Positron Emission Tomography–Computed Tomography Imaging, Genomic Profile, and Survival in Patients With Head and Neck Cancer Receiving Immunotherapy

Abstract

This study analyzes the association between maximum standardized uptake value on positron emission tomography–computed tomography and survival in patients with head and neck cancer who are receiving immune checkpoint inhibitor drugs.

Identifying biomarkers that predict survival outcomes and response to treatment is an unmet need in oncology. Few biomarkers predictive of response to immune checkpoint inhibitor (ICI) drugs have been described for patients with head and neck squamous cell cancer (HNSCC). Maximum standardized uptake value (SUVmax) on positron emission tomography–computed tomography (PET-CT) has been previously reported as a potential clinical predictor for survival outcome among patients with HNSCC.¹ Our objective was to analyze the association between SUVmax on PET-CT and survival in patients with head and neck cancer receiving ICI therapies.

Methods

After approval by the Memorial Sloan Kettering Cancer Center Institutional Review Board and patient written informed consent, we analyzed clinical and genomic data from patients treated with ICI for HNSCC between 2013 and 2018. Patients who received PET-CT imaging within 180 days prior to ICI start date were included, and SUVmax was categorized as above or below the median. Other clinical and genomic covariates with prior evidence of prognostic value were analyzed, including pretreatment peripheral blood neutrophil-to-lymphocyte ratio (NLR) and tumor mutation burden (TMB), as assessed using Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT), a targeted genomic sequencing panel of 341-468 genes.² The outcomes analyzed were overall and progression-free survival, and response to treatment was assessed using RECIST version 1.1 criteria. Analyses were performed using Stata version 16.

Results

Of 143 patients with mucosal HNSCC treated with ICI, 98 met inclusion criteria for analysis. Median age was 62.2 years (interquartile range [IQR], 51.8-69.3 years), and 77 (78.6%) were male. Median SUVmax was 11 (IQR, 8.4-15.0), and was localized to the head and neck in 47 patients (48.0%); lung, 30 (30.6%); bone, 11 (11.2%); and gastrointestinal tract/liver, 10 (10.2%). The site of SUVmax was at a primary recurrence site rather than metastatic site in 31 patients (31.6%). In patients with metastases, the mean number of metastatic sites was 1.9. Immune checkpoint inhibitor treatment was given for recurrent rather than metastatic disease in 19 patients. Average time from PET-CT to start of ICI therapy was 52 days.

High SUVmax (above median) was associated with poorer overall survival (HR, 1.65; 95% CI, 1.03-2.66; P = .04) (Figure 1). This association was similar when analyzing SUVmax as a continuous variable (HR, 1.05; 95% CI, 1.00-1.09; P = .03). The association between high SUVmax (above median) and poorer overall survival remained significant with adjustment for Eastern Cooperative Oncology Group performance status (HR, 1.63; 95% CI, 1.01-2.63; P = .046). In this study, SUVmax was not associated with progression-free survival (HR, 0.92; 95% CI, 0.60-1.40; P = .69) or the probability of tumor response to ICI (20.4% vs 26.5%, P = .48). In addition, SUVmax was not associated with TMB (r = −0.01, P = .89), but was significantly associated with pretreatment NLR (r = 0.32, P = .002) (Figure 2).

Figure 1. Kaplan-Meier Curve Demonstrating Overall Survival Based on SUVmax Low vs High Using Median as Cutoff Among Patients With HNSCC Treated With Immune Checkpoint Inhibitors.

HNSCC indicates head and neck squamous cell carcinoma; SUVmax, maximum standardized uptake value.

Figure 2. Scatter Plots of PET-CT SUVmax and (A) Tumor Mutational Burden or (B) Neutrophil-to-Lymphocyte Ratio .

PET-CT indicates positron emission tomography–computed tomography; SUVmax, maximum standardized uptake value.

Discussion

In patients treated with ICI for recurrent or metastatic HNSCC, high SUVmax was associated with poorer prognosis (Figure 1). High SUVmax was also significantly correlated with NLR, a biomarker with both prognostic and predictive value in patients with HNSCC treated with ICI.^3,4 There was no association between SUVmax and TMB, despite previous studies suggesting an association in other cancer types.⁵

Studies examining the prognostic value of SUVmax in HNSCC have produced mixed results. In the context of ICI, we hypothesize that higher levels of systemic inflammation, of which NLR is a surrogate marker, may in part be reflected in higher lesional SUVmax. This correlation has been identified in other cancer types.⁶ These data suggest that windows into systemic and tumor inflammation with laboratory and tumor metabolic imaging tests may provide useful prognostic information for HNSCC patients treated with ICI. An important limitation of this analysis is that several other factors can influence SUVmax, such as lean body mass, blood glucose level, and postinjection uptake time, meaning interpretation must be taken in context for each patient. Other factors can also influence survival, including overall burden of comorbid illness. Additional research, including advanced radiomic technologies, may provide insights to allow clinicians to more effectively predict response to ICI therapy based on radiologic findings.