A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action.

Current machine learning techniques enable robust association of biological signals with measured phenotypes, but these approaches are incapable of identifying causal relationships. Here, we develop an integrated "white-box" biochemical screening, network modeling, and machine learning approach for revealing causal mechanisms and apply this approach to understanding antibiotic efficacy. We counter-screen diverse metabolites against bactericidal antibiotics in Escherichia coli and simulate their corresponding metabolic states using a genome-scale metabolic network model. Regression of the measured screening data on model simulations reveals that purine biosynthesis participates in antibiotic lethality, which we validate experimentally. We show that antibiotic-induced adenine limitation increases ATP demand, which elevates central carbon metabolism activity and oxygen consumption, enhancing the killing effects of antibiotics. This work demonstrates how prospective network modeling can couple with machine learning to identify complex causal mechanisms underlying drug efficacy.

A combinatorially complete epistatic fitness landscape in an enzyme active site.

Protein engineering often targets binding pockets or active sites which are enriched in epistasis-nonadditive interactions between amino acid substitutions-and where the combined effects of multiple single substitutions are difficult to predict. Few existing sequence-fitness datasets capture epistasis at large scale, especially for enzyme catalysis, limiting the development and assessment of model-guided enzyme engineering approaches. We present here a combinatorially complete, 160,000-variant fitness landscape across four residues in the active site of an enzyme. Assaying the native reaction of a thermostable ß-subunit of tryptophan synthase (TrpB) in a nonnative environment yielded a landscape characterized by significant epistasis and many local optima. These effects prevent simulated directed evolution approaches from efficiently reaching the global optimum. There is nonetheless wide variability in the effectiveness of different directed evolution approaches, which together provide experimental benchmarks for computational and machine learning workflows. The most-fit TrpB variants contain a substitution that is nearly absent in natural TrpB sequences-a result that conservation-based predictions would not capture. Thus, although fitness prediction using evolutionary data can enrich in more-active variants, these approaches struggle to identify and differentiate among the most-active variants, even for this near-native function. Overall, this work presents a large-scale testing ground for model-guided enzyme engineering and suggests that efficient navigation of epistatic fitness landscapes can be improved by advances in both machine learning and physical modeling.

High Bacillary Burden and the ESX-1 Type VII Secretion System Promote MHC Class I Presentation by Mycobacterium tuberculosis-Infected Macrophages to CD8 T Cells.

We used a mouse model to study how Mycobacterium tuberculosis subverts host defenses to persist in macrophages despite immune pressure. CD4 T cells can recognize macrophages infected with a single bacillus in vitro. Under identical conditions, CD8 T cells inefficiently recognize infected macrophages and fail to restrict M. tuberculosis growth, although they can inhibit M. tuberculosis growth during high-burden intracellular infection. We show that high intracellular M. tuberculosis numbers cause macrophage death, leading other macrophages to scavenge cellular debris and cross-present the TB10.4 Ag to CD8 T cells. Presentation by infected macrophages requires M. tuberculosis to have a functional ESX-1 type VII secretion system. These data indicate that phagosomal membrane damage and cell death promote MHC class I presentation of the immunodominant Ag TB10.4 by macrophages. Although this mode of Ag presentation stimulates cytokine production that we presume would be host beneficial, killing of uninfected cells could worsen immunopathology. We suggest that shifting the focus of CD8 T cell recognition to uninfected macrophages would limit the interaction of CD8 T cells with infected macrophages and impair CD8 T cell-mediated resolution of tuberculosis.

A Prognostic Model To Predict Survival After Recurrence Among Patients With Recurrent Hepatocellular Carcinoma.

OBJECTIVE: We sought to develop and validate a preoperative model to predict survival after recurrence (SAR) in hepatocellular carcinoma (HCC). BACKGROUND: Although HCC is characterized by recurrence as high as 60%, models to predict outcomes after recurrence remain relatively unexplored. METHODS: Patients who developed recurrent HCC between 2000 and 2020 were identified from an international multi-institutional database. Clinicopathologic data on primary disease and laboratory and radiologic imaging data on recurrent disease were collected. Multivariable Cox regression analysis and internal bootstrap validation (5000 repetitions) were used to develop and validate the SARScore. Optimal Survival Tree analysis was used to characterize SAR among patients treated with various treatment modalities. RESULTS: Among 497 patients who developed recurrent HCC, median SAR was 41.2 months (95% CI 38.1-52.0). The presence of cirrhosis, number of primary tumors, primary macrovascular invasion, primary R1 resection margin, AFP>400 ng/mL on the diagnosis of recurrent disease, radiologic extrahepatic recurrence, radiologic size and number of recurrent lesions, radiologic recurrent bilobar disease, and early recurrence (≤24 months) were included in the model. The SARScore successfully stratified 1-, 3- and 5-year SAR and demonstrated strong discriminatory ability (3-year AUC: 0.75, 95% CI 0.70-0.79). While a subset of patients benefitted from resection/ablation, Optimal Survival Tree analysis revealed that patients with high SARScore disease had the worst outcomes (5-year AUC; training: 0.79 vs. testing: 0.71). The SARScore model was made available online for ease of use and clinical applicability ( https://yutaka-endo.shinyapps.io/SARScore/ ). CONCLUSION: The SARScore demonstrated strong discriminatory ability and may be a clinically useful tool to help stratify risk and guide treatment for patients with recurrent HCC.

