Sujith Baliga

Background: Head and neck squamous cell carcinomas (HNSCCs) cause considerable morbidity and mortality. Multimodal treatment strategies can cause significant toxicity, and therapy options are limited for recurrent disease. Immunotherapy has emerged as a promising approach. However, patient response variability underscores the need for better predictive markers. Objective: This study aims to use artificial intelligence to develop two predictive models in patients with HNSCC to assess (1) progression or recurrence following primary curative treatment and (2) long-term survival after immunotherapy schemes in recurrent and metastatic disease. This study will also describe the characteristics of patients with early, locally advanced, and recurrent or metastatic cancers. Methods: This is a retrospective, observational study of data captured in electronic health records (EHRs) from participating hospitals between January 1, 2014, and December 31, 2021. This study’s population comprises adults diagnosed with HNSCC at any stage. Study variables, including demographics, comorbidities, clinical variables, treatments, and outcomes, will be extracted using EHRead, a technology that applies natural language processing and machine learning to extract and analyze structured and unstructured clinical information in deidentified EHRs. Predictive models based on dynamic risk stratification for treatment response and progression or recurrence will be developed using multivariable logistic regressions, decision tree classifiers, and random forest approaches. Descriptive and outcome analyses will be shown for different anatomic subsites and stratified by stage and treatment. Results: This study began enrolling sites in July 2021 and is currently ongoing. By December 2025, data from 10 centers has been collected, comprising a total of 151,934,990 EHRs from 2,159,719 patients. Conclusions: Development of predictive models using artificial intelligence will advance clinical understanding of HNSCC to improve patient outcomes.

PURPOSE: Systemic inflammation is an important factor for carcinogenesis and tumor progression. Systemic Inflammation Response Index (SIRI) was suggested as one of the peripheral blood biomarkers to assess the magnitude of inflammation. Herein, we aimed to evaluate the prognostic value of SIRI in head and neck cancer. METHODS: A single-institution database at a comprehensive cancer center was queried for patients with nonmetastatic head and neck cancer who underwent definitive chemoradiation between December 2011 and February 2024 with SIRI available for analysis. With the top quartile as a reference to evaluate high versus low SIRI, logistic multivariable analysis (MVA) was performed to identify variables associated with high SIRI. Cox MVA was performed to evaluate survival. Interaction term analysis was performed to evaluate whether there is a heterogeneous association of SIRI with outcomes stratified by p16 status. RESULTS: A total of 638 patients met our criteria. The median follow-up was 45.9 months (95% CI, 44.4 to 47.7). Using the top quartile cutoff of 3.22 for SIRI, those who were male, underweight, had poor performance status, and had advanced T staging were more likely to have high SIRI, whereas those with other racial background, p16-positive, and larynx primary site were less likely to have high SIRI. Those with high SIRI had worse overall survival (OS; adjusted hazard ratio [aHR], 1.59 [95% CI, 1.15 to 2.25], P = .008), but not PFS (aHR, 1.18 [95% CI, 0.86 to 1.61], P = .320). Interaction term analysis did not reach statistical significance ( P _interaction = .75 for OS, 0.84 for PFS). CONCLUSION: Elevated SIRI was an independent, adverse prognostic factor for survival without heterogeneous association with outcomes based on p16 status. Further validation of SIRI would be warranted in future studies.

Background/Objectives: Treatment delays have been shown to be associated with overall survival (OS) in head and neck squamous cell carcinomas (HNSCCs). Given the slow tumor growth kinetics of adenoid cystic carcinoma (ACC), it is unclear if delays have a similar impact in this tumor histology. Methods: We queried the National Cancer Database for patients diagnosed with non-metastatic ACC between the years 2004 and 2019 and treated with surgery followed by RT. A multivariable Cox regression model was used to examine the associations between the time from diagnosis to surgery, the duration of RT, and OS. Results: A total of 1449 patients were included for analysis. Increased time from diagnosis to surgery (HR: 1.02, 95% CI: 1.01–1.03, p < 0.001) and duration of RT (HR: 1.14, 95% CI: 1.04–1.25, p = 0.004) were associated with worse survival on UVA, while time from surgery to RT start was not (p = 0.647). Increased duration of RT (aHR: 1.13, 95% CI: 1.03–1.24, p = 0.012) remained significantly associated with OS on multivariable analysis, while time from diagnosis to surgery (aHR: 1.00, 95% CI: 0.98–1.02, p = 0.979) did not. Conclusions: Delays in treatment initiation and in the interval from surgery to radiation did not result in clinically significant differences in survival in this analysis, while prolonged duration of radiation therapy was significantly associated with worse survival. These findings are hypothesis-generating and suggest that treatment delays for ACC may have different effects on oncologic outcomes than those for HNSCC; however, prospective data is paramount to verify these results before strong conclusions can be made.

