Safety and Efficacy Outcomes of Early Cessation of Anti-PD1 Therapy in Patients 80 Years or Older: A Retrospective Cohort Study.

Older patients have similar immune checkpoint inhibitor efficacy and rates of adverse events as younger patients, but appear to have decreased tolerability, particularly in the oldest patient cohort (>80 years), often leading to early cessation of therapy. We aimed to determine whether early discontinuation impacts efficacy of anti-PD-1 therapy in patients ≥80 years old. In this retrospective, multicenter, international cohort study, we examined 773 patients with 4 tumor types who were at least 80 years old and treated with anti-PD-1 therapy. We determined response rate, overall survival (OS), and progression-free survival (PFS) in patients who discontinued therapy early (<12 months) for reasons other than progression or death. We used descriptive statistics for demographics, response, and toxicity rates. Survival statistics were described using Kaplan Meier curves. Median (range) age at anti-PD-1 initiation was 83.0 (75.8-97.0) years. The cancer types included were melanoma (n=286), non-small cell lung cancer (NSCLC) (n=345), urothelial cell carcinoma (UCC) (n=108), and renal cell carcinoma (RCC) (n=34). Of these, 102 met the primary endpoint of <12 months to discontinuation for reasons other than death or progression. Median PFS and OS, respectively, for these patients were 34.4 months and 46.6 months for melanoma, 15.8 months and 23.4 months for NSCLC, and 10.4 months and 15.8 months for UCC. This study suggests geriatric patients who have demonstrated therapeutic benefit and discontinued anti-PD-1 therapy at less than 12 months of duration for reasons other than progression may have durable clinical benefit without additional therapy.

Long-term survival follow-up for tebentafusp in previously treated metastatic uveal melanoma.

BACKGROUND: Tebentafusp, a bispecific (gp100×CD3) ImmTAC, significantly improved overall survival (OS) outcomes for HLA-A*02:01+ adult patients with untreated metastatic uveal melanoma (mUM) and showed promising survival in previously treated mUM with 1-year OS of 62% in the primary analysis of study IMCgp100-102. Here we report long-term outcomes from this phase 1/2 study in pretreated mUM. PATIENTS AND METHODS: Patients with previously treated mUM received tebentafusp weekly intravenous at 20 µg dose 1, 30 µg dose 2 and either 54, 64, 68, or 73 µg (phase 1) or 68 µg (phase 2) dose 3+. The primary objective was overall response rate. Secondary objectives included OS and safety. OS was estimated by Kaplan-Meier methods. Association between OS and baseline covariates, on-treatment Response Evaluation Criteria in Solid Tumors (RECIST) response, baseline tumor biopsy and circulating-tumor DNA (ctDNA) changes were assessed. RESULTS: 146 patients were treated with tebentafusp: 19 in phase 1 and 127 in phase 2. With a median follow-up duration of 48.5 months, the median OS was 17.4 months (95% CI, 13.1 to 22.8), and the 1-year, 2-year, 3-year and 4-year OS rates were 62%, 40%, 23% and 14%, respectively. Improved survival was associated with lower ctDNA baseline levels and greater ctDNA reductions by week 9 on-treatment, with 100% 1-year, 73% 2-year and 45% 3-year OS rates for patients with ctDNA clearance. Baseline gp100 expression was not associated with survival, despite more RECIST responses among patients with higher expression. No new safety signals were reported with long-term dosing. CONCLUSIONS: This study represents the longest follow-up of a Tcell receptor bispecific to date and confirms the durable survival benefits achieved with tebentafusp in previously treated mUM with good tolerability long-term. A role for ctDNA reduction as an early indicator of clinical benefit was again suggested for patients treated with tebentafusp.

Oral care practices and hospital-acquired pneumonia prevention: A national survey of Australian nurses.

