Disease and toxicity outcomes for a modern cohort of patients with squamous cell carcinoma of cutaneous origin involving the parotid gland: Comparison of volumetric modulated arc therapy and pencil beam scanning proton therapy.

OBJECTIVES: We sought to describe outcomes for locally advanced cutaneous squamous cell carcinoma (SCC) involving the parotid treated with volumetric modulated arc therapy (VMAT) versus pencil beam scanning proton beam therapy (PBT). MATERIALS AND METHODS: Patients were gathered from 2016 to 2022 from 5 sites of a large academic RT department; included patients were treated with RT and had parotid involvement by: direct extension of a cutaneous primary, parotid regional spread from a previously or contemporaneously resected but geographically separate cutaneous primary, or else primary parotid SCC (with a cutaneous primary ostensibly occult). Acute toxicities were provider-reported (CTCAE v5.0) and graded at each on treatment visit. Statistical analyses were conducted. RESULTS: Median follow-up was 12.9 months (1.3 - 72.8); 67 patients were included. Positive margins/extranodal extension were present in 34 cases; gross disease in 17. RT types: 39 (58.2 %) VMAT and 28 (41.8 %) PBT. Concurrent systemic therapy was delivered in 10 (14.9 %) patients. There were 17 treatment failures (25.4 %), median time of 168 days. Pathologically positive neck nodes were associated with locoregional recurrence (p = 0.015). Oral cavity, pharyngeal constrictor, and contralateral parotid doses were all significantly lower for PBT. Median weight change was -3.8 kg (-14.1 - 5.1) for VMAT and -3 kg (-16.8 - 3) for PBT (p = 0.013). Lower rates of ≥ grade 1 xerostomia (p = 0.002) and ≥ grade 1 dysguesia (p < 0.001) were demonstrated with PBT. CONCLUSIONS: Cutaneous SCC involving the parotid can be an aggressive clinical entity despite modern multimodal therapy. PBT offers significantly lower dose to organs at risk compared to VMAT, which seemingly yields diminished acute toxicities.

Assessment of Barriers and Challenges to Screening, Diagnosis, and Biomarker Testing in Early-Stage Lung Cancer.

Management of lung cancer has transformed over the past decade and is no longer considered a singular disease as it now has multiple sub-classifications based on molecular markers. The current treatment paradigm requires a multidisciplinary approach. One of the most important facets of lung cancer outcomes however relies on early detection. Early detection has become crucial, and recent effects have shown success in lung cancer screening programs and early detection. In this narrative review, we evaluate low-dose computed tomography (LDCT) screening and how this screening modality may be underutilized. The barriers to broader implementation of LDCT screening is also explored as well as approaches to address these barriers. Current developments in diagnosis, biomarkers, and molecular testing in early-stage lung cancer are evaluated as well. Improving approaches to screening and early detection can ultimately lead to improved outcomes for patients with lung cancer.

Utilization of model-agnostic explainable artificial intelligence frameworks in oncology: a narrative review.

Background and Objective: Machine learning (ML) models are increasingly being utilized in oncology research for use in the clinic. However, while more complicated models may provide improvements in predictive or prognostic power, a hurdle to their adoption are limits of model interpretability, wherein the inner workings can be perceived as a "black box". Explainable artificial intelligence (XAI) frameworks including Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) are novel, model-agnostic approaches that aim to provide insight into the inner workings of the "black box" by producing quantitative visualizations of how model predictions are calculated. In doing so, XAI can transform complicated ML models into easily understandable charts and interpretable sets of rules, which can give providers with an intuitive understanding of the knowledge generated, thus facilitating the deployment of such models in routine clinical workflows. Methods: We performed a comprehensive, non-systematic review of the latest literature to define use cases of model-agnostic XAI frameworks in oncologic research. The examined database was PubMed/MEDLINE. The last search was run on May 1, 2022. Key Content and Findings: In this review, we identified several fields in oncology research where ML models and XAI were utilized to improve interpretability, including prognostication, diagnosis, radiomics, pathology, treatment selection, radiation treatment workflows, and epidemiology. Within these fields, XAI facilitates determination of feature importance in the overall model, visualization of relationships and/or interactions, evaluation of how individual predictions are produced, feature selection, identification of prognostic and/or predictive thresholds, and overall confidence in the models, among other benefits. These examples provide a basis for future work to expand on, which can facilitate adoption in the clinic when the complexity of such modeling would otherwise be prohibitive. Conclusions: Model-agnostic XAI frameworks offer an intuitive and effective means of describing oncology ML models, with applications including prognostication and determination of optimal treatment regimens. Using such frameworks presents an opportunity to improve understanding of ML models, which is a critical step to their adoption in the clinic.

