Brigham And Women's Hospital Research Articles

Evaluation of prognostic prediction ability of the novel Japanese risk factor scoring system in a Japanese cohort of resectable cutaneous squamous cell carcinoma: A retrospective cross-sectional study.

Japanese patients with very high-risk cutaneous squamous cell carcinomas (cSCCs), based on the National Comprehensive Cancer Network guidelines, have been reported to display a higher cumulative incidence of relapse and disease-specific death (DSD) than those with high-risk cSCC. Therefore, prognosis prediction is crucial for Japanese patients with very high-risk cSCCs. Herein, we aimed to evaluate the prognostic prediction ability of our novel Japanese Risk Factor Scoring Systems (JARF scoring) in a Japanese cohort of cSSC patients. Data of 424 Japanese patients with resectable very high-risk cSCCs were analysed. We compared the prognostic ability of the following three staging systems: Brigham and Women's Hospital (BWH) tumour staging, number of NCCN very high-risk factors, and JARF scoring, including recurrent tumour, high-risk histological features, deep tumour invasion and lymphatic or vascular involvement as risk factors. The prognostic ability of these staging systems was evaluated according to the cumulative incidence of local recurrence (LR), regional lymph node metastasis (RLNM), DSD, and overall survival (OS). When BWH staging was used, high T stage led to significantly poor outcomes only in the cumulative incidence of RLNM (p = 0.01). The presence of very high-risk NCCN factors led to significantly poor outcomes in terms of RLNM (p = 0.03) and OS (p = 0.02). Meanwhile, a high number of risk factors in the JARF scoring system clearly led to poor outcomes in terms of LR (p = 0.01), RLNM (p < 0.01), DSD (p = 0.03), and OS (p < 0.01). The JARF scoring system may accurately predict the risk of recurrence and death in very high-risk cSCC patients in Japan.

Comparative Performance of Four Staging Classifications to Select «High-Risk» Head and Neck Cutaneous Squamous Cell Carcinomas.

Many classifications exist to select patients with "high-risk" head and neck cutaneous squamous cell carcinoma (HNCSCC). To compare the performance of the Brigham and Women's Hospital (BWH) classification with the performance of the American Joint Committee on Cancer 8th Edition (AJCC8), the Union for International Cancer Control 8th Edition (UICC8), and the National Comprehensive Cancer Network (NCCN) classifications. In this single-center retrospective study, HNCSCC resected in a tertiary care center were classified as "low-risk" or "high-risk" tumors according to the four classifications. Rates of local recurrence (LR), lymph node recurrence (NR), and disease-specific death (DSD) were collected. The performance of each classification was then calculated in terms of homogeneity, monotonicity, and discrimination and compared. Two hundred and seventeen HNCSCC from 160 patients, with a mean age of 80 years, were included. For predicting the risk of any poor outcome and risk of NR, the BWH classification had the best specificity and positive predictive value. However, its concordance index was not significantly higher than that of the AJCC8 and UICC8 classifications. The NCCN classification was the least discriminant. This study suggests that the BWH classification is the most appropriate for predicting the risk of poor outcomes in patients with HNCSCC when compared with the NCCN, UICC8, and AJCC8 classifications.

Evaluating embedded geriatric care in gastrointestinal (GI) oncology program for older adults (OA): Feasibility and key learnings.