BACKGROUND: Hospital-acquired pneumonia (HAP) also known as non-ventilator associated pneumonia, is one of the most common infections acquired in hospitalised patients. Improving oral hygiene appears to reduce the incidence of HAP. This study aimed to describe current practices, barriers and facilitators, knowledge and educational preferences of registered nurses performing oral health care in the Australian hospital setting, with a focus on the prevention of HAP. We present this as a short research report. METHODS: We undertook a cross sectional online anonymous survey of Australian registered nurses. Participants were recruited via electronic distribution through existing professional networks and social media. The survey used was modified from an existing survey on oral care practice. RESULTS: The survey was completed by 179 participants. Hand hygiene was considered a very important strategy to prevent pneumonia (n = 90, 58%), while 45% (n = 71) felt that oral care was very important. The most highly reported barriers for providing oral care included: an uncooperative patient; inadequate staffing; and a lack of oral hygiene requisite. Patients' reminders, prompts and the provision of toothbrushes were common ways believed to help facilitate improvements in oral care. CONCLUSION: Findings from this survey will be used in conjunction with consumer feedback, to help inform a planned multi-centre randomised trial, the Hospital Acquired Pneumonia PrEveNtion (HAPPEN) study, aimed at reducing the incidence of HAP. Findings may also be useful for informing studies and quality improvement initiatives aimed at improving oral care to reduce the incidence of HAP.

Real-World Multicenter Study of PD-1 Blockade in HIV-Associated Classical Hodgkin Lymphoma Across the United States.

BACKGROUND: Despite a higher risk of classical Hodgkin lymphoma (cHL) in people with HIV and the demonstrated safety and efficacy of PD-1 blockade in cHL, there are limited data on the use of these agents in HIV-associated cHL (HIV-cHL). PATIENTS/METHODS: We retrospectively identified patients with HIV-cHL from the "Cancer Therapy using Checkpoint inhibitors in People with HIV-International (CATCH-IT)" database who received nivolumab or pembrolizumab, alone or in combination with other agents, and reviewed records for demographics, disease characteristics, immune-mediated adverse events (imAEs), and treatment outcomes. Changes in CD4+ T-cell counts with treatment were measured via Wilcoxon signed-rank tests. Overall response rate (ORR) was defined as the proportion of patients with partial or complete response (PR/CR) per 2014 Lugano classification. RESULTS: We identified 23 patients with HIV-cHL who received a median of 6 cycles of PD-1 blockade: 1 as 1st-line, 6 as 2nd-line, and 16 as ≥3rd-line therapy. Seventeen (74%) patients received monotherapy, 5 (22%) received nivolumab plus brentuximab vedotin, and 1 received nivolumab plus ifosfamide, carboplatin, and etoposide. The median baseline CD4+ T-cell count was 155 cells/µL, which increased to 310 cells/µL at end-of-treatment (P = .009). Three patients had grade 3 imAEs; none required treatment discontinuation. The ORR was 83% with median duration of response of 19.7 months. The median progression-free survival was 21.2 months and did not differ between patients with <200 versus ≥200 CD4+ cells/µL (P = .95). CONCLUSION: Our findings support the use of PD-1 blockade in HIV-cHL for the same indications as the general population with cHL.

Anti-PD-1 alone or in combination with anti-CTLA-4 for advanced melanoma patients with liver metastases.

BACKGROUND: The combination of anti-PD-1 and anti-CTLA-4 has been associated with improvement in response and survival over anti-PD-1 monotherapy in unselected patients with advanced melanoma. Whether patients with liver metastases also benefit from the combination of anti-PD-1 and anti-CTLA-4 over anti-PD-1, is unclear. In this study, we sought to assess whether the combination of anti-PD-1 and anti-CTLA-4 leads to better response, progression-free survival and overall survival, compared with anti-PD-1 monotherapy for patients with liver metastases. METHODS: We have conducted an international multicentre retrospective study. Patients with advanced melanoma with liver metastases treated with 1st line anti-PD1 monotherapy or with anti-CTLA-4 were included. The endpoints of this study were: objective response rate, progression-free survival and overall survival. RESULTS: With a median follow-up from commencement of anti-PD-1 monotherapy or in combination with anti-CTLA-4 of 47 months (95% CI, 42-51), objective response rate was higher with combination therapy (47%) versus anti-PD-1 monotherapy (35%) (p = 0.0027), while progression-free survival and overall survival were not statistically different between both treatment groups. However, on multivariable analysis with multiple imputation for missing values and adjusting for predefined variables, combination of anti-PD1 and anti-CTLA-4 was associated with higher objective response (OR 2.21, 1.46 - 3.36; p < 0.001), progression-free survival (HR 0.73, 0.57 - 0.92; p = 0.009) and overall survival (HR 0.71, 0.54 - 0.94; p = 0.018) compared to anti-PD1 monotherapy. CONCLUSIONS: Findings from this study will help guide treatment selection for patients who present with liver metastases, suggesting that combination therapy should be considered for this group of patients.