The prostate tissue-based telomere biomarker as a prognostic tool for metastasis and death from prostate cancer after prostatectomy.

Current biomarkers are inadequate prognostic predictors in localized prostate cancer making treatment decision-making challenging. Previously, we observed that the combination of more variable telomere length among prostate cancer cells and shorter telomere length in prostate cancer-associated stromal cells - the telomere biomarker - is strongly associated with progression to metastasis and prostate cancer death after prostatectomy independent of currently used pathologic indicators. Here, we optimized our method allowing for semi-automated telomere length determination in single cells in fixed tissue, and tested the telomere biomarker in five cohort studies of men surgically treated for clinically localized disease (N = 2,255). We estimated the relative risk (RR) of progression to metastasis (N = 311) and prostate cancer death (N = 85) using models appropriate to each study's design adjusting for age, prostatectomy stage, and tumor grade, which then we meta-analyzed using inverse variance weights. Compared with men who had less variable telomere length among prostate cancer cells and longer telomere length in prostate cancer-associated stromal cells, men with the combination of more variable and shorter telomere length had 3.76 times the risk of prostate cancer death (95% confidence interval [CI] 1.37-10.3, p = 0.01) and had 2.23 times the risk of progression to metastasis (95% CI 0.99-5.02, p = 0.05). The telomere biomarker was associated with prostate cancer death in men with intermediate risk disease (grade groups 2/3: RR = 9.18, 95% CI 1.14-74.0, p = 0.037) and with PTEN protein intact tumors (RR = 6.74, 95% CI 1.46-37.6, p = 0.015). In summary, the telomere biomarker is robust and associated with poor outcome independent of current pathologic indicators in surgically treated men.

Machine learning to refine prognostic and predictive nodal burden thresholds for post-operative radiotherapy in completely resected stage III-N2 non-small cell lung cancer.

BACKGROUND: The role of post-operative radiotherapy (PORT) for completely resected N2 non-small-cell lung cancer (NSCLC) is controversial in light of recent randomized data. We sought to utilize machine learning to identify a subset of patients who may still benefit from PORT based on extent of nodal involvement. MATERIALS/METHODS: Patients with completely resected N2 NSCLC were identified in the National Cancer Database. We trained a machine-learning based model of overall survival (OS). SHapley Additive exPlanation (SHAP) values were used to identify prognostic and predictive thresholds of number of positive lymph nodes (LNs) involved and lymph node ratio (LNR). Cox proportional hazards regression was used for confirmatory analysis. RESULTS: A total of 16,789 patients with completely resected N2 NSCLC were identified. Using the SHAP values, we identified thresholds of 3+ positive LNs and a LNR of 0.34+. On multivariate analysis, PORT was not significantly associated with OS (p = 0.111). However, on subset analysis of patients with 3+ positive LNs, PORT improved OS (HR: 0.91; 95% CI: 0.86-0.97; p = 0.002). On a separate subset analysis in patients with a LNR of 0.34+, PORT improved OS (HR: 0.90; 95% CI: 0.85-0.96; p = 0.001). Patients with 3+ positive lymph nodes had a 5-year OS of 38% with PORT compared to 31% without PORT. Patient with positive lymph node ratio 0.34+ had a 5-year OS of 38% with PORT compared to 29% without PORT. CONCLUSIONS: Patients with a high lymph node burden or lymph node ratio may present a subpopulation of patients who could benefit from PORT. To our knowledge, this is the first study to use machine learning algorithms to address this question with a large national dataset. These findings address an important question in the field of thoracic oncology and warrant further investigation in prospective studies.

