Machine Learning Identifies Variation in Timing of Palliative Care Consultations Among Traumatic Brain Injury Patients.

Background and Objective Timely palliative care involvement offers demonstrable benefits for traumatic brain injury (TBI) patients; however, palliative care consultations (PCCs) are used inconsistently during TBI management. This study aimed to employ advanced machine learning techniques to elucidate the primary drivers of PCC timing variability for TBI patients. Methods Data on admission, hospital course, and outcomes were collected for a cohort of 232 TBI patients who received both PCCs and neurosurgical consultations during the same hospitalization. Principal Component Analysis (PCA) and K-means clustering were used to identify patient phenotypes, which were then compared using Kaplan-Meier analysis. An extreme gradient boosting model (XGBoost) was employed to determine drivers of PCC timing, with model interpretation performed using SHapley Additive exPlanations (SHAP). Results Cluster A (n = 86) consisted mainly of older (median [IQR] = 87 [78, 94] years), White females with mild TBIs and demonstrated the shortest time-to-PCC (2.5 [1.0, 7.0] days). Cluster B (n = 108) also sustained mild TBIs but comprised moderately younger (81 [75, 86] years) married White males with later PCC (5.0 [3.0, 10.8] days). Cluster C (n = 38) represented much younger (46.5 [29.5, 59.8] years), more severely injured, non-White patients with the latest PCC initiation (9.0 [4.2, 17.0] days). The clusters did not differ by discharge disposition (p = 0.4) or frequency inpatient mortality (p > 0.9); however, Kaplan-Meier analysis revealed a significant difference in the time from admission to PCC (p < 0.001), despite no differences in time from admission to mortality (p = 0.18). SHAP analysis of the XGBoost model identified age, sex, and race as the most influential drivers of PCC timing. Conclusions This study highlights crucial disparities in PCC timing for TBI patients and underscores the need for targeted strategies to ensure timely and equitable palliative care integration for this vulnerable population.

Clonally expanded CD38hi cytotoxic CD8 T cells define the T cell infiltrate in checkpoint inhibitor-associated arthritis.

Immune checkpoint inhibitor (ICI) therapies used to treat cancer, such as anti-PD-1 antibodies, can induce autoimmune conditions in some individuals. The T cell mechanisms mediating such iatrogenic autoimmunity and their overlap with spontaneous autoimmune diseases remain unclear. Here, we compared T cells from the joints of 20 patients with an inflammatory arthritis induced by ICI therapy (ICI-arthritis) with two archetypal autoimmune arthritides, rheumatoid arthritis (RA) and psoriatic arthritis (PsA). Single-cell transcriptomic and antigen receptor repertoire analyses highlighted clonal expansion of an activated effector CD8 T cell population in the joints and blood of patients with ICI-arthritis. These cells were identified as CD38hiCD127- CD8 T cells and were uniquely enriched in ICI-arthritis joints compared with RA and PsA and also displayed an elevated interferon signature. In vitro, type I interferon induced CD8 T cells to acquire the ICI-associated CD38hi phenotype and enhanced cytotoxic function. In a cohort of patients with advanced melanoma, ICI therapy markedly expanded circulating CD38hiCD127- T cells, which were frequently bound by the therapeutic anti-PD-1 drug. In patients with ICI-arthritis, drug-bound CD8 T cells in circulation showed marked clonal overlap with drug-bound CD8 T cells from synovial fluid. These results suggest that ICI therapy directly targets CD8 T cells in patients who develop ICI-arthritis and induces an autoimmune pathology that is distinct from prototypical spontaneous autoimmune arthritides.

Anticitrullinated peptide antibody epitope expansion and the HLA DRB1 'shared epitope' are less common in seropositive checkpoint inhibitor-induced inflammatory arthritis than in longstanding rheumatoid arthritis.

BACKGROUND: Immune checkpoint inhibitors (ICI) can potentially cause ICI-inflammatory arthritis (ICI-IA), which often resembles rheumatoid arthritis (RA). In this study, we examined the degree of anticitrullinated peptide antibodies (ACPA) epitope expansion in CCP+ICI-IA and patients with RA. METHODS: We used clinical data and serum from ICI-IA and patients with RA with early disease as well as longstanding disease. A custom, bead-based antigen array was used to identify IgG ACPA reactivities to 18 putative RA-associated citrullinated proteins. Hierarchical clustering software was used to create a heatmap to identify ACPA levels. Additionally, HLA DRB1 typing was performed on ICI-IA patients as well as controls of patients treated with ICI that did not develop ICI-IA (ICI controls). RESULTS: Compared to patients with CCP+RA, patients with CCP+ICI-IA were older (p<0.001), less likely to have positive rheumatoid factor (p<0.001) and had a shorter duration of symptoms (p<0.001). There were less ACPA levels and a lower number of distinct ACPA epitopes in the serum of patients with ICI-IA compared with longstanding patients with RA (p<0.001). Among those tested for HLA DRB1, there were no differences in the frequency of the shared epitope between those with ICI-IA and ICI controls. CONCLUSION: Patients with ICI-IA had lower ACPA titres and targeted fewer ACPA epitopes than longstanding patients with RA, and there were no significant differences in the presence of the shared epitope between those that developed ICI-IA and ICI controls. It remains to be determined if ICI-IA represents an accelerated model of RA pathogenesis with ICI triggering a transition from preclinical to clinical disease.

Comparative safety and effectiveness of TNF inhibitors, IL6 inhibitors and methotrexate for the treatment of immune checkpoint inhibitor-associated arthritis.

OBJECTIVES: To compare the safety and effectiveness of biologic and conventional disease-modifying antirheumatic drugs (DMARDs) for immune checkpoint inhibitor-associated inflammatory arthritis (ICI-IA). METHODS: The retrospective multicentre observational study included patients with a diagnosis of ICI-IA treated with a tumour necrosis factor inhibitor (TNFi), interleukin-6 receptor inhibitor (IL6Ri) and/or methotrexate (MTX); patients with pre-existing autoimmune disease were excluded. The primary outcome was time to cancer progression from ICI initiation; the secondary outcome was time to arthritis control from DMARD initiation. Cox proportional hazard models were used to compare medication groups, adjusting for confounders. RESULTS: 147 patients were included (mean age 60.3 (SD 11.9) years, 66 (45%) women). ICI-IA treatment was TNFi in 33 (22%), IL6Ri 42 (29%) and MTX 72 (49%). After adjustment for time from ICI initiation to DMARD initiation, time to cancer progression was significantly shorter for TNFi compared with MTX (HR 3.27 (95% CI 1.21 to 8.84, p=0.019)) while the result for IL6Ri was HR 2.37 (95% CI 0.94 to 5.98, p=0.055). Time to arthritis control was faster for TNFi compared with MTX (HR 1.91 (95% CI 1.06 to 3.45, p=0.032)) while the result for IL6Ri was HR 1.66 (95% CI 0.93 to 2.97, p=0.089). A subset analysis in patients with melanoma gave similar results for both cancer progression and arthritis control. CONCLUSION: The treatment of ICI-IA with a biologic DMARD is associated with more rapid arthritis control than with MTX, but may be associated with a shorter time to cancer progression.