Citations

Purpose: The understanding of cardiac-involved ECD (ECD-CV) and how this affects outcomes and pre-existing diseases remains limited. We aimed to determine the association, burden, and severity of pre-existing and de novo cardiovascular (CV) comorbidities with ECD-CV compared to patients (pts) without cardiac-involved ECD (ECD-noCV). Methods: Records of pts with ECD diagnosed from Jan-1990 to Dec-2021, consecutively seen at Mayo Clinic, AZ, FL, and MN were reviewed. Cardiac involvement was diagnosed by imaging. CV comorbidities were captured before and after ECD diagnosis. CTCAE v3.0 based grading system was used to assess CV comorbidities which graded the conditions as mild (grade 1), moderate (grade 2), severe (grade 3), and life-threatening or disabling (grade 4). Chi-square test was used to determine if having ECD-CV worsens pre-existing CV comorbidities in pts with ECD. As per the established methods, the cumulative CV disease burden was defined after considering 8 CV morbidities that are known to have direct structural and functional implications on the heart, including hypertension (HTN), coronary artery disease (CAD), myocardial infarction (MI), congestive heart failure (CHF), atrial fibrillation (afib), pericardial effusion, valvular disease, and conduction disorders. A composite CV morbidity grade was assigned based on the highest grade in one of the 8 CV comorbidities (Oeffinger, NEJM, 2006). Lastly, a multivariant analysis was conducted using the presence of BRAFV600E mutation, smoking, age ≥ 60 years (yr), conventional vs targeted therapy, and ECD-CV in pts with a composite grade of 4. Results: Among 106 ECD pts, 38% (n=40) had cardiac involvement. The median age at diagnosis of ECD was 57 yr (range: 49-67); 61% were males. The median follow-up was 5 yr (95% CI: 3.5-6.1), and 4.1 yr (95% CI: 3.2-6.5) for ECD-CV and ECD-noCV pts, respectively p=0.92. The median time of CV diagnoses for the entire cohort before and after ECD diagnosis was 4.8 yr (range: 0.06-42.5) and 3.4 yr (range: 0.05-18.7), respectively. BRAFV600E mutation was present in 65% (n=26) of ECD-CV vs 42% (n=28) of ECD-noCV, p=0.02. Eight pts with ECD-CV died; one from sudden cardiac death, while 9 ECD-noCV pts died, p=0.39. Smoking was present in 43% (n=18) of ECD-CV pts compared to 24% with ECD-noCV (n=16), p=0.05. In comparison to ECD-noCV pts, increased prevalence of pre-existing HTN [59% (n=23) vs 35% (n=23), p=0.016] and MI [10% (n=4) vs 2% (n=1), p=0.046] was seen in ECD-CV. There was no difference in hyperlipidemia, diabetes mellitus, CAD, CHF, valvular disease, afib, and conduction disorders among the two cohorts. After ECD diagnosis, de novo diagnosis of hyperlipidemia [67% (n=26) vs 38% (n=25); p=0.004] and CHF [35% (n=14) vs 7% (n=5); p=<0.001] was seen more frequently in ECD-CV pts compared to ECD-noCV. Following the diagnosis of ECD, the severity of pre-existing conditions did not change in pts with or without ECD-CV. ECD-CV pts had a greater cumulative CV disease burden as the total number of CV morbidities was greater in ECD-CV pts both before and after ECD diagnosis (Table 1). Additionally, pts with ECD-CV had greater severity of CV morbidity (composite grade ≥4) than ECD-noCV before (35% v 3%; p=<0.001) and after (42.5% vs 7.6%; p=<0.001) ECD diagnosis. The composite grade of 4 before ECD diagnosis was most attributable to pericardial effusion (42% in pts with ECD-CV vs 0% in pts with ECD-noCV) and CAD (50% in pts with ECD-CV vs 100% in pts with ECD-noCV); after ECD diagnosis pericardial effusion (42% in pts with ECD-CV vs 0% in pts with ECD-noCV) and CAD (41% in pts with ECD-CV vs 100% in pts with ECD-noCV). After considering BRAFV600E mutation, age ≥60, smoking, ECD-CV, and first-line therapy (conventional vs targeted therapy), ECD-CV was the only factor independently associated with a composite grade of ≥4, before (OR 25.6, 95% CI: 4.8-137) and after (OR 9.7, 95% CI: 3.0-30.8) ECD diagnosis. Nonetheless, smoking, age >60, and ECD-CV were independent factors associated when considered a composite grade of ≥3, before and after ECD diagnosis. Conclusion: ECD-CV did not prognosticate pts and did not worsen the pre-existing CV conditions compared to pts with ECD-noCV. However, ECD-CV pts had a higher cumulative CV disease burden and a higher grade of pre-existing and de novo CV morbidities compared to ECD-noCV. The presence of cardiac involvement in ECD was independently associated with a severe cumulative CV disease burden grade.