Osmosensor-mediated control of Ca2+ spiking in pollen germination.

Higher plants survive terrestrial water deficiency and fluctuation by arresting cellular activities (dehydration) and resuscitating processes (rehydration). However, how plants monitor water availability during rehydration is unknown. Although increases in hypo-osmolarity-induced cytosolic Ca2+ concentration (HOSCA) have long been postulated to be the mechanism for sensing hypo-osmolarity in rehydration1,2, the molecular basis remains unknown. Because osmolarity triggers membrane tension and the osmosensing specificity of osmosensing channels can only be determined in vivo3-5, these channels have been classified as a subtype of mechanosensors. Here we identify bona fide cell surface hypo-osmosensors in Arabidopsis and find that pollen Ca2+ spiking is controlled directly by water through these hypo-osmosensors-that is, Ca2+ spiking is the second messenger for water status. We developed a functional expression screen in Escherichia coli for hypo-osmosensitive channels and identified OSCA2.1, a member of the hyperosmolarity-gated calcium-permeable channel (OSCA) family of proteins6. We screened single and high-order OSCA mutants, and observed that the osca2.1/osca2.2 double-knockout mutant was impaired in pollen germination and HOSCA. OSCA2.1 and OSCA2.2 function as hypo-osmosensitive Ca2+-permeable channels in planta and in HEK293 cells. Decreasing osmolarity of the medium enhanced pollen Ca2+ oscillations, which were mediated by OSCA2.1 and OSCA2.2 and required for germination. OSCA2.1 and OSCA2.2 convert extracellular water status into Ca2+ spiking in pollen and may serve as essential hypo-osmosensors for tracking rehydration in plants.

Discovery and Preclinical Characterization of BIIB129, a Covalent, Selective, and Brain-Penetrant BTK Inhibitor for the Treatment of Multiple Sclerosis.

Multiple sclerosis (MS) is a chronic disease with an underlying pathology characterized by inflammation-driven neuronal loss, axonal injury, and demyelination. Bruton's tyrosine kinase (BTK), a nonreceptor tyrosine kinase and member of the TEC family of kinases, is involved in the regulation, migration, and functional activation of B cells and myeloid cells in the periphery and the central nervous system (CNS), cell types which are deemed central to the pathology contributing to disease progression in MS patients. Herein, we describe the discovery of BIIB129 (25), a structurally distinct and brain-penetrant targeted covalent inhibitor (TCI) of BTK with an unprecedented binding mode responsible for its high kinome selectivity. BIIB129 (25) demonstrated efficacy in disease-relevant preclinical in vivo models of B cell proliferation in the CNS, exhibits a favorable safety profile suitable for clinical development as an immunomodulating therapy for MS, and has a low projected total human daily dose.

Use of artificial intelligence chatbots in clinical management of immune-related adverse events.

BACKGROUND: Artificial intelligence (AI) chatbots have become a major source of general and medical information, though their accuracy and completeness are still being assessed. Their utility to answer questions surrounding immune-related adverse events (irAEs), common and potentially dangerous toxicities from cancer immunotherapy, are not well defined. METHODS: We developed 50 distinct questions with answers in available guidelines surrounding 10 irAE categories and queried two AI chatbots (ChatGPT and Bard), along with an additional 20 patient-specific scenarios. Experts in irAE management scored answers for accuracy and completion using a Likert scale ranging from 1 (least accurate/complete) to 4 (most accurate/complete). Answers across categories and across engines were compared. RESULTS: Overall, both engines scored highly for accuracy (mean scores for ChatGPT and Bard were 3.87 vs 3.5, p<0.01) and completeness (3.83 vs 3.46, p<0.01). Scores of 1-2 (completely or mostly inaccurate or incomplete) were particularly rare for ChatGPT (6/800 answer-ratings, 0.75%). Of the 50 questions, all eight physician raters gave ChatGPT a rating of 4 (fully accurate or complete) for 22 questions (for accuracy) and 16 questions (for completeness). In the 20 patient scenarios, the average accuracy score was 3.725 (median 4) and the average completeness was 3.61 (median 4). CONCLUSIONS: AI chatbots provided largely accurate and complete information regarding irAEs, and wildly inaccurate information ("hallucinations") was uncommon. However, until accuracy and completeness increases further, appropriate guidelines remain the gold standard to follow.