Targeted inhibition of kisspeptin neurons reverses hyperandrogenemia and abnormal hyperactive LH secretion in a preclinical mouse model of polycystic ovary syndrome.

STUDY QUESTION: Do hyperactive kisspeptin neurons contribute to abnormally high LH secretion and downstream hyperandrogenemia in polycystic ovary syndrome (PCOS)-like conditions and can inhibition of kisspeptin neurons rescue such endocrine impairments? SUMMARY ANSWER: Targeted inhibition of endogenous kisspeptin neuron activity in a mouse model of PCOS reduced the abnormally hyperactive LH pulse secretion and hyperandrogenemia to healthy control levels. WHAT IS KNOWN ALREADY: PCOS is a reproductive disorder characterized by hyperandrogenemia, anovulation, and/or polycystic ovaries, along with a hallmark feature of abnormal LH hyper-pulsatility, but the mechanisms underlying the endocrine impairments remain unclear. A chronic letrozole (LET; aromatase inhibitor) mouse model recapitulates PCOS phenotypes, including polycystic ovaries, anovulation, high testosterone, and hyperactive LH pulses. LET PCOS-like females also have increased hypothalamic kisspeptin neuronal activation which may drive their hyperactive LH secretion and hyperandrogenemia, but this has not been tested. STUDY DESIGN, SIZE, DURATION: Transgenic KissCRE+/hM4Di female mice or littermates Cre- controls were treated with placebo, or chronic LET (50 µg/day) to induce a PCOS-like phenotype, followed by acute (once) or chronic (2 weeks) clozapine-N-oxide (CNO) exposure to chemogenetically inhibit kisspeptin cells (n = 6 to 10 mice/group). PARTICIPANTS/MATERIALS, SETTING, METHODS: Key endocrine measures, including in vivo LH pulse secretion patterns and circulating testosterone levels, were assessed before and after selective kisspeptin neuron inhibition and compared between PCOS groups and healthy controls. Alterations in body weights were measured and pituitary and ovarian gene expression was determined by qRT-PCR. MAIN RESULTS AND THE ROLE OF CHANCE: Acute targeted inhibition of kisspeptin neurons in PCOS mice successfully lowered the abnormally hyperactive LH pulse secretion (P < 0.05). Likewise, chronic selective suppression of kisspeptin neuron activity reversed the previously high LH and testosterone levels (P < 0.05) down to healthy control levels and rescued reproductive gene expression (P < 0. 05). LARGE SCALE DATA: N/A. LIMITATIONS, REASONS FOR CAUTION: Ovarian morphology was not assessed in this study. Additionally, mouse models can offer mechanistic insights into neuroendocrine processes in PCOS-like conditions but may not perfectly mirror PCOS in women. WIDER IMPLICATIONS OF THE FINDINGS: These data support the hypothesis that overactive kisspeptin neurons can drive neuroendocrine PCOS-like impairments, and this may occur in PCOS women. Our findings complement recent clinical investigations using NKB receptor antagonists to lower LH in PCOS women and suggest that pharmacological dose-dependent modulation of kisspeptin neuron activity may be a valuable future therapeutic target to clinically treat hyperandrogenism and lower elevated LH in PCOS women. STUDY FUNDING/COMPETING INTEREST(S): This research was supported by NIH grants R01 HD111650, R01 HD090161, R01 HD100580, P50 HD012303, R01 AG078185, and NIH R24 HD102061, and a pilot project award from the British Society for Neuroendocrinology. There are no competing interests.

The Association of Food Insecurity and Surgical Outcomes Among Patients Undergoing Surgery for Colorectal Cancer.

