Development and Validation of an Automated Image-Based Deep Learning Platform for Sarcopenia Assessment in Head and Neck Cancer.

Importance: Sarcopenia is an established prognostic factor in patients with head and neck squamous cell carcinoma (HNSCC); the quantification of sarcopenia assessed by imaging is typically achieved through the skeletal muscle index (SMI), which can be derived from cervical skeletal muscle segmentation and cross-sectional area. However, manual muscle segmentation is labor intensive, prone to interobserver variability, and impractical for large-scale clinical use. Objective: To develop and externally validate a fully automated image-based deep learning platform for cervical vertebral muscle segmentation and SMI calculation and evaluate associations with survival and treatment toxicity outcomes. Design, Setting, and Participants: For this prognostic study, a model development data set was curated from publicly available and deidentified data from patients with HNSCC treated at MD Anderson Cancer Center between January 1, 2003, and December 31, 2013. A total of 899 patients undergoing primary radiation for HNSCC with abdominal computed tomography scans and complete clinical information were selected. An external validation data set was retrospectively collected from patients undergoing primary radiation therapy between January 1, 1996, and December 31, 2013, at Brigham and Women's Hospital. The data analysis was performed between May 1, 2022, and March 31, 2023. Exposure: C3 vertebral skeletal muscle segmentation during radiation therapy for HNSCC. Main Outcomes and Measures: Overall survival and treatment toxicity outcomes of HNSCC. Results: The total patient cohort comprised 899 patients with HNSCC (median [range] age, 58 [24-90] years; 140 female [15.6%] and 755 male [84.0%]). Dice similarity coefficients for the validation set (n = 96) and internal test set (n = 48) were 0.90 (95% CI, 0.90-0.91) and 0.90 (95% CI, 0.89-0.91), respectively, with a mean 96.2% acceptable rate between 2 reviewers on external clinical testing (n = 377). Estimated cross-sectional area and SMI values were associated with manually annotated values (Pearson r = 0.99; P < .001) across data sets. On multivariable Cox proportional hazards regression, SMI-derived sarcopenia was associated with worse overall survival (hazard ratio, 2.05; 95% CI, 1.04-4.04; P = .04) and longer feeding tube duration (median [range], 162 [6-1477] vs 134 [15-1255] days; hazard ratio, 0.66; 95% CI, 0.48-0.89; P = .006) than no sarcopenia. Conclusions and Relevance: This prognostic study's findings show external validation of a fully automated deep learning pipeline to accurately measure sarcopenia in HNSCC and an association with important disease outcomes. The pipeline could enable the integration of sarcopenia assessment into clinical decision making for individuals with HNSCC.

Screening for extranodal extension in HPV-associated oropharyngeal carcinoma: evaluation of a CT-based deep learning algorithm in patient data from a multicentre, randomised de-escalation trial.

BACKGROUND: Pretreatment identification of pathological extranodal extension (ENE) would guide therapy de-escalation strategies for in human papillomavirus (HPV)-associated oropharyngeal carcinoma but is diagnostically challenging. ECOG-ACRIN Cancer Research Group E3311 was a multicentre trial wherein patients with HPV-associated oropharyngeal carcinoma were treated surgically and assigned to a pathological risk-based adjuvant strategy of observation, radiation, or concurrent chemoradiation. Despite protocol exclusion of patients with overt radiographic ENE, more than 30% had pathological ENE and required postoperative chemoradiation. We aimed to evaluate a CT-based deep learning algorithm for prediction of ENE in E3311, a diagnostically challenging cohort wherein algorithm use would be impactful in guiding decision-making. METHODS: For this retrospective evaluation of deep learning algorithm performance, we obtained pretreatment CTs and corresponding surgical pathology reports from the multicentre, randomised de-escalation trial E3311. All enrolled patients on E3311 required pretreatment and diagnostic head and neck imaging; patients with radiographically overt ENE were excluded per study protocol. The lymph node with largest short-axis diameter and up to two additional nodes were segmented on each scan and annotated for ENE per pathology reports. Deep learning algorithm performance for ENE prediction was compared with four board-certified head and neck radiologists. The primary endpoint was the area under the curve (AUC) of the receiver operating characteristic. FINDINGS: From 178 collected scans, 313 nodes were annotated: 71 (23%) with ENE in general, 39 (13%) with ENE larger than 1 mm ENE. The deep learning algorithm AUC for ENE classification was 0·86 (95% CI 0·82-0·90), outperforming all readers (p<0·0001 for each). Among radiologists, there was high variability in specificity (43-86%) and sensitivity (45-96%) with poor inter-reader agreement (κ 0·32). Matching the algorithm specificity to that of the reader with highest AUC (R2, false positive rate 22%) yielded improved sensitivity to 75% (+ 13%). Setting the algorithm false positive rate to 30% yielded 90% sensitivity. The algorithm showed improved performance compared with radiologists for ENE larger than 1 mm (p<0·0001) and in nodes with short-axis diameter 1 cm or larger. INTERPRETATION: The deep learning algorithm outperformed experts in predicting pathological ENE on a challenging cohort of patients with HPV-associated oropharyngeal carcinoma from a randomised clinical trial. Deep learning algorithms should be evaluated prospectively as a treatment selection tool. FUNDING: ECOG-ACRIN Cancer Research Group and the National Cancer Institute of the US National Institutes of Health.