Interoperability of phenome-wide multimorbidity patterns: a comparative study of two large-scale EHR systems.

Background: Electronic health records (EHR) are increasingly used for studying multimorbidities. However, concerns about accuracy, completeness, and EHRs being primarily designed for billing and administrative purposes raise questions about the consistency and reproducibility of EHR-based multimorbidity research. Methods: Utilizing phecodes to represent the disease phenome, we analyzed pairwise comorbidity strengths using a dual logistic regression approach and constructed multimorbidity as an undirected weighted graph. We assessed the consistency of the multimorbidity networks within and between two major EHR systems at local (nodes and edges), meso (neighboring patterns), and global (network statistics) scales. We present case studies to identify disease clusters and uncover clinically interpretable disease relationships. We provide an interactive web tool and a knowledge base combining data from multiple sources for online multimorbidity analysis. Findings: Analyzing data from 500,000 patients across Vanderbilt University Medical Center and Mass General Brigham health systems, we observed a strong correlation in disease frequencies (Kendall's τ = 0.643) and comorbidity strengths (Pearson ρ = 0.79). Consistent network statistics across EHRs suggest similar structures of multimorbidity networks at various scales. Comorbidity strengths and similarities of multimorbidity connection patterns align with the disease genetic correlations. Graph-theoretic analyses revealed a consistent core-periphery structure, implying efficient network clustering through threshold graph construction. Using hydronephrosis as a case study, we demonstrated the network's ability to uncover clinically relevant disease relationships and provide novel insights. Interpretation: Our findings demonstrate the robustness of large-scale EHR data for studying phenome-wide multimorbidities. The alignment of multimorbidity patterns with genetic data suggests the potential utility for uncovering shared biology of diseases. The consistent core-periphery structure offers analytical insights to discover complex disease interactions. This work also sets the stage for advanced disease modeling, with implications for precision medicine. Funding: VUMC Biostatistics Development Award, the National Institutes of Health, and the VA CSRD.

Defining D-irAEs: consensus-based disease definitions for the diagnosis of dermatologic adverse events from immune checkpoint inhibitor therapy.

With an increasing number of patients eligible for immune checkpoint inhibitors, the incidence of immune-related adverse events (irAEs) is on the rise. Dermatologic immune-related adverse events (D-irAEs) are the most common and earliest to manifest, often with important downstream consequences for the patient. Current guidelines lack clarity in terms of diagnostic criteria for D-irAEs. The goal of this project is to better define D-irAE for the purposes of identification, diagnosis, and future study of this important group of diseases.The objectives of this project were to develop consensus guidance for an approach to D-irAEs including disease definitions and severity grading. Knowing that consensus among oncologists, dermatologists, and irAE subspecialists would be critical for usability, we formed a Dermatologic irAE Disease Definition Panel. The panel was composed of 34 experts, including oncologists, dermatologists, a rheumatologist, and an allergist/immunologist from 22 institutions across the USA and internationally. A modified Delphi consensus process was used, with two rounds of anonymous ratings by panelists and two virtual meetings to discuss areas of controversy. Panelists rated content for usability, appropriateness, and accuracy on 9-point scales in electronic surveys and provided free text comments. A working group aggregated survey responses and incorporated them into revised definitions. Consensus was based on numeric ratings using the RAND/UCLA Appropriateness Method with prespecified definitions.Following revisions based on panelist feedback, all items received consensus in the second round of ratings. Consensus definitions were achieved for 10 core D-irAE diagnoses: ICI-vitiligo, ICI-lichen planus, ICI-psoriasis, ICI-exanthem, ICI-bullous pemphigoid, ICI-Grover's, ICI-eczematous, ICI-eruptive atypical squamous proliferation, ICI-pruritus without rash, and ICI-erosive mucocutaneous. A standard evaluation for D-irAE was also found to reach consensus, with disease-specific exceptions detailed when necessary. Each disorder's description includes further details on disease subtypes, symptoms, supportive exam findings, and three levels of diagnostic certainty (definite, probable, and possible).These consensus-driven disease definitions standardize D-irAE classification in a useable framework for multiple disciplines and will be the foundation for future work. Given consensus on their accuracy and usability from a representative panel group, we anticipate that they can be used broadly across clinical and research settings.