Purpose: Advances in targeted therapeutics for histiocytic neoplasms have led to a growing survivor population, especially for Erdheim-Chester disease (ECD), Langerhans cell histiocytosis (LCH), and Rosai-Dorfman disease (RDD). This growing population is at risk of adverse health-related quality of life (HRQoL) as a result of the disease or treatments received. Patient-reported outcomes in this population remain understudied, and we aim to define their HRQoL using a validated tool. Methods: We enrolled patients from 2 histiocytosis centers and an international conference from Aug 2022-April 2023 to complete the PROMIS © (Patient-Reported Outcomes [PROs] Measurement Information System) 29+2 profile v2.1, which captures 8 domains of physical/mental health and functioning. Domains are reported as T-scores, with 50 being the general population mean. We calculated summary T-scores for Physical Health (PH) and Mental Health (MH), and differences of 3 from general population were considered clinically meaningful. We then used these T-scores to calculate each patient's PROMIS-Preference (PROPr) score, which is a health-utility score that measures overall HRQoL. A mean PROPr score difference of 0.4 from the US population mean of 0.52 was considered meaningful. We ran independent-samples t-tests to examine differences in PH/MH summary scores and PROPr between sub-cohorts. Finally, we performed a multinomial logistic regression analysis to identify factors associated with worse scores, and calculated odds ratios (OR) for each factor. Results: We included a total of 77 participants. Specific diagnoses included ECD (n=33, 42.9%), LCH (n=21, 27.3%), RDD (n=19, 24.7%), and other/unknown (n=4, 5.2%). Median age at diagnosis was 50y (9-79) and at survey was 54y (18-82), with a median 4 year difference between diagnosis and survey completion. Racial demographics included white (n=64, 83.1%), Black (n=12, 15.6%), and Asian (n=1, 1.3%). 4 patients (5.2%) identified as having a Hispanic ethnicity. Disease extent included unifocal (n=13, 16.9%), single system multifocal (ss-M, n=8, 10.4%), single system lung LCH (ss-L, n=3, 3.9%), or multisystem (n=53, 68.8%) disease. The most commonly involved organs included bone (n=44, 57.1%), kidney (n=24, 31.2%), skin (n=24, 31.2%), CNS (n=24, 31.2%), and pituitary (n=19, 24.7%). PH and MH summary scores were lower in various diagnostic categories as compared with the US population, and PROPr scores were reduced in all diagnostic categories. Bone involvement was associated with worse PH (-6.16, 95%CI -10.81, -1.52, [p=0.005]), MH (-7.96, 95%CI -12.48, -3.44, [p=<0.001]), and PROPr (-0.184, 95%CI -0.313, -0.056, [p=0.003]) scores. CNS involvement was associated with worse PH (-4.78, 95%CI -9.14, -0.42, [p=0.016]). We initially constructed a multinomial regression model using diagnosis, race, age at survey completion, and CNS and bone involvement. This model was not statistically significant, and in particular, diagnosis and age did not significantly contribute to the model. We then removed diagnosis and age and constructed a new model using the remaining variables, which was statistically significant for all 3 scores (p<0.05). We found that factors associated with worse PH score were disease involvement of bone (OR=6.19, 95%CI 1.88, 20.43 [p=0.003]) and CNS (OR=10.36, 95%CI 2.01, 53.39 [p=0.005] Figure 2). Factors associated with worse MH scores included female sex (OR=4.12, 95%CI 1.24, 13.66 [p=0.021]) and CNS involvement (OR=10.74, 95%CI 2.47, 46.79 [p=0.002]). Having race/ethnic group other than white/Non-Hispanic or Black was associated with improved MH. However, as this group had only 4 members, the significance of this is uncertain. Finally, CNS involvement was found to be associated with worse odds of having a low PROPr score (OR=10.46, 95%CI 2.17, 50.37 [p=0.003]). Conclusions: Several PROs were worse in patients with histiocytic neoplasms as compared with the general population. We further found that CNS involvement was independently associated with worse HRQoL, regardless of the diagnosis type. Additionally, female sex was associated with worse MH scores, and bone involvement with worse PH scores. PROPr score was particularly low among patients with ECD. Future research is warranted to assess the impact of treatments on HRQoL, as well as the economic loss resulting from loss of productivity as highlighted by low PROPr scores.