BACKGROUND: Food insecurity predisposes individuals to suboptimal nutrition, leading to chronic disease and poor outcomes. OBJECTIVE: We sought to assess the impact of county-level food insecurity on colorectal surgical outcomes. DESIGN: Retrospective cohort study. SETTING: Data from the Surveillance, Epidemiology, and End Results-Medicare database was merged with county-level food insecurity obtained from the Feeding America: Mapping the Meal Gap report. Multiple logistic and Cox regression adjusted for patient-level covariates were implemented to assess outcomes. PATIENTS: Medicare beneficiaries diagnosed with colorectal cancer between 2010 and 2015. MAIN OUTCOME MEASURES: Surgical admission type (nonelective and elective admission), any complication, extended length of stay, discharge disposition (discharged to home and nonhome discharge), 90-day readmission, 90-day mortality, and textbook outcome. Textbook outcome was defined as no extended length of stay, postoperative complications, 90-day readmission, and 90-day mortality. RESULTS: Among 72,354 patients with colorectal cancer, 46,296 underwent resection. Within the surgical cohort, 9091 (19.3%) were in low, 27,716 (59.9%) were in moderate, and 9,489 (20.5%) were in high food insecurity counties. High food insecurity patients had greater odds of nonelective surgery (OR: 1.17; 95% CI, 1.09-1.26; p < 0.001), 90-day readmission (OR: 1.11; 95% CI, 1.04-1.19; p = 0.002), extended length of stay (OR: 1.32; 95% CI, 1.21-1.44; p < 0.001), and complications (OR: 1.11; 95% CI, 1.03-1.19; p = 0.002). High food insecurity patients also had decreased odds of home discharge (OR: 0.85; 95% CI, 0.79-0.91; p < 0.001) and textbook outcomes (OR: 0.81; 95% CI, 0.75-0.87; p < 0.001). High food insecurity minority patients had increased odds of complications (OR 1.59; 95% CI, 1.43-1.78) and extended length of stay (OR 1.89; 95% CI, 1.69-2.12) compared with low food insecurity white patients (all, p < 0.001). Notably, high food insecurity minority patients had 31% lower odds of textbook outcomes (OR: 0.69; 95% CI, 0.62-0.76; p < 0.001) compared with low food insecurity White patients ( p < 0.001). LIMITATIONS: This study was limited to Medicare beneficiaries aged 65 years or older; hence, it may not be generalizable to younger populations or those without insurance or with private insurance. CONCLUSIONS: County-level food insecurity was associated with suboptimal outcomes, demonstrating the importance of interventions to mitigate these inequities. See Video Abstract. LA ASOCIACIN DE INSEGURIDAD ALIMENTARIA Y RESULTADOS QUIRRGICOS ENTRE PACIENTES SOMETIDOS A CIRUGA DE CNCER COLORRECTAL: ANTECEDENTES:La inseguridad alimentaria predispone a las personas a una nutrición subóptima, lo que conduce a enfermedades crónicas y malos resultados.OBJETIVO:Intentamos evaluar el impacto de la inseguridad alimentaria a nivel de condado en resultados de la cirugía colorrectal.DISEÑO:Estudio de cohorte retrospectivo.ENTORNO CLINICO:La base de datos SEER-Medicare fusionada con la inseguridad alimentaria a nivel de condado obtenida del informe Feeding America: Mapping the Meal Gap. Para evaluar los resultados se implementaron regresiones logísticas múltiples y de Cox ajustadas según las covariables a nivel de paciente.PACIENTES:Beneficiarios de Medicare diagnosticados con cáncer colorrectal entre 2010 y 2015.PRINCIPALES MEDIDAS DE RESULTADO:Tipo de ingreso quirúrgico (ingreso no electivo y electivo), cualquier complicación, duración prolongada de la estancia hospitalaria, disposición del alta (alta al domicilio y alta no domiciliaria), reingreso a los 90 días, mortalidad a los 90 días y resultado del libro de texto. El resultado de los libros de texto se definió como ausencia de estancia hospitalaria prolongada, complicaciones postoperatorias, reingreso a los 90 días y mortalidad a los 90 días.RESULTADOS:Entre 72.354 pacientes con cáncer colorrectal, 46.296 se sometieron a resección. Dentro de la cohorte quirúrgica, 9.091 (19,3%) tenían inseguridad alimentaria baja, 27.716 (59,9%) eran moderadas y 9.489 (20,5%) tenían inseguridad alimentaria alta. Los pacientes con alta inseguridad alimentaria tuvieron mayores probabilidades de cirugía no electiva (OR: 1,17, IC 95%: 1,09-1,26, p <0,001), reingreso a los 90 días (OR: 1,11, IC95%: 1,04-1,19, p = 0,002), duración prolongada de la estancia hospitalaria (OR: 1,32; IC95%: 1,21-1,44, p < 0,001) y complicaciones (OR: 1,11; IC95%: 1,03-1,19, p = 0,002). Los pacientes con alta inseguridad alimentaria también tuvieron menores probabilidades de ser dados de alta a domicilio (OR: 0,85, IC del 95%: 0,79-0,91, p <0,001) y resultados de los libros de texto (OR: 0,81, IC del 95%: 0,75-0,87, p <0,001). Los pacientes minoritarios con alta inseguridad alimentaria tuvieron mayores probabilidades de complicaciones (OR 1,59, IC 95%, 1,43-1,78) y duración prolongada de la estadía (OR 1,89, IC 95%, 1,69-2,12) en comparación con los individuos blancos con baja inseguridad alimentaria (todos, p < 0,001). En particular, los pacientes minoritarios con alta inseguridad alimentaria tenían un 31% menos de probabilidades de obtener resultados según los libros de texto (OR: 0,69, IC del 95%, 0,62-0,76, p <0,001) en comparación con los pacientes blancos con baja inseguridad alimentaria ( p <0,001).LIMITACIONES:Limitado a beneficiarios de Medicare mayores de 65 años, por lo tanto, puede no ser generalizable a poblaciones más jóvenes o a aquellos sin seguro o con seguro privado.CONCLUSIONES:La inseguridad alimentaria a nivel de condado se asoció con resultados subóptimos, lo que demuestra la importancia de las intervenciones para mitigar estas desigualdades. (Dr. Francisco M. Abarca-Rendon ).