Utilization and Outcomes of Multigene Panel Testing in Patients With Pancreatic Ductal Adenocarcinoma.

PURPOSE: Guidelines recommend germline genetic testing (GT) for patients with pancreatic ductal adenocarcinoma (PDAC). This study aims to evaluate the utilization and outcomes of multigene panel GT in patients with PDAC. METHODS: This retrospective, multisite study included patients with PDAC diagnosed between May 2018 and August 2020 at Mayo Clinic Arizona, Florida, and Minnesota. Discussion, uptake, and outcomes of GT were compared before (May 1, 2018-May 1, 2019) and after (August 1, 2019-August 1, 2020) the guideline update, accounting for a transition period. RESULTS: The study identified 533 patients with PDAC, with 321 (60.2%) preguideline and 212 (39.8%) postguideline. Patient characteristics did not differ between the preguideline and postguideline periods. GT was discussed in 34.3% (110 of 321) of preguideline and 39.6% (84 of 212) of postguideline patients (odds ratio [OR], 1.26 [95% CI, 0.88 to 1.80]) and subsequently performed in 80.9% (89 of 110) of preguideline and 75.0% (63 of 84) of postguideline patients (OR, 1.10 [95% CI, 0.75 to 1.61]). Of 152 tested patients, 26 (17.1%) had a pathogenic variant (PV), of whom 17 (11.2%; 17 of 152) were PDAC-associated. Over the entire study period, GT was more likely in younger patients (65 v 70 years; P < .001), those seen by a medical oncologist (82.9% v 69.0%; P < .001), and those surviving more than 12 months from diagnosis (70.4% v 43.4%; P < .001). Demographics and personal/family cancer history were comparable between patients with and without a PDAC PV. CONCLUSION: GT remains underutilized despite National Comprehensive Cancer Network guideline recommendations. Given the poor prognosis of PDAC and potential implications of GT, efforts to increase utilization are needed to provide surveillance and support to both patients with PDAC and at-risk family members.

Investigational Approaches for Treatment of Melanoma Patients Progressing After Standard of Care.

ABSTRACT: The advent of effective immunotherapy, specifically cytotoxic T-lymphocyte associated protein 4 and programmed cell death 1 inhibitors, as well as targeted therapy including BRAF/MEK inhibitors, has dramatically changed the prognosis for metastatic melanoma patients. Up to 50% of patients may experience long-term survival currently. Despite these advances in melanoma treatment, many patients still progress and die of their disease. As such, there are many studies aimed at providing new treatment options for this population. Therapies currently under investigation include, but are not limited to, novel immunotherapies, targeted therapies, tumor-infiltrating lymphocytes and other cellular therapies, oncolytic viral therapy and other injectables, and fecal microbiota transplant. In this review, we discuss the emerging treatment options for metastatic melanoma patients who have progressed on standard of care treatments.

Anti-PD-(L)1 plus BRAF/MEK inhibitors (triplet therapy) after failure of immune checkpoint inhibition and targeted therapy in patients with advanced melanoma.

BACKGROUND: Effective treatment options are limited for patients with advanced melanoma who have progressed on immune checkpoint inhibitors (ICI) and targeted therapies (TT). Preclinical models support the combination of ICI with TT; however, clinical trials evaluating the efficacy of triplet combinations in first-line setting showed limited advantage compared to TT only. METHODS: We conducted a retrospective, multicenter study, that included patients with advanced melanoma who were treated with BRAF/MEK inhibitors in combination with an anti-PD-(L)1 antibody (triplet therapy) after failure of at least one anti-PD-(L)1-based therapy and one TT in seven major melanoma centers between February 2016 and July 2022. RESULTS: A total of 48 patients were included, of which 32 patients, 66.7% had brain metastases, 37 patients (77.1%) had three or more metastatic organs and 21 patients (43.8%) had three or more treatment lines. The median follow-up time was 31.4 months (IQR, 22.27-40.45 months). The treatment with triplet therapy resulted in an ORR of 35.4% (n = 17) and a DCR of 47.9% (n = 23). The median DOR was 5.9 months (range, 3.39-14.27 months). Patients treated with BRAF/MEK inhibitors as the last treatment line showed a slightly lower ORR (29.6%) compared to patients who received ICI or chemotherapy last (ORR: 42.9%). Grade 3-4 treatment-related adverse events occurred in 25% of patients (n = 12), with seven patients (14.6%) requiring discontinuation of treatment with both or either drug. CONCLUSIONS: Triplet therapy has shown activity in heavily pretreated patients with advanced melanoma and may represent a potential treatment regimen after failure of ICI and TT.