e24015 Background: Older adults with GI cancers experience frequent interruptions to life-improving care due to emergency department (ED) or unplanned hospital visits (UHV), with estimates ranging from 34-84% for advanced disease. Geriatric assessment can predict survival, toxicity, and completion of planned therapy but has been considered cumbersome for use in oncology practice. Geriatric co-management is now standard within certain primary care and subspecialty practices at the Brigham and Women’s Hospital (BWH), and has resulted in reduced mortality at 1-year, lower incidence of delirium, and improved provider/patient satisfaction. The REACH study (Resilience and Equity in Aging, Cancer, and Health) is a type III hybrid implementation mixed methods study that evaluates the ability of geriatric consultation to improve quality of life and reduce ED/UHV for pre-frail and frail OA diagnosed with GI cancer. Methods: We reviewed early data from the REACH study based in the Gastrointestinal Cancer Center at Dana-Farber Cancer Institute (DFCI) between Jan. 2022–Jan. 2023. Data included REACH-eligible patients scheduled for a new consult visit, OA patients continuing care at DFCI, percentage of patients consented to participate, and percentage of consented patients who completed the frailty questionnaire (EFA). Results: As of 1/31/2023, a total of 815 OA patients age 70+ were screened initially eligible. Consented patients completed the REACH EFA. Among the 815 patients initially eligible, 73.8% of OA patients aged 70-74 continued care at DFCI, along with 74.7% of patients age 75-79, and 73.5% of patients age 80-84. For patients 85 or older, 82.7% of patients continued their oncology care at DFCI. Patients who screened ineligible or declined to participate were due to either not continuing care at DFCI, preferred language other than English/Spanish, or initiating hospice care. From Jan. 2022 to Jan. 2023, n = 30 consented to participate with a 90% completion rate of the full EFA questionnaire (n = 27/30). Conclusions: Data for the REACH study is ongoing. We are now transitioning to a clinical program where consent is no longer needed to complete the REACH EFA questionnaire. DFCI has pivoted to EFA questionnaire being standard of care for OA patients due to its acceptability and importance. Future program efforts will improve patient language access and expand to additional DFCI regional sites around New England to ensure a more diverse population and provide tailored support for those patients who have a higher level of frailty and more complex needs throughout their cancer care journey. Clinical trial information: NCT04674267 .

The Prognostic Value and Clinical Utility of the 40-Gene Expression Profile (40-GEP) Test in Cutaneous Squamous Cell Carcinoma: Systematic Review and Meta-Analysis.

The current tumor staging systems for cutaneous squamous cell carcinoma (cSCC) are considered inadequate and insufficient for evaluating the risk of metastasis and for identifying patients at high risk of cSCC. This meta-analysis aimed to assess the prognostic significance of a 40-gene expression profile (40-GEP) both independently and integrated with clinicopathologic risk factors and established staging systems (American Joint Committee on Cancer, eighth edition (AJCC8) and Brigham and Women's Hospital (BWH)). Electronic databases, including PubMed (MEDLINE), Embase, the Cochrane Library, and Google Scholar, were systematically searched to identify cohort studies and randomized controlled trials on evaluations of the prediction value of 40-GEP in cSCC patients up to January 2023. The metastatic risk analysis of a given 40-GEP class combined with tumor stage and/or other clinicopathologic risk factors was based upon log hazard ratios (HRs) and their standard error (SE). Heterogeneity and subgroup analyses were performed, and data quality was assessed. A total of 1019 patients from three cohort studies were included in this meta-analysis. The overall three-year metastatic-free survival rates were 92.4%, 78.9%, and 45.4% for class 1 (low risk), class 2A (Intermediate risk), and class 2B (high risk) 40-GEP, respectively, indicating a significant variation in survival rates between the risk classification groups. The pooled positive predictive value was significantly higher in class 2B when compared to AJCC8 or BWH. The subgroup analyses demonstrated significant superiority of integrating 40-GEP with clinicopathologic risk factors or AJCC8/BWH, especially for class 2B patients. The integration of 40-GEP with staging systems can improve the identification of cSCC patients at high risk of metastasis, potentially leading to improved care and outcomes, especially in the high-risk class 2B group.

Importance of external validation and subgroup analysis of artificial intelligence in the detection of low ejection fraction from electrocardiograms.

Left ventricular systolic dysfunction (LVSD) carries an increased risk for overt heart failure and mortality, yet treatable to mitigate disease progression. An artificial intelligence (AI)-enabled 12-lead electrocardiogram (ECG) model demonstrated promise in LVSD screening, but the performance dropped unexpectedly in external validation. We thus sought to train de novo models for LVSD detection and investigated their performance across multiple institutions and across a broader set of patient strata. ECGs taken within 14 days of an echocardiogram were obtained from four academic hospitals (three in the United States and one in Japan). Four AI models were trained to detect patients with ejection fraction (EF) <40% using ECGs from each of the four institutions. All the models were then evaluated on the held-out test data set from the same institution and data from the three external institutions. Subgroup analyses stratified by patient characteristics and common ECG abnormalities were performed. A total of 221 846 ECGs were identified from the 4 institutions. While the Brigham and Women's Hospital (BWH)-trained and Keio-trained models yielded similar accuracy on their internal test data [area under the receiver operating curve (AUROC) 0.913 and 0.914, respectively], external validity was worse for the Keio-trained model (AUROC: 0.905-0.915 for BWH trained and 0.849-0.877 for Keio-trained model). Although ECG abnormalities including atrial fibrillation, left bundle branch block, and paced rhythm-reduced detection, the models performed robustly across patient characteristics and other ECG features. While using the same model architecture, different data sets produced models with different performances for detecting low-EF highlighting the importance of external validation and extensive stratification analysis.