Fully-automated sarcopenia assessment in head and neck cancer: development and external validation of a deep learning pipeline.

Purpose: Sarcopenia is an established prognostic factor in patients diagnosed with head and neck squamous cell carcinoma (HNSCC). The quantification of sarcopenia assessed by imaging is typically achieved through the skeletal muscle index (SMI), which can be derived from cervical neck skeletal muscle (SM) segmentation and cross-sectional area. However, manual SM segmentation is labor-intensive, prone to inter-observer variability, and impractical for large-scale clinical use. To overcome this challenge, we have developed and externally validated a fully-automated image-based deep learning (DL) platform for cervical vertebral SM segmentation and SMI calculation, and evaluated the relevance of this with survival and toxicity outcomes. Materials and Methods: 899 patients diagnosed as having HNSCC with CT scans from multiple institutes were included, with 335 cases utilized for training, 96 for validation, 48 for internal testing and 393 for external testing. Ground truth single-slice segmentations of SM at the C3 vertebra level were manually generated by experienced radiation oncologists. To develop an efficient method of segmenting the SM, a multi-stage DL pipeline was implemented, consisting of a 2D convolutional neural network (CNN) to select the middle slice of C3 section and a 2D U-Net to segment SM areas. The model performance was evaluated using the Dice Similarity Coefficient (DSC) as the primary metric for the internal test set, and for the external test set the quality of automated segmentation was assessed manually by two experienced radiation oncologists. The L3 skeletal muscle area (SMA) and SMI were then calculated from the C3 cross sectional area (CSA) of the auto-segmented SM. Finally, established SMI cut-offs were used to perform further analyses to assess the correlation with survival and toxicity endpoints in the external institution with univariable and multivariable Cox regression. Results: DSCs for validation set (n = 96) and internal test set (n = 48) were 0.90 (95% CI: 0.90 - 0.91) and 0.90 (95% CI: 0.89 - 0.91), respectively. The predicted CSA is highly correlated with the ground-truth CSA in both validation (r = 0.99, p < 0.0001) and test sets (r = 0.96, p < 0.0001). In the external test set (n = 377), 96.2% of the SM segmentations were deemed acceptable by consensus expert review. Predicted SMA and SMI values were highly correlated with the ground-truth values, with Pearson r ß 0.99 (p < 0.0001) for both the female and male patients in all datasets. Sarcopenia was associated with worse OS (HR 2.05 [95% CI 1.04 - 4.04], p = 0.04) and longer PEG tube duration (median 162 days vs. 134 days, HR 1.51 [95% CI 1.12 - 2.08], p = 0.006 in multivariate analysis. Conclusion: We developed and externally validated a fully-automated platform that strongly correlates with imaging-assessed sarcopenia in patients with H&N cancer that correlates with survival and toxicity outcomes. This study constitutes a significant stride towards the integration of sarcopenia assessment into decision-making for individuals diagnosed with HNSCC. SUMMARY STATEMENT: In this study, we developed and externally validated a deep learning model to investigate the impact of sarcopenia, defined as the loss of skeletal muscle mass, on patients with head and neck squamous cell carcinoma (HNSCC) undergoing radiotherapy. We demonstrated an efficient, fullyautomated deep learning pipeline that can accurately segment C3 skeletal muscle area, calculate cross-sectional area, and derive a skeletal muscle index to diagnose sarcopenia from a standard of care CT scan. In multi-institutional data, we found that pre-treatment sarcopenia was associated with significantly reduced overall survival and an increased risk of adverse events. Given the increased vulnerability of patients with HNSCC, the assessment of sarcopenia prior to radiotherapy may aid in informed treatment decision-making and serve as a predictive marker for the necessity of early supportive measures.

