most influence your treatment decisions for patients with recurrent or metastatic HNC, and how do these considerations affect your clinical practice?

Continue

Poll

1.

How frequently do you evaluate human papillomavirus status in patients with HNC?

Submit