NREM Parasomnias: Retrospective Analysis of Treatment Approaches and Comorbidities.

Brief SummaryNon-rapid eye movement (NREM) parasomnias are relatively common among children and can often persist into adulthood, where they can pose as a clinical challenge:, the behaviors can result in injury or have negative impacts on functioning and quality of life, thus necessitating treatment, but the choice of treatment is complicated by the lack of evidence—based guidelines, as well as potential side effects. The aim of this retrospective analysis is to examine the most frequently used treatment strategies and pharmaceuticals for NREM parasomnias and evaluate perceived outcomes based on the patient’s subjective reports of the frequency and severity of symptoms.The aim of this retrospective analysis is to determine the most frequently prescribed medications for the treatment of NREM parasomnias and evaluate reported outcomes. We performed a retrospective chart review of all patients with NREM parasomnia diagnosed within Brigham and Women’s Hospital (BWH) clinics examining the date of diagnosis, date of starting therapy, comorbidities, type of medication prescribed, and the reported change in symptoms or side effects at follow-up visits. From 2012 to 2019, 110 patients (59 females, 51 male) at BWH clinics received a diagnosis of NREM parasomnia, including sleepwalking and night terrors. The mean age was 44. Comorbidities included obstructive sleep apnea (OSA) (46%), periodic limb movement syndrome (PLMS) (13%), insomnia (19%), Restless leg syndrome (RLS) (9%), epilepsy (4%), and REM behavior disorder (RBD) (9%). Initial treatment strategies include behavioral and safety counseling only (34%), pharmacological treatment (29%), treatment of any comorbidity (28%), and combined treatment of any of the above (9%). Improvement was reported with: treatment of OSA (n = 23 52% reported improvement), melatonin (n = 8, improvement reported by 88%.,benzodiazepine (n = 7, improvement reported by 57%). Treating comorbid conditions is a frequent treatment strategy, often associated with symptom improvement. The pharmacologic treatment most commonly included melatonin and benzodiazepines. Comprehensive management should include behavioral and safety recommendations, assessment of comorbid conditions, and individually tailored pharmaceutical treatment.

144 Clinical epidemiology of Masson tumor

Intravascular papillary endothelial hyperplasia (IPEH), also known as Masson tumor–a rare, benign vascular lesion typically presenting in the skin as a subcutaneous nodule – may be clinically mistaken for other neoplasms such as hemangiomas and lipomas. IPEH is classically categorized into three types:1) pure type arising in dilated endovascular spaces; 2) mixed type developing from preexisting vascular abnormalities; 3) extravascular type. While the prognosis of IPEH is excellent, it must be differentiated from malignant tumors such as angiosarcomas, which may require intensive treatments. Because the literature on IPEH is limited, we sought to characterize clinical and pathological features of IPEH. Subjects were identified using the Mass General Brigham (MGB) Research Patient Data Registry and included individuals with pathologically proven diagnosis of IPEH from 1/1980 to 8/2021 at Massachusetts General Hospital, Brigham and Women’s Hospital (BWH), and the BWH Faulkner Hospital. Demographic information, clinical documentation, and pathology reports were reviewed for data extraction. 261 individuals were diagnosed with IPEH, with the majority being women (60%) and white (74%). The average age at diagnosis was 53 years old [4-98 years old]. The most frequently involved anatomic sites were the upper (29%) and lower (24%) extremities. Common initial clinical diagnoses of lesions were cysts, hemangiomas, and lipomas. The pure subtype of IPEH was the most common (50%), followed by the mixed (46%) and extravascular subtypes (4%). Extravascular IPEH occurred more frequently in women (5%) compared to men (1%). We found that most clinicians’ initial impressions prior to biopsy did not include the final diagnosis of IPEH -- often using vague terms such as “soft tissue mass” -- indicating a potential need for greater awareness of this condition. Given the differential diagnosis of IPEH often includes conditions such as melanoma or angiosarcoma, clinicopathologic correlation is of utmost importance for this uncommon vascular lesion.