Chronological and Geographical Variations in the Incidence and Acceptance of COVID-19-Positive Donors and Outcomes Among Abdominal Transplant Patients.

INTRODUCTION: Given the importance of understanding COVID-19-positive donor incidence and acceptance, we characterize chronological and geographic variations in COVID-19 incidence relative to COVID-19-positive donor acceptance. METHODS: Data on deceased donors and recipients of liver and kidney transplants were obtained from the UNOS database between 2020 and 2023. Hierarchical cluster analysis was used to assess trends in COVID-19-positive donor incidence. Posttransplant graft and patient survival were assessed using Kaplan-Meier curves. RESULTS: From among 38 429 deceased donors, 1517 were COVID-19 positive. Fewer kidneys (72.4% vs. 76.5%, p < 0.001) and livers (56.4% vs. 62.0%, p < 0.001) were used from COVID-19-positive donors versus COVID-19-negative donors. Areas characterized by steadily increased COVID-19 donor incidence exhibit the highest transplantation acceptance rates (92.33%), followed by intermediate (84.62%) and rapidly increased (80.00%) COVID-19 incidence areas (p = 0.016). Posttransplant graft and patient survival was comparable among recipients, irrespective of donor COVID-19 status. CONCLUSIONS: Regions experiencing heightened rates of COVID-19-positive donors are associated with decreased acceptance of liver and kidney transplantation. Similar graft and patient survival is noted among recipients, irrespective of donor COVID-19 status. These findings emphasize the need for adaptive practices and unified medical consensus in navigating a dynamic pandemic.

Effect of the opioid crisis on the liver transplantation donor pool: A national analysis.

BACKGROUND: Over the last decade there has been a surge in overdose deaths due to the opioid crisis. We sought to characterize the temporal change in overdose donor (OD) use in liver transplantation (LT), as well as associated post-LT outcomes, relative to the COVID-19 era. METHODS: LT candidates and donors listed between January 2016 and September 2022 were identified from the Scientific Registry of Transplant Recipients database. Trends in LT donors and changes related to OD were assessed pre- versus post-COVID-19 (February 2020). RESULTS: Between 2016 and 2022, most counties in the United States experienced an increase in overdose-related deaths (n = 1284, 92.3%) with many counties (n = 458, 32.9%) having more than a doubling in drug overdose deaths. Concurrently, there was an 11.2% increase in overall donors, including a 41.7% increase in the number of donors who died from drug overdose. In pre-COVID-19 overdose was the 4th top mechanism of donor death, while in the post-COVID-19 era, overdose was the 2nd most common cause of donor death. OD was younger (OD: 35 yrs, IQR 29-43 vs. non-OD: 43 yrs, IQR 31-56), had lower body mass index (≥35 kg/cm2, OD: 31.2% vs. non-OD: 33.5%), and was more likely to be HCV+ (OD: 28.9% vs. non-OD: 5.4%) with lower total bilirubin (≥1.1 mg/dL, OD: 12.9% vs. non-OD: 20.1%) (all p < .001). Receipt of an OD was not associated with worse graft survival (HR .94, 95% CI .88-1.01, p = .09). CONCLUSIONS: Opioid deaths markedly increased following the COVID-19 pandemic, substantially altering the LT donor pool in the United States.

Erdheim-Chester disease: misdiagnosed as multiple sclerosis.

Erdheim-Chester disease is a rare histiocytic neoplasm with a wide range of clinical manifestations. Due to its rarity and protean characteristics, this condition often presents a diagnostic challenge. A Caucasian woman in her late 60s presented with unsteadiness, dysphagia and dysarthria. She was initially diagnosed with secondary progressive multiple sclerosis but deteriorated over 2 years with a potential lack of therapeutic response. Subsequent investigations resulted in the diagnosis of Erdheim-Chester disease. She received targeted therapy with BRAF and MAPK-pathway inhibitors. Her initial response to treatment has been positive with functional gains and reduced disease burden on MR brain imaging, and with no significant adverse effects.

Prognostic significance of postoperative complications for patients with hepatocellular carcinoma relative to alpha-feto protein and tumor burden score.

