PD-1 blockade represents a promising treatment in patients with head and neck squamous cell carcinoma (HNSCC). We analyzed results of a neoadjuvant randomized window-of-opportunity trial of nivolumab plus/minus tadalafil to investigate whether immunotherapy-mediated treatment effects vary by site of involvement (primary tumor, lymph nodes) and determine how radiographic tumor shrinkage correlates with pathologic treatment effect.
Forty-four patients enrolled in trial NCT03238365 were treated with nivolumab 240 mg intravenously on days 1 and 15 with or without oral tadalafil, as determined by random assignment, followed by surgery on day 31. Radiographic volumetric response (RVR) was defined as percent change in tumor volume from pretreatment to posttreatment CT scan. Responders were defined as those with a 10% reduction in the volume of the primary tumor or lymph nodes (LN). Pathologic treatment effect (PTE) was defined as the area showing fibrosis or lymphohistiocytic inflammation divided by total tumor area.
Sixteen of 32 patients (50%) with pathologic evidence of LN involvement exhibited discordant PTE between primary sites and LN. In four patients with widely discordant adjacent LN, increased PTE was associated with increased infiltration of tumor CD8+ T cells and CD163+ macrophages, whereas stromal regulatory T cells were associated with low nodal PTE. RVR correlated with PTE at both primary tumor (slope = 0.55,
Nivolumab often induces discordant treatment effects between primary tumor sites and metastatic lymph nodes within subjects. This treatment discordance was also demonstrated in adjacent lymph nodes, which may correlate with local immune cell makeup. Finally, although these data were generated by a relatively small population size, our data support the use of early radiographic response to assess immunotherapy treatment effect in HNSCC.