220 Clinical epidemiology of ecthyma gangrenosum

Ecthyma gangrenosum (EG) is an uncommon infectious cutaneous eruption. It typically presents with painless macules, transitions to hemorrhagic vesicles, and eventually evolves into necrotic ulcers. EG should be recognized early and treated aggressively as it is associated with significant morbidity. Because the literature on this disease is limited, we sought to characterize clinical and pathologic features of EG in a large, multi-institution academic center. Patients were identified using the Mass General Brigham Research Patient Data Registry and included individuals with a diagnosis of EG from 1/1980 to 9/2021 seen at Massachusetts General Hospital, Brigham and Women’s Hospital (BWH), and the BWH Faulkner Hospital.

Protocol Abstract of Mama Sana: Optimizing Health Outcomes for Spanish-speaking Mothers and Their Babies at Brigham and Women’s Hospital

ObjectivesMama Sana is a quality improvement program that will provide perinatal nutrition and breastfeeding (BF) counseling in a culturally and language-specific manner to Spanish-speaking patients at Brigham and Women's Hospital (BWH) (Boston, MA). Our goal is to increase rates of exclusive BF at discharge and 6-weeks postpartum (PP), decrease PP weight retention, and improve diet quality for all patients enrolled.MethodsSpanish-speaking patients screened and identified at a prenatal visit will be able to attend group virtual classes conducted in Spanish on BF, nutrition, and nutrition for gestational diabetes conducted by a bilingual registered dietitian and certified lactation counselor. Data on infant feeding intentions, diet quality, and perceived module effectiveness will be gathered using REDCap questionnaires. All Spanish-speaking patients who deliver at BWH will receive bedside lactation counseling, as well as follow-up counseling at 7–14 days, 3, and 6 months after delivery. We will collect information on BF status and lifestyle habits using standardized questionnaires during each call. Patients will have access to a Spanish BF “warmline” until 1 year PP. Information on maternal and infant health, delivery data, and comorbidities will be gathered from the electronic medical records of participants. Baseline data will be collected from Spanish-speaking patients who delivered at BWH and were discharged prior to program implementation.ResultsWe aim to perform 100 retrospective chart reviews and enroll 100 birthing parents prospectively to reach a total of 200 participants over 12 months. We will examine demographics and compare characteristics between the pre-and post-Mama Sana epochs using the Student's T-test or chi-squared test for continuous and categorical data, respectively. We will compare process measures, outcomes, and balancing measures by epoch using run charts and will examine changes over time.ConclusionsMama Sana combines language and culturally concordant BF and nutrition counseling to target health issues affecting Spanish-speaking mother-infant dyads. We anticipate this wrap-around model will decrease disparities in health care delivery and improve BF initiation and duration rates, as well as diet quality and weight retention in enrolled mothers.Funding SourcesBWH Department of Pediatric of Newborn Medicine.

Deep Learning to Predict Mortality After Cardiothoracic Surgery Using Preoperative Chest Radiographs

BackgroundThe Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) estimates mortality risk only for certain common procedures (eg, coronary artery bypass or valve surgery) and is cumbersome, requiring greater than 60 inputs. We hypothesized that deep learning can estimate postoperative mortality risk based on a preoperative chest radiograph for cardiac surgeries in which STS-PROM scores were available (STS index procedures) or unavailable (non–STS index procedures). MethodsWe developed a deep learning model (CXR-CTSurgery) to predict postoperative mortality based on preoperative chest radiographs in 9283 patients at Massachusetts General Hospital (MGH) having cardiac surgery before April 8, 2014. CXR-CTSurgery was tested on 3615 different MGH patients and externally tested on 2840 patients from Brigham and Women’s Hospital (BWH) having surgery after April 8, 2014. Discrimination for mortality was compared with the STS-PROM using the C-statistic. Calibration was assessed using the observed-to-expected ratio (O/E ratio). ResultsFor STS index procedures, CXR-CTSurgery had a C-statistic similar to STS-PROM at MGH (CXR-CTSurgery: 0.83 vs STS-PROM: 0.88; P = .20) and BWH (0.74 vs 0.80; P = .14) testing cohorts. The CXR-CTSurgery C-statistic for non–STS index procedures was similar to STS index procedures in the MGH (0.87 vs 0.83) and BWH (0.73 vs 0.74) testing cohorts. For STS index procedures, CXR-CTSurgery had better calibration than the STS-PROM in the MGH (O/E ratio: 0.74 vs 0.52) and BWH (O/E ratio: 0.91 vs 0.73) testing cohorts. ConclusionsCXR-CTSurgery predicts postoperative mortality based on a preoperative CXR with similar discrimination and better calibration than the STS-PROM. This may be useful when the STS-PROM cannot be calculated or for non–STS index procedures.