BACKGROUND: We sought to elucidate the impact of postoperative complications on patient outcomes relative to differences in alpha-fetoprotein-tumor burden score (ATS) among patients with hepatocellular carcinoma (HCC). METHODS: Patients who underwent resection of HCC between 2000 and 2020 were identified from an international database. Moderate/severe complications were defined using the optimal cut-off value of the comprehensive complication index (CCI) based on the log-rank test. RESULTS: A total of 1124 patients was included. CCI cut-off value of 16.6 was identified as the optimal prognostic threshold. Patients who experienced moderate/severe complications were more likely to have worse recurrence free survival [RFS] versus individuals who had no/mild complications (2-year RFS; no/mild complication: 55.9% vs. moderate/severe complication: 38.1% p < 0.001). Of note, low and medium ATS patients who experienced moderate/severe complications had a higher risk of recurrence (2-year RFS; no/mild complication: postoperative complications 70.0% vs. moderate/severe complication: 51.1%, p = 0.006; medium: no/mild complication: 50.8% vs moderate/severe complication: 56.7%, p = 0.01); however, postoperative complications were not associated with worse outcomes among patients with high ATS (no/mild complication: 39.1% vs. moderate/severe complication: 29.2%, p = 0.20). CONCLUSION: These data serve to emphasize how reduction in postoperative complications may be crucial to improve prognosis, particularly among patients with favorable HCC characteristics.

Efficacy and safety of pralsetinib in patients with advanced RET fusion-positive non-small cell lung cancer.

BACKGROUND: Pralsetinib is a potent, selective RET inhibitor targeting oncogenic RET alterations. As part of the global, phase 1/2 ARROW trial (NCT03037385), the efficacy and safety of pralsetinib in Chinese patients with advanced RET fusion-positive non-small cell lung cancer (NSCLC) were evaluated. METHODS: Adult patients with advanced, RET fusion-positive NSCLC with or without prior platinum-based chemotherapy were enrolled into two cohorts receiving 400-mg once-daily oral pralsetinib. Primary end points were objective response rates assessed by blinded independent central review and safety. RESULTS: Of 68 patients enrolled, 37 had received prior platinum-based chemotherapy (48.6% with ≥3 prior systemic regimens) and 31 were treatment-naïve. As of March 4, 2022 (data cutoff), of the patients with measurable lesions at baseline, a confirmed objective response was observed in 22 (66.7%; 95% confidence interval [CI], 48.2-82.0) of 33 pretreated patients, including 1 (3.0%) complete response and 21 (63.6%) partial responses; and in 25 (83.3%; 95% CI, 65.3-94.4) of 30 treatment-naïve patients, including two (6.7%) complete responses and 23 (76.7%) partial responses. Median progression-free survival was 11.7 months (95% CI, 8.7-not estimable) in pretreated patients and 12.7 months (95% CI, 8.9-not estimable) in treatment-naïve patients. The most common grade 3/4 treatment-related adverse events in 68 patients were anemia (35.3%) and decreased neutrophil count (33.8%). Eight (11.8%) patients discontinued pralsetinib because of treatment-related adverse events. CONCLUSION: Pralsetinib showed robust and durable clinical activity with a well-tolerated safety profile in Chinese patients with RET fusion-positive NSCLC. CLINICAL TRIAL REGISTRATION: NCT03037385.

The Impact of a Liver Transplant Program on the Outcomes of Hepatocellular Carcinoma.

OBJECTIVE: We sought to evaluate the impact of liver transplantation (LT) programs on the prognosis of hepatocellular carcinoma (HCC) patients who underwent liver resection (LR) and noncurative intent treatment. BACKGROUND: LT programs have an array of resources and services that would positively affect the prognosis of patients with HCC. METHODS: Patients who underwent LT, LR, radiotherapy (RT), or chemotherapy (CTx) for HCC between 2004 and 2018 were included in the National Cancer Database. Institutions with LT programs were defined as those that performed 1 or more LT for at least 5 years. Centers were stratified by hospital volume. The impact of LT programs was assessed after propensity score matching to achieve covariate balance. RESULTS: A total of 71,735 patients were identified, of which 7997 received LT (11.1%), 12,683 LR (17.7%), 15,675 RT (21.9%), and 35,380 CTx (49.3%). Among a total of 1267 distinct institutions, 94 (7.4%) were categorized as LT programs. Designation as an LT program was also associated with a high volume of LR and noncurative intent treatment (both P <0.001). After propensity score matching, LT programs were associated with better survival among LR and noncurative intent treatment patients. Although hospital volume was also associated with improved prognosis, LT programs were associated with additional survival benefits in noncurative intent treatment. On the other hand, no such benefit was noted in patients who underwent LR. CONCLUSIONS: The presence of an LT program was associated with a higher volume of LR and noncurative intent treatment. Furthermore, designation as an LT program had a "halo effect" on the prognosis of patients undergoing RT/CTx that went beyond the procedure-volume effect.