PURPOSE: National Comprehensive Cancer Network guidelines recommend weekly cisplatin and weekly carboplatin with paclitaxel as alternative concurrent systemic therapy regimens among patients with head and neck cancer undergoing definitive chemoradiation. However, there is a paucity of literature comparing outcomes with these two regimens. We performed an observational cohort study of patients with head and neck cancer treated with definitive chemoradiation receiving either cisplatin or carboplatin with paclitaxel. METHODS: A single-institution database was queried for patients with head and neck cancer diagnosed between October 2011 and December 2023 who underwent chemoradiation with either once a week cisplatin (40 mg/m ² ) or once a week carboplatin (AUC 1) with paclitaxel (30 mg/m ² ). Cox multivariable analysis (MVA), Kaplan-Meier, Fine-Gray MVA, logistic MVA, and propensity score matching were performed to evaluate treatment outcomes. RESULTS: A total of 488 patients were identified. Median follow-up was 43.1 months (95% CI, 41.0 to 45.4). The majority of patients received at least five cycles. Those with older age, former smoking history, and worse performance status were more likely to undergo carboplatin and paclitaxel. Carboplatin and paclitaxel were associated with no statistically significant overall survival (adjusted hazards ratio [aHR], 1.12 [95% CI, 0.69 to 1.82]; P = .64), progression-free survival (aHR, 1.45 [95% CI, 0.96 to 2.19]; P = .08), locoregional failure (aHR, 1.91 [95% CI, 0.85 to 4.32]; P = .12), and distant failure (aHR, 1.33 [95% CI, 0.70 to 2.54]; P = .39) compared with cisplatin. Similar findings were noted on 109 matched pairs as well as subgroups stratified by p16 status. CONCLUSION: Our study suggested that those treated with weekly carboplatin and paclitaxel had no statistically significant survival and tumor control outcomes compared with those with weekly cisplatin. Older age and poor performance status were independently associated with the receipt of weekly carboplatin and paclitaxel.

Circulating tumor tissue modified (TTMV)‐HPV DNA has emerged as a promising biomarker in human papillomavirus associated oropharyngeal squamous cell carcinoma (HPV‐OPSCC). The objective of this study was to assess ctHPVDNA TTMV clearance kinetics during RT and its relationship with progression in HPV‐OPSCC. We identified 80 non‐metastatic HPV‐OPSCC patients with 366 TTMV samples who underwent prospective plasma TTMV testing before, during and after curative intent RT or CRT between June 2021 and February 2023. Patients with rapid favorable clearance (>95% decline from pre‐treatment value) of TTMV were compared to those with suboptimal clearance (p = .042). Among patients who had complete clearance of their TTMV at 3 months, those who had a negative, equivocal, and incomplete PET response had a 2‐year PFS of 94.3%, 77.8%, and 59.3%, respectively (log‐rank, p = .029). Our study indicates that persistent TTMV‐HPV DNA is associated with worse PFS and may portend unfavorable outcomes. Thus, monitoring TTMV clearance kinetics could be a valuable biomarker for guiding treatment decisions in patients with HPV‐OPSCC.

Cisplatin, one of the most ototoxic anti-neoplastic agents, causes permanent hearing loss in up to 90% of patients. We assessed ototoxicity rates and prospectively collected audiologic outcomes of patients receiving low-dose or high-dose cisplatin with concurrent cochlear-sparing intensity-modulated radiation therapy (IMRT). Patients with head and neck squamous cell carcinoma (HNSCC) receiving definitive or adjuvant cisplatin-based chemoradiotherapy (CRT) were analyzed. Cisplatin was administered either in low doses weekly (40 mg/m²) for up to seven doses or in high doses triweekly (100 mg/m²) for up to three doses. Cochlear-sparing IMRT was delivered in all cases. Audiologic data were prospectively collected before, during, and after treatment completion. The primary endpoint was a hearing change grade of ≥3 after CRT completion. Of the 96 HNSCC patients evaluated, 69 received weekly cisplatin and 58 received definitive CRT. Of patients receiving weekly cisplatin, 13% developed ≥G3 ototoxicity vs. 56% of patients who received triweekly cisplatin (p < 0.001). In multivariable modeling, the cisplatin dose schedule remained significant (OR: 8.4, 95%CI: 2.8–27.8, p < 0.001) for risk of severe irreversible ototoxicity. Triweekly cisplatin CRT significantly increased the ≥G3 severe irreversible ototoxicity risk compared to low-dose weekly cisplatin, irrespective of the cumulative cisplatin dose, even with the use of cochlear-sparing IMRT. No significant difference in oncologic outcomes was observed between the two schedules.