Long-term outcomes of high-risk basal cell carcinoma treated with Mohs micrographic surgery

To the Editor: Mohs micrographic surgery (MMS) is indicated for the treatment of high-risk basal cell carcinoma (BCC) based on superior long-term outcomes compared to wide local excision (WLE).1,2 Factors that may predict poor outcomes include tumor diameter of ≥2 cm, head and neck location, deep invasion, recurrence, aggressive histologic subtype, and perineural invasion.3,4 The American Joint Committee on Cancer (AJCC, 8th edition) and the Brigham and Women Hospital (BWH) staging systems have incorporated some of these parameters to risk-stratify BCC.

Noninvasive Detection of Hippocampal Epileptiform Activity on Scalp Electroencephalogram

The hippocampus is a highly epileptogenic brain region, yet over 90% of hippocampal epileptiform activity (HEA) cannot be identified on scalp electroencephalogram (EEG) by human experts. Currently, detection of HEA requires intracranial electrodes, which limits our understanding of the role of HEA in brain diseases. To develop and validate a machine learning algorithm that accurately detects HEA from a standard scalp EEG, without the need for intracranial electrodes. In this diagnostic study, conducted from 2008 to 2021, EEG data were used from patients with temporal lobe epilepsy (TLE) and healthy controls (HCs) to train and validate a deep neural network, HEAnet, to detect HEA on scalp EEG. Participants were evaluated at tertiary-level epilepsy centers at 2 academic hospitals: Massachusetts General Hospital (MGH) or Brigham and Women's Hospital (BWH). Included in the study were patients aged 12 to 78 years with a clinical diagnosis of TLE and HCs without epilepsy. Patients with TLE and HCs with a history of intracranial surgery were excluded from the study. Simultaneous intracranial EEG and/or scalp EEG. Performance was assessed using cross-validated areas under the receiver operating characteristic curve (AUC ROC) and precision-recall curve (AUC PR) and additional clinically relevant metrics. HEAnet was trained and validated using data sets that were derived from a convenience sample of 141 eligible participants (97 with TLE and 44 HCs without epilepsy) whose retrospective EEG data were readily available. Data set 1 included the simultaneous scalp EEG and intracranial electrode recordings of 51 patients with TLE (mean [SD] age, 40.7 [15.9] years; 30 men [59%]) at MGH. An automatically generated training data set with 972 095 positive HEA examples was created, in addition to a held-out expert-annotated testing data set with 22 762 positive HEA examples. HEAnet's performance was validated on 2 independent scalp EEG data sets: (1) data set 2 (at MGH; 24 patients with TLE and 20 HCs; mean [SD] age, 42.3 [16.2] years; 17 men [39%]) and (2) data set 3 (at BWH; 22 patients with TLE and 24 HCs; mean [SD] age, 43.0 [14.4] years; 20 men [43%]). For single-event detection of HEA on data set 1, HEAnet achieved a mean (SD) AUC ROC of 0.89 (0.01) and a mean (SD) AUC PR of 0.39 (0.03). On external validation with data sets 2 and 3, HEAnet accurately distinguished TLE from HC (AUC ROC of 0.88 and 0.95, respectively) and predicted epilepsy lateralization with 100% and 92% accuracy, respectively. HEAnet tracked dynamic changes in HEA in response to seizure medication adjustments and performed comparably with human experts in diagnosing TLE from 1-hour scalp EEG recordings, diagnosing TLE in several individuals that experts missed. Without reducing specificity, addition of HEAnet to human expert EEG review increased sensitivity for diagnosing TLE in humans from 50% to 58% to 63% to 67%. Results of this diagnostic study suggest that HEAnet provides a novel, noninvasive, quantitative, and clinically relevant biomarker of hippocampal hyperexcitability in humans.