Efficacy and Safety of Avapritinib in Treating Unresectable or Metastatic Gastrointestinal Stromal Tumors: A Phase I/II, Open-Label, Multicenter Study.

BACKGROUND: Avapritinib is a type 1 kinase inhibitor designed to potently and selectively inhibit oncogenic KIT/PDGFRA mutants by targeting the kinase active conformation. This multicenter, single-arm, open-label, phase I/II bridging study of NAVIGATOR in Chinese patients evaluated the safety and the antineoplastic activity of avapritinib in Chinese patients with unresectable/metastatic gastrointestinal stromal tumors (GIST). METHODS: Phase I comprised dose escalation for safety and phase II dose determination. Phase II comprised dose expansion for safety/efficacy evaluations in patients with PDGFRA D842V mutations or patients having received at least 3 lines of therapy without PDGFRA D842V mutations. The primary endpoints were recommended phase II dose, safety, and Independent Radiology Review Committee (IRRC)-assessed objective response rate (ORR). RESULTS: No dose-limiting toxicities occurred (n = 10); the recommended phase II dose was avapritinib 300 mg once daily orally. Fifty-nine patients initially received avapritinib 300 mg. Common grade ≥3 treatment-related adverse events were anemia, decreased white blood cell count, increased blood bilirubin levels, and decreased neutrophil count. In patients with PDGFRA D842V mutations, IRRC- and investigator-assessed ORRs were 75% and 79%, respectively; clinical benefit rates were both 86%. Median duration of response/progression-free survival were not reached. IRCC- and investigator-assessed ORRs in patients in the fourth- or later-line setting were 22% and 35%, respectively. Median progression-free survivals were 5.6 months for both. Overall survival data were immature and not calculated. CONCLUSION: Avapritinib was generally well tolerated and showed marked anti-tumor activity in Chinese patients with GIST bearing PDGFRA D842V mutations and notable efficacy as fourth- or later-line monotherapy (ClinicalTrials.gov Identifier: NCT04254939).

Suicidal Ideation Among Patients with Gastrointestinal Cancer.

BACKGROUND: Mental illness (MI) and suicidal ideation (SI) often are associated with a diagnosis of cancer. We sought to define the incidence of MI and SI among patients with gastrointestinal cancers, as well as ascertain the predictive factors associated with SI. METHODS: Patients diagnosed between 2004 and 2016 with stomach, liver, pancreatic, and colorectal cancer were identified from the SEER-Medicare database. County-level social vulnerability index (SVI) was extracted from the Centers for Disease Control database. Multivariable logistic regression was used to identify factors associated with SI. RESULTS: Among 382,266 patients, 83,514 (21.9%) individuals had a diagnosis of MI. Only 1410 (0.4%) individuals experienced SI, and 359 (0.1%) committed suicide. Interestingly, SI was least likely among patients with pancreatic cancer (ref: hepatic cancer; odds ratio [OR] 0.67, 95% confidence interval [CI] 0.52-0.86; p = 0.002), as well as individuals with stage III/IV disease (OR 0.59, 95% CI 0.52-067; p < 0.001). In contrast, male (OR 1.34, 95% CI 1.19-1.50), White (OR 1.34, CI 1.13-1.59), and single (OR 2.03, 95% CI 1.81-2.28) patients were at higher odds of SI risk (all p < 0.001). Furthermore, individuals living in relative privilege (low SVI) had markedly higher risk of SI (OR 1.33, 95% CI 1.14-1.54; p < 0.001). Moreover, living in a county with a shortage of mental health professionals was associated with increased odds of developing SI (OR 1.21, 95% CI 1.04-1.40; p = 0.012). CONCLUSIONS: Oncology care teams should incorporate routine mental health and SI screening in the treatment of patients with gastrointestinal cancers, as well as target suicide prevention towards patients at highest risk.

Development and Validation of a Machine-Learning Model to Predict Early Recurrence of Intrahepatic Cholangiocarcinoma.

BACKGROUND: The high incidence of early recurrence after hepatectomy for intrahepatic cholangiocarcinoma (ICC) has a detrimental effect on overall survival (OS). Machine-learning models may improve the accuracy of outcome prediction for malignancies. METHODS: Patients who underwent curative-intent hepatectomy for ICC were identified using an international database. Three machine-learning models were trained to predict early recurrence (< 12 months after hepatectomy) using 14 clinicopathologic characteristics. The area under the receiver operating curve (AUC) was used to assess their discrimination ability. RESULTS: In this study, 536 patients were randomly assigned to training (n = 376, 70.1%) and testing (n = 160, 29.9%) cohorts. Overall, 270 (50.4%) patients experienced early recurrence (training: n = 150 [50.3%] vs testing: n = 81 [50.6%]), with a median tumor burden score (TBS) of 5.6 (training: 5.8 [interquartile range {IQR}, 4.1-8.1] vs testing: 5.5 [IQR, 3.7-7.9]) and metastatic/undetermined nodes (N1/NX) in the majority of the patients (training: n = 282 [75.0%] vs testing n = 118 [73.8%]). Among the three different machine-learning algorithms, random forest (RF) demonstrated the highest discrimination in the training/testing cohorts (RF [AUC, 0.904/0.779] vs support vector machine [AUC, 0.671/0.746] vs logistic regression [AUC, 0.668/0.745]). The five most influential variables in the final model were TBS, perineural invasion, microvascular invasion, CA 19-9 lower than 200 U/mL, and N1/NX disease. The RF model successfully stratified OS relative to the risk of early recurrence. CONCLUSIONS: Machine-learning prediction of early recurrence after ICC resection may inform tailored counseling, treatment, and recommendations. An easy-to-use calculator based on the RF model was developed and made available online.