Obesity is Associated with Shorter Telomere Length in Prostate Stromal Cells in Men with Aggressive Prostate Cancer.

In our prior studies, obesity was associated with shorter telomeres in prostate cancer-associated stromal (CAS) cells, and shorter CAS telomeres were associated with an increased risk of prostate cancer death. To determine whether the association between obesity and shorter CAS telomeres is replicable, we conducted a pooled analysis of 790 men who were surgically treated for prostate cancer, whose tissue samples were arrayed on five tissue microarray (TMA) sets. Telomere signal was measured using a quantitative telomere-specific FISH assay and normalized to 4',6-diamidino-2-phenylindole for 351 CAS cells (mean) per man; men were assigned their median value. Weight and height at surgery, collected via questionnaire or medical record, were used to calculate body mass index (BMI; kg/m2) and categorize men as normal (<25), overweight (25 ≤ BMI < 30), or obese (≥30). Analyses were stratified by grade and stage. Men were divided into tertiles of TMA- (overall) or TMA- and disease aggressiveness- (stratified) specific distributions; short CAS telomere status was defined by the bottom two tertiles. We used generalized linear mixed models to estimate the association between obesity and short CAS telomeres, adjusting for age, race, TMA set, pathologic stage, and grade. Obesity was not associated with short CAS telomeres overall, or among men with nonaggressive disease. Among men with aggressive disease (Gleason≥4+3 and stage>T2), obese men had a 3-fold increased odds of short CAS telomeres (OR: 3.06; 95% confidence interval: 1.07-8.75; P trend = 0.045) when compared with normal weight men. Telomere shortening in prostate stromal cells may be one mechanism through which lifestyle influences lethal prostate carcinogenesis. PREVENTION RELEVANCE: This study investigates a potential mechanism underlying the association between obesity and prostate cancer death. Among men with aggressive prostate cancer, obesity was associated with shorter telomeres prostate cancer associated stromal cells, and shorter CAS telomeres have been associated with an increased risk of prostate cancer death.

Racial Difference in Prostate Cancer Cell Telomere Lengths in Men with Higher Grade Prostate Cancer: A Clue to the Racial Disparity in Prostate Cancer Outcomes.

BACKGROUND: Black men have worse prostate cancer outcomes following treatment than White men even when accounting for prognostic factors. However, biological explanations for this racial disparity have not been fully identified. We previously showed that more variable telomere lengths among cancer cells and shorter telomere lengths in cancer-associated stromal (CAS) cells individually and together ("telomere biomarker") are associated with prostate cancer-related death in surgically treated men independent of currently used prognostic indicators. Here, we hypothesize that Black-White differences in the telomere biomarker and/or in its components may help explain the racial disparity in prostate cancer outcomes. METHODS: Black [higher grade (Gleason ≥4+3) = 34 and lower grade = 93] and White (higher grade = 34 and lower grade = 89) surgically treated men were frequency matched on age, pathologic stage, and grade. We measured telomere lengths in cancer and CAS cells using a robust telomere-specific FISH assay. Tissue microarray and grade-specific distributional cutoff points without regard to race were evaluated. RESULTS: Among men with higher grade disease, the proportion of Black men (47.1%) with more variable cancer cell telomere lengths was 2.3-times higher (P = 0.02) than that in White men (20.6%). In contrast, among men with lower grade disease, cancer cell telomere length variability did not differ by race. The proportion of men with shorter CAS cell telomeres did not differ by race for either higher or lower grade disease. CONCLUSIONS: A greater proportion of Black men with higher grade disease have an adverse prostate cancer cell telomere phenotype than White men with higher grade disease. IMPACT: Our findings suggest a possible explanation for the racial disparity in prostate cancer outcomes.