Heterogeneity of Drug Allergies and Reaction Lists in Two U.S. Health Care Systems' Electronic Health Records.

Health care institutions have their own "picklist" for clinicians to document adverse drug reactions (ADRs) into the electronic health record (EHR) allergy list. Whether the lack of a nationally standardized picklist impacts clinician data entries is unknown. The objective of this study was to assess the impact of defined reaction picklists on clinical documentation and, therefore, downstream analytics and clinical research using these data at two institutions. ADR data were obtained from the EHRs of patients who visited the emergency department or outpatient clinics at Brigham and Women's Hospital (BWH) and University of Colorado Hospital (UCH) from 2013 to 2018. Reported drug class ADR prevalences were calculated. We investigated the reactions on each picklist and compared the top 40 reactions at each institution, as well as the top 10 reactions within each drug class. Of 2,160,116 patients, 640,444 (30%) had 928,973 active drug allergies. The most commonly reported drug class allergens were similar between BWH and UCH. BWH's picklist had 48 reactions, and UCH's had 160 reactions; 29 reactions were shared by both picklists. While the top four reactions overall (rash, GI upset/nausea/vomiting, hives, itching) were identical between sites, reactions by drug class exhibited greater documentation diversity. For example, while the summed prevalence of swelling-related reactions to angiotensin-converting-enzyme inhibitors was comparable across sites, swelling was represented by two terms ("swelling," "angioedema") at BWH but 11 terms at UCH (e.g., "swelling," "edema," by body locality). The availability and granularity of reaction picklists impact ADR documentation in the EHR by health care providers; picklists may partially explain variations in reported ADRs across health care systems.

Validation of four cutaneous squamous cell carcinoma staging systems using nationwide data.

SummaryBackgroundCutaneous squamous cell carcinoma (cSCC) is the second most common cancer worldwide with relatively low metastatic potential (2–5%). Developments in therapeutic options have highlighted the need to better identify high‐risk patients who could benefit from closer surveillance, adjuvant therapies and baseline/follow‐up imaging, while at the same time safely omitting low‐risk patients from further follow‐up. Controversy remains regarding the predictive performance of current cSCC staging systems and which methodology to adopt.ObjectivesTo validate the performance of four cSCC staging systems [American Joint Committee on Cancer 8th edition (AJCC8), Brigham and Women’s Hospital (BWH), Tübingen and Salamanca T3 refinement] in predicting metastasis using a nationwide cohort.MethodsA nested case–control study using data from the National Disease Registration Service, England, 2013–2015 was conducted. Metastatic cSCC cases were identified using an algorithm to identify all potential cases for manual review. These were 1 : 1 matched on follow‐up time to nonmetastatic controls randomly selected from 2013. Staging systems were analysed for distinctiveness, homogeneity, monotonicity, specificity, positive predictive value (PPV), negative predictive value (NPV) and c‐index.ResultsWe included 887 metastatic cSCC cases and 887 nonmetastatic cSCC controls. The BWH system showed the highest specificity [92.8%, 95% confidence interval (CI) 90.8–94.3%, PPV (13.2%, 95% CI 10.6–16.2) and c‐index (0.84, 95% CI 0.82–0.86). The AJCC8 showed superior NPV (99.2%, 95% CI 99.2–99.3), homogeneity and monotonicity compared with the BWH and Tübingen diameter and thickness classifications (P < 0.001). Salamanca refinement did not show any improvement in AJCC8 T3 cSCC staging.ConclusionsWe validated four cSCC staging systems using the largest nationwide dataset of metastatic cSCC so far. Although the BWH system showed the highest overall discriminative ability, PPV was low for all staging systems, which shows the need for further improvement and refining of current cSCC staging systems.

Sentinel lymph node biopsy for head and neck cutaneous squamous cell carcinoma using the Brigham and Women's staging system: a cost analysis.