The Association of Persistent Poverty and Outcomes Among Patients Undergoing Cardiac Surgery.

INTRODUCTION: We sought to evaluate the association of county-level poverty duration and cardiac surgical outcomes. METHODS: Patients who underwent coronary artery bypass graft, surgical aortic valve replacement, and mitral valve repair and replacement between 2016 and 2020 were identified using the Medicare Standard Analytical Files Database. County-level poverty data were acquired from the American Community Survey and US Department of Agriculture (1980-2015). High poverty was defined as ≥19.5% of residents in poverty. Patients were stratified into never-high poverty (NHP), intermittent low poverty, intermittent high poverty, and persistent poverty (PP). A mixed-effect hierarchical generalized linear model and Cox regression models that adjusted for patient-level covariates were used to evaluate outcomes. RESULTS: Among 237,230 patients, 190,659 lived in NHP counties, while 10,273 resided in PP counties. Compared with NHP patients, PP patients were more likely to present at a younger median age (NHP: 75 y versus PP: 74 y), be non-Hispanic Black (5388, 2.9% versus PP: 1030, 10.1%), and live in the south (NHP: 66,012, 34.6% versus PP: 87,815, 76.1%) (all P < 0.001). PP patients also had more nonelective surgical operations (NHP: 58,490, 30.8% versus 3645, 35.6%, P < 0.001). Notably, PP patients had increased odds of 30-d mortality (odds ratio 1.13, 95% confidence interval [CI] 1.02-1.26), 90-d mortality (odds ratio 1.14, 95% CI 1.05-1.24), and risk of long-term mortality (hazard ratio 1.13, 95% CI 1.09-1.19) compared with patients in NHP counties (all P < 0.05). CONCLUSIONS: County-level poverty was associated with a greater risk of short- and long-term mortality among cardiac surgical patients.

Travel distance and social vulnerability index: Impact on liver-related mortality among patients with end-stage liver disease.

INTRODUCTION: The reasons for the geographic disparities in liver-related mortality across the US remain ill-defined. We sought to investigate the impact of travel distance to liver transplantation (LT) programs and social vulnerability on county differences in liver-related mortality. METHODS: Data on LT registrants were obtained from the Scientific Registry of Transplant Recipients Standard Analytic Files (SRTR SAFs) between 2004 and 2019. Liver-related mortality data were obtained from the Center for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) platform. Spatial epidemiological clustering of county-level LT registration and liver-related mortality rates was determined using local Moran's I. Comparison analyses assessed social vulnerability index (SVI) and travel distance within various county clusters. RESULTS: Among 151 864 LT waitlist registrants who were diagnosed with liver disease due to hepatitis C virus (HCV) or hepatitis B virus (HBV) (n = 68 479, 45.1%), alcohol (n = 38 328, 25.2%), non-alcoholic steatohepatitis (NASH) (n = 17 485, 11.5%), liver tumors (n = 16 644, 11.0%), and other diseases (n = 10 928, 7.2%), median SVI was 59.3 (IQR, 40.1-83.4). SVI (76.2 vs. 24.3, p < .001) was greater in the highest versus lowest liver-related mortality quartiles. The travel distances to LT centers (143.1 miles vs. 107.2 miles, p < .001) was longer in the lowest versus highest LT registration quartiles. Counties with low LT registration rates and high liver-related mortality rates were associated with long travel distances and high SVI. In contrast, while counties with high LT registration rates and high liver-related mortality rates had comparable SVI, travel distance was relatively shorter. CONCLUSION: Counties with greater SVIs were associated with higher liver-related mortality, with the highest SVI  counties having the highest overall liver-related mortality. Longer travel distances were associated with higher liver-related mortality. These findings highlight the impact of social determinants of health (SDOH) on liver disease outcomes.

Association between the environmental quality index and textbook outcomes among Medicare beneficiaries undergoing surgery for colorectal cancer (CRC).