Polygenic risk score for ulcerative colitis predicts immune checkpoint inhibitor-mediated colitis.

Immune checkpoint inhibitor-mediated colitis (IMC) is a common adverse event of treatment with immune checkpoint inhibitors (ICI). We hypothesize that genetic susceptibility to Crohn's disease (CD) and ulcerative colitis (UC) predisposes to IMC. In this study, we first develop a polygenic risk scores for CD (PRSCD) and UC (PRSUC) in cancer-free individuals and then test these PRSs on IMC in a cohort of 1316 patients with ICI-treated non-small cell lung cancer and perform a replication in 873 ICI-treated pan-cancer patients. In a meta-analysis, the PRSUC predicts all-grade IMC (ORmeta=1.35 per standard deviation [SD], 95% CI = 1.12-1.64, P = 2×10-03) and severe IMC (ORmeta=1.49 per SD, 95% CI = 1.18-1.88, P = 9×10-04). PRSCD is not associated with IMC. Furthermore, PRSUC predicts severe IMC among patients treated with combination ICIs (ORmeta=2.20 per SD, 95% CI = 1.07-4.53, P = 0.03). Overall, PRSUC can identify patients receiving ICI at risk of developing IMC and may be useful to monitor patients and improve patient outcomes.

Benefit, recurrence pattern, and toxicity to adjuvant anti-PD-1 monotherapy varies by ethnicity and melanoma subtype: An international multicenter cohort study.

Background: Anti-Program-Death-1 (PD-1) is a standard adjuvant therapy for patients with resected melanoma. We hypothesized that there are discrepancies in survival, recurrence pattern and toxicity to adjuvant PD-1 between different ethnicities and melanoma subtypes. Objective: We performed a multicenter cohort study incorporating 6 independent institutions in Australia, China, Japan, and the United States. The primary outcomes were recurrence free survival (RFS) and overall survival (OS). Secondary outcomes were disease recurrence patterns and toxicities. Results: In total 534 patients were included. East-Asian/Hispanic/African reported significantly poorer RFS/OS. Nonacral cutaneous or melanoma of unknown primary reported the best RFS/OS, followed by acral, and mucosal was the poorest. Within the nonacral cutaneous or melanoma of unknown primary subtypes, East-Asian/Hispanic/African reported significantly poorer RFS/OS than Caucasian. In the multivariate analysis incorporating ethnicity/melanoma-subtype/age/sex/stage/lactate dehydrogenase/BRAF (v-Raf murine sarcoma viral oncogene homolog B)-mutation/adjuvant radiotherapy, East-Asian/Hispanic/African had independently significantly poorer outcomes (RFS: HR, 1.71; 95% CI, 1.19-2.44 and OS: HR, 2.34; 95% CI, 1.39-3.95), as was mucosal subtype (RFS: HR, 3.25; 95% CI, 2.04-5.17 and OS: HR, 3.20; 95% CI, 1.68-6.08). Mucosal melanoma was an independent risk factor for distant metastasis, especially liver metastasis. East-Asian/Hispanic/African had significantly lower incidence of gastrointestinal/musculoskeletal/respiratory/other-rare-type-toxicities; but higher incidences of liver toxicities. Limitations: A retrospective study. Conclusions: Ethnicity and melanoma subtype are associated with survival and recurrence pattern in melanoma patients treated with adjuvant anti-PD-1. Toxicity profile differs by ethnicity and may require a precision toxicity surveillance strategy.

Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data.