The objective is to determine the cost-effectiveness of sentinel lymph node biopsy (SLNB) for cutaneous squamous cell carcinoma (CSCC) according to the Brigham and Women's Hospital (BWH) Tumor Staging system. A decision analysis was utilized to examine costs and outcomes associated with the use of SLNB in patients with high-risk head and neck CSCC. Decision tree outcome probabilities were obtained from published literature. Costs were derived from Medicare reimbursement rates (US$) and effectiveness was represented by quality-adjusted life-years (QALYs). The primary outcome was the incremental cost-effectiveness ratio (ICER), with a willingness-to-pay set at $100,000 per QALY gained. SLNB was found to be a cost-effective tool for patients with T3 tumors, with an ICER of $18,110.57. Withholding SLNB was the dominant strategy for both T2a and T2b lesions, with ICERs of - $2468.99 and - $16,694.00, respectively. Withholding SLNB remained the dominant strategy when examining immunosuppressed patients with T2a or T2b lesions. In patients with head and neck CSCC, those with T3 or T2b lesionswith additional risk factors not accounted for in the staging system alone, may be considered for SLNB, while in other tumor stages it may be impractical. SLNB should only be offered on an individual patient basis.

Prediction of Poor Outcome for Cutaneous Squamous Cell Carcinoma of the Head and Neck Comparing Classification Systems: A Competing Risk Analysis

The most commonly used staging system for cutaneous squamous cell carcinoma (cSCC) of the head and neck is the American Joint Committee on Cancer eighth edition (AJCC8) system (NVDV, 2018; Stratigos et al., 2020). This system is predominantly based on tumor size in lower stages and extradermal or perineural invasion in higher stages. An alternative classification system for cSCC on all locations is the Brigham and Women’s Hospital (BWH) classification system (Jambusaria-Pahlajani et al., 2013). This system is based on high-risk factors derived from tumor characteristics.

An Explainable Machine Learning Approach Reveals Prognostic Significance of Right Ventricular Dysfunction in Nonischemic Cardiomyopathy

ObjectivesThe authors implemented an explainable machine learning (ML) model to gain insight into the association between cardiac magnetic resonance markers and adverse outcomes of cardiovascular hospitalization and all-cause death (composite endpoint) in patients with nonischemic dilated cardiomyopathy (NICM). BackgroundRisk stratification of patients with NICM remains challenging. An explainable ML model has the potential to provide insight into the contributions of different risk markers in the prediction model. MethodsAn explainable ML model based on extreme gradient boosting (XGBoost) machines was developed using cardiac magnetic resonance and clinical parameters. The study cohorts consist of patients with NICM from 2 academic medical centers: Beth Israel Deaconess Medical Center (BIDMC) and Brigham and Women’s Hospital (BWH), with 328 and 214 patients, respectively. XGBoost was trained on 70% of patients from the BIDMC cohort and evaluated based on the other 30% as internal validation. The model was externally validated using the BWH cohort. To investigate the contribution of different features in our risk prediction model, we used Shapley additive explanations (SHAP) analysis. ResultsDuring a mean follow-up duration of 40 months, 34 patients from BIDMC and 33 patients from BWH experienced the composite endpoint. The area under the curve for predicting the composite endpoint was 0.71 for the internal BIDMC validation and 0.69 for the BWH cohort. SHAP analysis identified parameters associated with right ventricular (RV) dysfunction and remodeling as primary markers of adverse outcomes. High risk thresholds were identified by SHAP analysis and thus provided thresholds for top predictive continuous clinical variables. ConclusionsAn explainable ML-based risk prediction model has the potential to identify patients with NICM at risk for cardiovascular hospitalization and all-cause death. RV ejection fraction, end-systolic and end-diastolic volumes (as indicators of RV dysfunction and remodeling) were determined to be major risk markers.

Gene expression profiling for metastatic risk in head and neck cutaneous squamous cell carcinoma.