BACKGROUND: Quality of cancer care received by individuals may be influenced by environmental factors resulting in inequalities within the healthcare system. We sought to investigate the association between the Environmental Quality Index (EQI) and achievement of textbook outcomes (TOs) among Medicare beneficiaries who underwent surgical resection for colorectal cancer (CRC). METHODS: Patients diagnosed with CRC from 2004 to 2015 were identified using the Surveillance, Epidemiology, and End Results-Medicare database and merged with the US Environmental Protection Agency's EQI data. A high EQI category indicated poor environmental quality, whereas a low EQI indicated better environmental conditions. RESULTS: Among 40 939 patients, 33 699 (82.3%) were diagnosed with colon cancer, 7240 (17.7%) were diagnosed with rectal cancer, and 652 (1.6%) were diagnosed with both cancers. Median age was 76 years old (interquartile range: 70-82 years) with roughly half of patients being female (n = 22 033, 53.8%). Most patients self-reported as White (n = 32 404, 79.2%) and resided in the West region of the United States (n = 20 308, 49.6%). On multivariable analysis, patients residing in high EQI areas were less likely to achieve TO (referent: low EQI; odds ratio [OR]: 0.94, 95% confidence interval [95% CI]: 0.89-0.99; p = 0.02). Of note, Black patients living in moderate-to-high EQI counties had a 31% decreased likelihood of reaching a TO compared with White patients in low EQI counties (OR: 0.69, 95% CI: 0.55-0.87). CONCLUSION: Patients residing in high EQI counties and Black race were associated with a lower likelihood of TO following resection of CRC among Medicare patients. Environmental factors may be important contributors to health care disparities and affect postoperative outcomes following CRC resection.

The Effects of Chronic Steroid Use on Postoperative Complications Following Thyroidectomy.

BACKGROUND: Patients undergoing thyroidectomy are sometimes on chronic steroids for underlying disease. This study examined the postoperative risk profile of thyroidectomy patients on chronic steroids. METHODS: Patients in the National Surgical Quality Improvement Program (NSQIP) database who underwent thyroidectomy were sorted by presence or absence of chronic steroid use. Clinicodemographics, comorbidities, and postoperative complications were recorded and compared between the two. Univariate and multivariate analyses compared the groups and calculated odds ratios (OR). RESULTS: We identified 42,857 patients. 41,903 (97.8%) patients were not on chronic steroids, while 954 (2.2%) were. Most underwent total thyroidectomy (18,748, 43.75%) or total lobectomy (16,323, 38.09%). Following univariate and multivariate analyses, patients on chronic steroids had increased risk of postoperative bleeding and transfusions (OR = 0.375, p = 0.046, 95% CI 0.223-0.988), open wound infection (OR = 0.226, p < 0.001, 95% CI 0.117-0.437), pulmonary embolism (OR = 0.312, p = 0.034, 95% CI 0.106-0.918), and ventilator use > 48 h (OR = 0.401, p < 0.008, 95% CI 0.205-0.785). CONCLUSIONS: Chronic steroid use prior to thyroidectomy is an independent risk factor for multiple postoperative complications, namely postoperative bleeding and transfusions, open wound infection, pulmonary embolism, and ventilator use over 48 h. Patients on chronic steroids should be medically optimized before thyroidectomy to reduce the risk of potentially life-threatening complications.

Psychosocial Burden of Pediatric and Adult Patients With Congenital Versus Traumatic Facial Differences: Assessment of Psychiatric Distress and Healthcare Utilization in the United States From 2004 to 2012.

BACKGROUND: Psychiatric distress and its effects on healthcare utilization in pediatric patients with congenital and traumatic facial differences remain poorly understood. This study analyzes the psychosocial burden along with mental health and reconstructive surgery services utilization of this patient population in comparison with adult patients with such facial differences. METHODS: The 2004-2012 Medical Expenditures Panel Survey was queried for all patients with facial differences. Socioeconomic variables, Patient Health Questionnaire 2 and Kessler 6 scores, responses from validated screening surveys, and utilization of mental health and reconstructive surgery (ie, plastic surgery and otolaryngology) services were compared between pediatric and adult patients with congenital and traumatic facial differences. RESULTS: Children ages 5 to 12 years were more likely to be affected by facial trauma, whereas adolescents aged 13 to 17 years were more affected by congenital facial conditions. Pediatric patients with congenital facial conditions had higher rates of medical care, education, and special therapy utilization ( P < 0.0001), although their facial trauma counterparts used mental health services more often ( P < 0.0001). In adults, more facial trauma patients reported poorer perceived mental health status ( P = 0.01). Among patients with any facial difference, distressed adult patients were less likely to see a reconstructive surgeon even when controlling for socioeconomic variables (0.55 [0.31-0.97], P = 0.04). CONCLUSIONS: In the pediatric population, psychosocial considerations should include both age and etiology of facial differences to best optimize care. Among adults with facial trauma, poor mental health may contribute to lower rates of surgical follow-up, highlighting a potential benefit for provision of mental health services earlier for these patient populations.