Imaging and Biomarker Surveillance for Head and Neck Squamous Cell Carcinoma: A Systematic Review and AUC Document.

Surveillance for survivors of head and neck cancer (HNC) is focused on early detection of recurrent or second primary malignancies. After initial restaging confirms disease-free status, the use of surveillance imaging for asymptomatic patients with HNC is controversial. Our objective was to comprehensively review literature pertaining to imaging and biomarker surveillance of asymptomatic patients treated for head and neck squamous cell carcinoma and to convene a multidisciplinary expert panel to provide appropriate use criteria for surveillance in representative clinical scenarios. The evidence base for the appropriate use criteria was gathered through a librarian-mediated search of literature published from 1990 to 2022 focused on surveillance imaging and circulating tumor-specific DNA for nonmetastatic head and neck squamous cell carcinoma using MEDLINE (Ovid), Embase, Web of Science Core Collection, and the Cochrane Central Register of Controlled Trials. The systematic review was reported according to PRISMA guidelines. Using the modified Delphi process, the expert panel voted on appropriate use criteria, providing recommendations for appropriate use of surveillance imaging and human papillomavirus (HPV) circulating tumor DNA. Of 5178 studies identified, 80 met inclusion criteria (5 meta-analyses/systematic reviews, 1 randomized control trial, 1 post hoc analysis, 25 prospective, and 48 retrospective cohort studies [with ≥50 patients]), reporting on 27,525 patients. No large, randomized, prospective trials examined whether asymptomatic patients who receive surveillance imaging or HPV circulating tumor DNA monitoring benefit from earlier detection of recurrence or second primary tumors in terms of disease-specific or quality-of-life outcomes. In the absence of prospective data, surveillance imaging for HNC survivors should rely on individualized recurrence-risk assessment accounting for initial disease staging, HPV disease status, and tobacco use history. There is an emerging surveillance role for circulating tumor biomarkers.

Association of Pretreatment Circulating Tumor Tissue-Modified Viral HPV DNA With Clinicopathologic Factors in HPV-Positive Oropharyngeal Cancer.

Importance: Circulating tumor tissue-modified viral (TTMV) human papillomavirus (HPV) DNA is a dynamic, clinically relevant biomarker for HPV-positive oropharyngeal squamous cell carcinoma. Reasons for its wide pretreatment interpatient variability are not well understood. Objective: To characterize clinicopathologic factors associated with TTMV HPV DNA. Design, Setting, and Participants: This cross-sectional study included patients evaluated for HPV-positive oropharyngeal squamous cell carcinoma at Dana-Farber Cancer Institute in Boston, Massachusetts, between December 2019 and January 2022 and who were undergoing curative-intent treatment. Exposures: Clinicopathologic characteristics including demographic variables, tumor and nodal staging, HPV genotype, and imaging findings. Main Outcomes and Measures: Pretreatment circulating TTMV HPV DNA from 5 genotypes (16, 18, 31, 33, and 35) assessed using a commercially available digital droplet polymerase chain reaction-based assay, considered as either detectable/undetectable or a continuous score (fragments/mL). Results: Among 110 included patients, 96 were men (87%) and 104 were White (95%), with a mean (SD) age of 62.2 (9.4) years. Circulating TTMV HPV DNA was detected in 98 patients (89%), with a median (IQR) score of 315 (47-2686) fragments/mL (range, 0-60 061 fragments/mL). Most detectable TTMV HPV DNA was genotype 16 (n = 86 [88%]), while 12 patients (12%) harbored other genotypes. Circulating TTMV HPV DNA detection was most strongly associated with clinical N stage. Although few patients had clinical stage N0 disease, only 4 of these 11 patients (36%) had detectable DNA compared with 94 of 99 patients (95%) with clinical stage N1 to N3 disease (proportion difference, 59%; 95% CI, 30%-87%). Among patients with undetectable TTMV HPV DNA, more than half (7 of 12 [58%]) had clinical stage N0 disease. The TTMV HPV DNA prevalence and score increased with progressively higher clinical nodal stage, diameter of largest lymph node, and higher nodal maximum standardized uptake value on positron emission tomography/computed tomography. In multivariable analysis, clinical nodal stage and nodal maximum standardized uptake value were each strongly associated with TTMV HPV DNA score. Among 27 surgically treated patients, more patients with than without lymphovascular invasion had detectable TTMV HPV DNA (12 of 12 [100%] vs 9 of 15 [60%]). Conclusions and Relevance: In this cross-sectional study, circulating TTMV HPV DNA was statistically significantly associated with nodal disease at HPV-positive OPSCC diagnosis. The few patients with undetectable levels had predominantly clinical stage N0 disease, suggesting assay sensitivity for diagnostic purposes may be lower among patients without cervical lymphadenopathy. Mechanisms underlying this association, and the use of this biomarker for surveillance of patients with undetectable baseline values, warrant further investigation.