ObjectiveOver 50% of newly diagnosed cutaneous squamous cell carcinoma (cSCC) lesions occur in the head and neck (cSCC‐HN), and metastasis to nodal basins in this region further complicates surgical and adjuvant treatment. The current study addressed whether the 40‐gene expression profile (40‐GEP) test can predict metastatic risk in cSCC‐HN with improved accuracy and provide independent prognostic value to complement current risk assessment methods.Study DesignMulticenter, retrospective cohort study.MethodsFormalin‐fixed paraffin‐embedded primary tumor tissue and associated clinical data from patients with cSCC‐HN (n = 278) were collected from 33 independent centers. Samples were analyzed via the 40‐GEP test. Cases were staged per American Joint Committee on Cancer, Eighth Edition (AJCC8) and Brigham and Women's Hospital (BWH) criteria after comprehensive medical record and pathology report review. Metastasis‐free survival (MFS) rates were determined, and risk factors were analyzed via Cox regression.ResultsThe 40‐GEP test classified the cohort into low (Class 1, n = 126; 45.3%), moderate (Class 2A, n = 134; 48.2%), and high (Class 2B, n = 18; 6.5%) metastatic risk at 3 years postdiagnosis. Regional/distant metastasis occurred in 54 patients (19.4%). MFS rates were 92.1% (Class 1), 76.1% (Class 2A), and 44.4% (Class 2B; p < .0001). Multivariate analysis of 40‐GEP results with AJCC8 or BWH tumor stage, or clinicopathologic risk factors, demonstrated independent prognostic value of the 40‐GEP test (p < .03). Accuracy of predicting metastatic risk was also improved using 40‐GEP classification (p < .02).ConclusionsImproved metastatic risk stratification through the 40‐GEP test could complement cSCC‐HN risk assessment for better‐informed decision‐making for treatment and surveillance and ultimately improve patient outcomes.Level of Evidence3

Identifying Brigham and Women's Hospital stage T2a cutaneous squamous cell carcinomas at risk of poor outcomes

Although most of the poor outcomes with cutaneous squamous cell carcinoma (CSCC) occur in high-stage tumors, 26% of nodal metastases and 8% of disease-specific deaths develop in Brigham and Women's Hospital (BWH) T2a tumors. To determine risk factors associated with poor outcomes (nodal metastasis, distant metastases, and disease-specific deaths) in BWH T2a CSCC. A 17-year retrospective multi-institutional cohort study of primary CSCC BWH T2a tumors. A predictive model based on tumor characteristics was developed to identify those at higher risk of poor outcomes. Presence of 1 major criterion (primary tumor diameter ≥40mm, invasion depth beyond subcutaneous fat, poor differentiation, or large-caliber perineural invasion) and ≥ 1 minor criterion (invasion depth in subcutaneous fat, moderate differentiation, small-caliber perineural invasion, or lymphovascular invasion) was most predictive of developing poor outcomes (area under the curve, 0.53; C-statistic, 0.60). This model has a sensitivity of 7.7%, specificity of 97.4%, and a positive and negative predictive value of 33.3% and 86.1%, respectively. The 5-year cumulative incidence of poor outcomes in these tumors is 8.0% (95% CI, 5.1-13.7) compared to 2.8% (95% CI, 1.9-4.1) in other T2a tumors (sub-hazard ratio, 3.0; 95% CI, 1.5-5.8). Multi-institutional cohort study was not externally validated. BWH T2a-high CSCCs have an 8% chance of developing poor outcomes.

Outcomes of Sentinel Lymph Node Biopsy for Primary Cutaneous Squamous Cell Carcinoma of the Head and Neck.

Limited data exist on sentinel lymph node biopsy (SLNB) for cutaneous squamous cell carcinoma (cSCC) of the head and neck. To review the results of SLNB for patients with cSCC of the head and neck at the authors' institution. A retrospective review was completed for patients who underwent SLNB for cSCC of the head and neck over 19 years. Patient demographics, immune status, tumor stage, total patients with positive SLNB, local recurrence, nodal recurrence, in-transit metastasis, and disease-specific death were recorded. Sixty patients underwent lymphoscintigraphy, and an SLN was identified in 58 patients. The mean follow-up was 3.2 years (range, 15 days-16 years). Four patients (6.9%) had a positive SLNB. All were Brigham and Women's Hospital (BWH) stage T2b tumors. Three of these patients were immunosuppressed, 3 patients underwent neck dissection, and 2 patients received adjuvant radiation. None developed local or regional recurrence. Of the 53 patients with a negative SLNB, there were 4 local recurrences, 2 in-transit metastases, and no nodal recurrences. Immunosuppressed patients with tumors BWH stage T2b or greater may be a reasonable cohort to focus future prospective studies on the utility of SLNB in cSCC of the head and neck.