Background: Pediatric patients with metastatic and/or recurrent solid tumors have poor survival outcomes despite standard-of-care systemic therapy. Stereotactic ablative radiation therapy (SABR) may improve tumor control. We report the outcomes with the use of SABR in our pediatric solid tumor population. Methods: This was a single-institutional study in patients < 30 years treated with SABR. The primary endpoint was local control (LC), while the secondary endpoints were progression-free survival (PFS), overall survival (OS), and toxicity. The survival analysis was performed using Kaplan–Meier estimates in R v4.2.3. Results: In total, 48 patients receiving 135 SABR courses were included. The median age was 15.6 years (interquartile range, IQR 14–23 y) and the median follow-up was 18.1 months (IQR: 7.7–29.1). The median SABR dose was 30 Gy (IQR 25–35 Gy). The most common primary histologies were Ewing sarcoma (25%), rhabdomyosarcoma (17%), osteosarcoma (13%), and central nervous system (CNS) gliomas (13%). Furthermore, 57% of patients had oligometastatic disease (≤5 lesions) at the time of SABR. The one-year LC, PFS, and OS rates were 94%, 22%, and 70%, respectively. No grade 4 or higher toxicities were observed, while the rates of any grade 1, 2, and 3 toxicities were 11.8%, 3.7%, and 4.4%, respectively. Patients with oligometastatic disease, lung, or brain metastases and those who underwent surgery for a metastatic site had a significantly longer PFS. LC at 1-year was significantly higher for patients with a sarcoma histology (95.7% vs. 86.5%, p = 0.01) and for those who received a biological equivalent dose (BED10) > 48 Gy (100% vs. 91.2%, p = 0.001). Conclusions: SABR is well tolerated in pediatric patients with 1-year local failure and OS rates of <10% and 70%, respectively. Future studies evaluating SABR in combination with systemic therapy are needed to address progression outside of the irradiated field.

Background: Salivary duct carcinomas (SDC) are a rare and aggressive subtype of salivary gland neoplasm. They can present with distinct immunoprofiles, such as androgen receptor (AR) and HER-2/Neu-positivity. To date, no consensus exists on how to best manage this entity. Methods: All patients diagnosed with nonmetastatic AR+ SDC of the parotid from 2013 to 2019 treated with curative intent were included. Immunologic tumor profiling was conducted using 24 distinct markers. Kaplan–Meier analyses were used to estimate locoregional recurrence (LRR), distant control, and overall survival (OS). Results: Fifteen patients were included. Nine (60%) patients presented with T4 disease and eight (53%) had positive ipsilateral cervical lymphadenopathy. Ten (67%) patients underwent trimodality therapy, including surgery followed by adjuvant radiation and concurrent systemic therapy. The median follow-up was 5.5 years (interquartile range, 4.8–6.1). The estimated 5-year rates of LRR, distant progression, and OS were 6%, 13%, and 87%, respectively. Conclusion: Despite only including AR+ SDC of the parotid, immunoprofiles, such as expression of HER-2, were highly variable, highlighting the potential to tailor systemic regimens based on individual histologic profiles in the future. Studies with larger patient numbers using tumor-specific molecular profiling and tumor heterogeneity analyses are justified to better understand the biology of these tumors. Molecularly informed treatment approaches, including the potential use of AR- and HER-2/Neu-directed therapies upfront in the definitive setting, may hold future promise to further improve outcomes for these patients.

Objectives: Patients with recurrent and metastatic head and neck cancer (HNC) have limited treatment options. ‘QuadShot’ (QS), a hypofractionated palliative radiotherapy regimen, can provide symptomatic relief and local control and may potentiate the effects of immune checkpoint inhibitors (ICIs). We compared outcomes of QS ± concurrent ICIs in the palliative treatment of HNC. Materials and Methods: We identified patients who received ≥three cycles of QS from 2017 to 2022 and excluded patients without post-treatment clinical evaluation or imaging. Outcomes for patients who received QS alone were compared to those treated with ICI concurrent with QS, defined as receipt of ICI within 4 weeks of QS. Results: Seventy patients were included, of whom 57% received concurrent ICI. Median age was 65.5 years (interquartile range [IQR]: 57.9–77.8), and 50% patients had received prior radiation to a median dose of 66 Gy (IQR: 60–70). Median follow-up was 8.8 months. Local control was significantly higher with concurrent ICIs (12-month: 85% vs. 63%, p = 0.038). Distant control (12-month: 56% vs. 63%, p = 0.629) and median overall survival (9.0 vs. 10.0 months, p = 0.850) were similar between the two groups. On multivariable analysis, concurrent ICI was a significant predictor of local control (HR for local failure: 0.238; 95% CI: 0.073–0.778; p = 0.018). Overall, 23% patients experienced grade 3 toxicities, which was similar between the two groups. Conclusions: The combination of QS with concurrent ICIs was well tolerated and significantly improved local control compared to QS alone. The median OS of 9.4 months compares favorably to historical controls for patients with HNC treated with QS. This approach represents a promising treatment option for patients with HNC unsuited for curative-intent treatment and warrants prospective evaluation.