Interventions for head and neck cancer survivors: Systematic review.

BACKGROUND: Interventions for head/neck cancer (HNC) survivors may not address their cancer-related and general health needs. METHODS: Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guided this systematic review of studies from 2000 to 2021 of interventions targeting cancer survivors treated with curative-intent, using MEDLINE, Embase, Emcare, and PsycINFO. Interventions were categorized into domains of the Quality of Cancer Survivorship Care Framework to characterize the scope and quality of interventions. RESULTS: We identified 28 studies for inclusion: 13 randomized and 15 non-randomized. Most targeted surveillance/management of physical effects (n = 24) including 13 that also targeted psychosocial effects. Four studies addressed prevention/surveillance for recurrence/new cancers, one addressed health promotion/disease prevention, and one addressed chronic medical conditions. Most studies (n = 27) had medium-high risk of bias. CONCLUSIONS: There are few high-quality studies addressing HNC survivorship. Future rigorously designed studies should address broader areas of care, including chronic disease management and health promotion/disease prevention.

Characterization of clinical outcomes after shorter course hypofractionated and standard-course radiotherapy for stage I-III curatively-treated Merkel cell carcinoma.

BACKGROUND: Limited data exists regarding the efficacy of curative hypofractionated radiotherapy (hypo-RT) regimens compared to conventionally-fractionated radiotherapy (conv-RT) for Merkel cell carcinoma (MCC). METHODS: A retrospective analysis of 241 patients diagnosed with non-metastatic MCC from 2005-2021 and who received RT at Dana-Farber/Brigham & Women's Cancer Center. The primary outcome was cumulative incidence of in-field locoregional relapse using Gray's test with competing risks of death and isolated out-of-field recurrence. Secondary outcomes included overall survival (OS) and MCC-specific survival using log-rank tests, and risk factors of recurrence using Cox-proportional hazards regression. RESULTS: There were 50 (20.6 %) and 193 (79.4 %) courses of hypo-RT and conv-RT, respectively. The hypo-RT cohort was older (≥73 years at diagnosis: 78.0 % vs 41.5 %, p < 0.01), and received a lower equivalent total RT dose in 2 Gy per fraction (<50 Gy: 58.0 % vs 5.2 %, p < 0.01). Median follow-up was 65.1 months (range: 1.2-194.5) for conv-RT and 25.0 months (range: 1.6-131.3) for hypo-RT cohorts. Two-year cumulative incidence of in-field locoregional relapse was low in both groups (1.1 % conv-RT vs 4.1 % hypo-RT, p = 0.114). While two-year OS was lower for the hypo-RT group (62.6 % vs 84.4 %, p = 0.0008), two-year MCC-specific survival was similar (84.7 % vs 86.6 %, p = 0.743). On multivariable analysis, immunosuppression, clinical stage III disease, and lymphovascular invasion were associated with any-recurrence when controlling for sex, age, and hypo-RT. CONCLUSIONS AND RELEVANCE: There was no difference in cumulative incidence of in-field locoregional relapse or MCC-specific survival between hypo-RT and conv-RT. Prospective studies are needed to confirm hypo-RT as an efficacious treatment option for MCC.