PURPOSE: Although immune checkpoint inhibitors (ICIs) have improved outcomes in certain patients with cancer, they can also cause life-threatening immunotoxicities. Predicting immunotoxicity risks alongside response could provide a personalized risk-benefit profile, inform therapeutic decision making, and improve clinical trial cohort selection. We aimed to build a machine learning (ML) framework using routine electronic health record (EHR) data to predict hepatitis, colitis, pneumonitis, and 1-year overall survival. METHODS: Real-world EHR data of more than 2,200 patients treated with ICI through December 31, 2018, were used to develop predictive models. Using a prediction time point of ICI initiation, a 1-year prediction time window was applied to create binary labels for the four outcomes for each patient. Feature engineering involved aggregating laboratory measurements over appropriate time windows (60-365 days). Patients were randomly partitioned into training (80%) and test (20%) sets. Random forest classifiers were developed using a rigorous model development framework. RESULTS: The patient cohort had a median age of 63 years and was 61.8% male. Patients predominantly had melanoma (37.8%), lung cancer (27.3%), or genitourinary cancer (16.4%). They were treated with PD-1 (60.4%), PD-L1 (9.0%), and CTLA-4 (19.7%) ICIs. Our models demonstrate reasonably strong performance, with AUCs of 0.739, 0.729, 0.755, and 0.752 for the pneumonitis, hepatitis, colitis, and 1-year overall survival models, respectively. Each model relies on an outcome-specific feature set, though some features are shared among models. CONCLUSION: To our knowledge, this is the first ML solution that assesses individual ICI risk-benefit profiles based predominantly on routine structured EHR data. As such, use of our ML solution will not require additional data collection or documentation in the clinic.

Management of infliximab refractory immune checkpoint inhibitor gastrointestinal toxicity: a multicenter case series.

BACKGROUND: Immune checkpoint inhibitor (ICI) gastrointestinal toxicity (gastritis, enteritis, colitis) is a major cause of morbidity and treatment-related death. Guidelines agree steroid-refractory cases warrant infliximab, however best management of infliximab-refractory ICI gastrointestinal toxicity (IRIGItox) is unknown. METHODS: We conducted an international multicenter retrospective case series. IRIGItox was defined as failure of symptom resolution ≤grade 1 (Common Terminology Criteria for Adverse Events V.5.0) following ≥2 infliximab doses or failure of symptom resolution ≤grade 2 after one dose. Data were extracted regarding demographics, steroid use, response to treatment, and survival outcomes. Toxicity was graded at symptom onset and time of infliximab failure. Efficacy of infliximab refractory therapy was assessed by symptom resolution, time to resolution and steroid wean duration. Survival outcomes were examined based on immunosuppressive therapy received. RESULTS: 78 patients were identified: median age 60 years; 56% men; majority melanoma (N=70, 90%); 60 (77%) received anti-cytotoxic T-lymphocyte-associated protein 4 alone or in combination with anti-programmed cell death protein-1 and most had colitis (N=74, 95%). 106 post-infliximab treatments were given: 31 calcineurin inhibitors (CNIs); 27 antimetabolites (mycophenolate, azathioprine); 16 non-systemic immunomodulatory agents (eg, mesalazine or budesonide); 15 vedolizumab; 5 other biologics (anti-interleukin-12/23, 16, Janus kinase inhibitors) and 7 interventional procedures (including colectomy); 5 did not receive post-infliximab therapy. Symptom resolution was achieved in most (N=23/31, 74%) patients treated with CNIs; 12/27 (44%) with antimetabolites; 7/16 (44%) with non-systemic immunomodulation, 8/15 (53%) with vedolizumab and 5/7 (71%) with interventional procedures. No non-vedolizumab biologics resulted in toxicity resolution. CNIs had the shortest time to symptom resolution (12 days) and steroid wean (43 days); however, were associated with poorer event-free survival (6.3 months) and overall survival (26.8 months) than other agents. Conversely, vedolizumab had the longest time to toxicity resolution and steroid wean, 66 and 124 days, but most favorable survival data: EFS 24.5 months; median OS not reached. Six death occurred (three due to IRIGItox or management of toxicity; three with persisting IRIGItox and progressive disease). CONCLUSIONS: IRIGItox causes major morbidity and mortality. Management is heterogeneous. CNIs appear most likely to result in toxicity resolution in the shortest time period, however, are associated with poorer oncological outcomes in contrast to vedolizumab.