Adjuvant radiation following clear margin resection of high T-stage cutaneous squamous cell carcinoma halves the risk of local and locoregional recurrence: A dual-center retrospective study.

BACKGROUND: Although adjuvant radiation (ART) following clear margin surgery is recommended for select high-risk cutaneous squamous cell carcinomas, efficacy data are limited. OBJECTIVE: To evaluate the impact of ART on outcomes following clear margin surgery for high T-stage cutaneous squamous cell carcinomas. METHODS: A 20-year retrospective cohort study at 2 academic centers of high T-stage cutaneous squamous cell carcinomas (Brigham and Women's Hospital T2b or T3) with negative histologic margins post resection. Local recurrence (LR) and locoregional recurrence (LRR) were compared by whether tumors received ART or observation. RESULTS: A total of 508 tumors were included, of which 96 underwent ART (ART+). ART+ had a lower 5-year cumulative incidence of LR (ART+, 3.6% [95% CI, 1.6%-7.7%] vs ART-, 8.7% [95% CI, 6.3%-12.0%]) and LRR (ART+, 7.5% [95% CI, 4.4%-11.9%] vs ART-, 15.3% [95% CI, 11.9%-22.1%]). Recurrent tumors ≥6 cm or Brigham and Women's Hospital T3 tumors were classified as high-risk due to a higher 5-year cumulative incidence of LRR (High-risk, 26.3% [95% CI, 19.0%-35.7%]). High-risk tumors treated with ART had a lower 5-year cumulative incidence of LRR (ART+, 17.2% [95% CI, 11.9%-26.4%] vs ART-, 31.0% [95% CI, 26.1%-40.8%]). LIMITATIONS: Retrospective design, heterogeneous population, variations in radiation protocols. CONCLUSION: ART following clear margin surgery for high T-stage cutaneous squamous cell carcinomas resulted in half the risk of LR and LRR.

Retreatment of Recurrent or Second Primary Head and Neck Cancer After Prior Radiation: Executive Summary of the American Radium Society Appropriate Use Criteria.

Retreatment of recurrent or second primary head and neck cancers occurring in a previously irradiated field is complex. Few guidelines exist to support practice. We performed an updated literature search of peer-reviewed journals in a systematic fashion. Search terms, key questions, and associated clinical case variants were formed by panel consensus. The literature search informed the committee during a blinded vote on the appropriateness of treatment options via the modified Delphi method. The final number of citations retained for review was 274. These informed 5 key questions, which focused on patient selection, adjuvant reirradiation, definitive reirradiation, stereotactic body radiation, and reirradiation to treat nonsquamous cancer. Results of the consensus voting are presented along with discussion of the most current evidence. This provides updated evidence-based recommendations and guidelines for the retreatment of recurrent or second primary cancer of the head and neck.

Association between radiation dose to organs at risk and acute patient reported outcome during radiation treatment for head and neck cancers.

BACKGROUND: Associations between patient-reported outcomes and dose to organs at risk (OARs) may promote management and guide future investigations. METHODS: We retrospectively evaluated PROs and OAR dose in head and neck (H&N) cancer. RESULTS: In 169 patients, we identified weak associations between: "Difficulty swallowing/chewing" and increased mean RT dose to the oral cavity, larynx, pharyngeal constrictor muscles (PCM) and contralateral parotid; "choking/coughing" and larynx mean dose; "problems with mucus in mouth and throat" and oral cavity, contralateral parotid mean dose and parotid V30, contralateral submandibular gland and PCM mean dose; "difficulty with voice/speech" and oral cavity, contralateral parotid, contralateral submandibular gland and larynx mean dose; and "dry mouth" and ipsilateral submandibular gland, oral cavity and PCM mean dose. CONCLUSION: We identified weak associations between PRO and dose to OARs-these data can guide on treatment management, patient counseling, and serve as a baseline for future investigations.

Neoadjuvant and Adjuvant Nivolumab and Lirilumab in Patients with Recurrent, Resectable Squamous Cell Carcinoma of the Head and Neck.

PURPOSE: Surgery often represents the best chance for disease control in locoregionally recurrent squamous cell carcinoma of the head and neck (SCCHN). We investigated dual immune-checkpoint inhibition [anti-PD-1, nivolumab (N), and anti-KIR, lirilumab (L)] before and after salvage surgery to improve disease-free survival (DFS). PATIENTS AND METHODS: In this phase II study, patients received N (240 mg) + L (240 mg) 7 to 21 days before surgery, followed by six cycles of adjuvant N + L. Primary endpoint was 1-year DFS; secondary endpoints were safety, pre-op radiologic response, and overall survival (OS). Correlatives included tumor sequencing, PD-L1 scoring, and immunoprofiling. RESULTS: Among 28 patients, the median age was 66, 86% were smokers; primary site: 9 oral cavity, 9 oropharynx, and 10 larynx/hypopharynx; 96% had prior radiation. There were no delays to surgery. Grade 3+ adverse events: 11%. At the time of surgery, 96% had stable disease radiologically, one had progression. Pathologic response to N + L was observed in 43% (12/28): 4/28 (14%) major (tumor viability, TV ≤ 10%) and 8/28 (29%) partial (TV ≤ 50%). PD-L1 combined positive score (CPS) at surgery was similar regardless of pathologic response (P = 0.71). Thirteen (46%) recurred (loco-regional = 10, distant = 3). Five of 28 (18%) had positive margins, 4 later recurred. At median follow-up of 22.8 months, 1-year DFS was 55.2% (95% CI, 34.8-71.7) and 1-year OS was 85.7% (95% CI, 66.3-94.4). Two-year DFS and OS were 64% and 80% among pathologic responders. CONCLUSIONS: (Neo)adjuvant N + L was well tolerated, with a 43% pathologic response rate. We observed favorable DFS and excellent 2-year OS among high-risk, previously treated patients exhibiting a pathologic response. Further evaluation of this strategy is warranted.See related commentary by Sacco and Cohen, p. 435.

Oligometastatic adenoid cystic carcinoma: Correlating tumor burden and time to treatment with outcomes.

BACKGROUND: There is limited information on the management and outcomes of oligometastases (OM) in adenoid cystic carcinoma (ACC). METHODS: Retrospective study of 42 patients with metastatic ACC of the head and neck. Imaging studies were analyzed to identify patients with OM (1-5 lesions) at any point during follow-up. RESULTS: There was radiographic evidence of OM in 33/42 (79%) patients. Eighteen patients had OM when treated for metastases, with median overall survival (OS) of 36.0 versus 9.2 years for patients with polymetastases (6+ lesions, HR 0.38, 95%CI 0.14-0.89). Earlier locally ablative treatment, but not systemic treatment, of patients with OM predicted improved survival 3 years after metastasis (HR 0.15, 95%CI 0.02-0.63) and postponed systemic treatment by 80 more months (HR 0.22, 95%CI 0.07-0.71). CONCLUSIONS: There is a considerable population of ACC patients with detectable oligometastases, and early locally ablative treatment of oligometastases may be associated with improved outcomes.

Epidemiologic distinctions between base of tongue and tonsil oropharyngeal carcinomas.

BACKGROUND: Human papillomavirus (HPV)-positive tonsil and base of tongue (BOT) cancers have been considered together. However, important differences may exist. METHODS: Demographic and tumor characteristics, and survival, were compared by oropharyngeal cancer subsite from 2004 to 2016 in the National Cancer Database (NCDB). Trends in tonsillectomy from the National Health and Nutrition Examination Survey (NHANES) 2011-2014 were examined. RESULTS: HPV-positive BOT (N = 13 081) were older than HPV-positive tonsil patients (N = 16 874; mean 61.5 vs. 58.4 years, p < 0.001), and individuals 70+ years were significantly more likely to have BOT tumors compared with individuals <50 (adjusted odd ratio [aOR] = 2.9, 95% confidence interval = 2.6-3.2). BOT patients were also more likely to be white, male, and have advanced tumor classification. Among 7418 NHANES participants, tonsillectomy was associated with older age and white race. CONCLUSIONS: There are epidemiologic and tumor-related differences among HPV-positive tonsil and BOT carcinomas. Demographic differences may be attributable to tonsillectomy trends.

Head and Neck Cancer Clinical Research on ClinicalTrials.gov: An Opportunity for Radiation Oncologists.

PURPOSE: Many improvements in head and neck cancer (HNC) outcomes are related to optimization of radiation therapy (RT) dose, fractionation, normal-tissue sparing, and technology. However, prior work has shown that the literature of randomized controlled trials is dominated by industry-sponsored trials that have lower rates of incorporating RT. We characterized HNC clinical trials, hypothesizing that RT-specific research questions may be relatively underrepresented among HNC randomized controlled trials. METHODS AND MATERIALS: A web query of all open interventional trials on www.ClinicalTrials.gov was performed using search terms "head and neck cancer" and specific HNC subsites. Trial details were captured including the modality used, principal investigator (PI) specialty, funding, and whether the study tested a RT-modality specific hypothesis. Chi-square testing and logistic regression were used to compare groups. RESULTS: There were 841 open HNC trials, including definitive (47.6%) and recurrent/metastatic (41.9%) populations. Most trials (71.7%) were phase I or nonrandomized phase II studies, rather than phase III or randomized phase II (28.3%). Among single-arm studies, most (79.6%) incorporated systemic therapy (ST), and fewer (25.2%) incorporated RT. Even fewer phase III and randomized phase II trials tested an RT-specific hypothesis (11.1%), compared with ST-related hypotheses (77.1%; P < .001); trials were more likely to test an RT-hypothesis if the study PI was a radiation oncologist (20.9% vs 6.0%; P < .001). Among RT trials, most early-phase studies tested novel modalities (eg, stereotactic body radiation therapy, proton therapy), whereas most later-phase studies tested dose and fractionation. RT-focused trials had low rates of federal (10.4%) or industry (2.6%) funding. CONCLUSIONS: RT-specific research hypotheses are a minority of phase II-III HNC trials, which mostly focus on incorporating ST in the definitive or recurrent/metastatic setting and have higher rates of industry funding. Radiation oncologist PI leadership and increased nonindustry funding access may ensure that RT-specific hypotheses are incorporated into trial design.

A Randomized Phase 2 Study of Pembrolizumab With or Without Radiation in Patients With Recurrent or Metastatic Adenoid Cystic Carcinoma.

PURPOSE: We evaluated the safety and efficacy of pembrolizumab (pembro) ± radiation therapy (RT) in a phase 2 study among patients with progressive, metastatic adenoid cystic carcinoma (ACC). METHODS AND MATERIALS: Eligible patients had metastatic ACC with progression within the last year and ≥1 measurable lesion. Patients were randomized to pembro alone or with RT to 30 Gy in 5 fractions (pembroRT). The primary endpoint was objective response rate outside the RT field. Secondary endpoints included progression-free survival (PFS), overall survival (OS), and local RT responses. RESULTS: We randomized 20 patients (10 per arm) from 2017 to 2018. We did not observe objective response outside of the radiation treatment field; stable disease (SD) was the best response in 12 (60%) patients and was not different per arm (7 pembro, 5 pembroRT, P = .65). A tumor growth rate decrease (TGR) of >25% was noted among 7 of 12 patients and >75% in 4 patients. There were local responses in the irradiated field among all evaluable pembroRT patients. Median PFS and OS were 4.5/not reached for pembroRT and 6.6 / 27.2 months for pembro patients. One patient developed grade 3 liver enzyme elevation after 27 cycles of therapy. Correlative analyses confirm low levels of programmed death-ligand 1 expression (PD-L1), and CD8 infiltrating T-cells. We identified associations between local response and both MYB/NFIB translocation and PD-L1 expression and between changes in systemic immune populations and RT. CONCLUSIONS: Pembrolizumab and pembroRT were well tolerated. We observed no objective responses, but 60% of patients with PD before the study achieved SD, the majority with decreased TGR and half (n = 10) with clinical benefit (SD >6 months). We observed favorable local responses within the RT field. Additional strategies are needed to further delay progression and effect response.