Abstract Advances in the treatment of Langerhans cell histiocytosis (LCH) have resulted in a growing survivor population. There is a lack of data on long-term outcomes among adults with LCH. We conducted a retrospective record review of 219 adults (aged ≥18 years) with LCH. Most common presentation was multisystem (34.2%), followed by single-system pulmonary (32%), unifocal (28.3%), and single-system multifocal (5.5%) LCH. Risk organ involvement (the liver, spleen, or bone marrow) was seen in 8.7% of cases, and 40 of 88 (45.5%) tested cases were BRAFV600E. At a median follow-up of 74 months, 5-year progression-free survival (PFS) was 58.3% and estimated median PFS was 83 months. Median overall survival (OS) was not reached; 5- and 10-year OS rates were 88.7% and 74.5%, respectively. Risk organ involvement was associated with worse PFS (hazard ratio [HR], 4.5) and OS (HR, 10.8). BRAFV600E was not associated with risk organ involvement or survival. When compared with matched unaffected US population, individuals with LCH had a significantly higher risk of overall mortality (standardized mortality ratio [SMR], 2.66), specifically among those aged <55 years at diagnosis (SMR, 5.94) and those with multisystem disease (SMR, 4.12). Second cancers occurred in 16.4% cases, including diverse hematologic and solid organ malignancies. LCH-associated deaths constituted 36.1% of deaths and occurred within 5 years of diagnosis. After 5 years, non-LCH causes of death, including second cancers, chronic obstructive pulmonary disease, and cardiovascular diseases, predominated. Our study highlights, to our knowledge, for the first time, that adults with LCH experience early and late mortality from non-LCH causes and the need for development of targeted survivorship programs to improve outcomes.

1552 Background: Immune checkpoint inhibitor therapy-related (ICI) pneumonitis poses a significant challenge in patients with cancer. There is a need for predictive modeling to identify those at high risk of developing serious adverse events of ICI pneumonitis. Methods: A newly created database of cancer patients diagnosed with ICI pneumonitis seen at Mayo Clinic from 2014-2022 was utilized for exploratory data analysis. Modeling was used to determine if three clinical outcomes could be predicted, namely risk of death from: pneumonitis prior to starting immunotherapy, pneumonitis at time of diagnosis of ICI pneumonitis, and any cause at time of ICI pneumonitis diagnosis. The data consisted of 170 patients with annotation for 71 clinical features. Data was divided into features available at time of cancer diagnosis and at time of ICI pneumonitis diagnosis. Exploratory modeling was performed using the gradient boosting technique Xgboost (Chen, 2015) and conducted using k-fold balanced cross validation best practices as well as a train/test/validate schema with 70%/20%/10% data proportions, respectively. Model reverse engineering was done with Shapley statistics (Lundberg, 2017) to determine which features had the largest contribution per model. Once identified, only those highly weighted features were used for logistic regression analysis providing more reproducible predictions by decreasing model variance. Results: Risk of death from pneumonitis prior to starting immunotherapy was modeled with an area under the curve of the receiver operator characteristic (AUC-ROC) of 0.79 with the most contributory features including lymphocyte count, oxygen dependence, PFT values, and PD-L1 expression. Logistic regression produced an AUC-ROC of 0.87 (95% CI 0.72-1.0, p < 0.0013). Risk of death from pneumonitis at the time of ICI pneumonitis diagnosis was modeled with an AUC-ROC of 0.85 and the most contributory features were similar to the list described in the previous endpoint. Logistic regression produced an AUC-ROC of 0.89 (95% CI 0.81-0.96, p < 0.0001). Risk of death from any cause at the time of ICI pneumonitis diagnosis until follow up concluded by 12/2022 produced a model with an AUC-ROC of 0.75 with the most contributory features including supplemental oxygen, PFT values, basic laboratory values, and PD-L1 expression, among others. Logistic regression produced an AUC-ROC of 0.85 (95% CI 0.76-0.93, p < 0.0001). Conclusions: We demonstrate that commonly available clinical data can be used to identify patients at high risk of death from ICI pneumonitis. This study identified clinical features that were predictive in each scenario from which further concerted effort could produce a new clinical model to provide clinician decision support when considering immunotherapy. Further studies should be done to further elucidate feature interdependence and generalizability.

Abstract Erdheim-Chester disease (ECD) and Rosai-Dorfman disease (RDD) are rare non-Langerhans cell histiocytoses (non-LCHs), for which therapeutic options are limited. MAPK pathway activation through BRAFV600E mutation or other genomic alterations is a histiocytosis hallmark and correlates with a favorable response to BRAF inhibitors and the MEK inhibitor cobimetinib. However, there has been no systematic evaluation of alternative MEK inhibitors. To assess the efficacy and safety of the MEK inhibitor trametinib, we retrospectively analyzed the outcomes of 26 adult patients (17 with ECD, 5 with ECD/RDD, 3 with RDD, and 1 with ECD/LCH) treated with orally administered trametinib at 4 major US care centers. The most common treatment-related toxicity was rash (27% of patients). In most patients, the disease was effectively managed at low doses (0.5-1.0 mg trametinib daily). The response rate of the 17 evaluable patients was 71% (73% [8/11] without a detectable BRAFV600E achieving response). At a median follow-up of 23 months, treatment effects were durable, with a median time-to-treatment failure of 37 months, whereas the median progression-free and overall survival were not reached (at 3 years, 90.1% of patients were alive). Most patients harbored mutations in BRAF (either classic BRAFV600E or other BRAF alterations) or alterations in other genes involved in the MAPK pathway, eg, MAP2K, NF1, GNAS, or RAS. Most patients required lower than standard doses of trametinib but were responsive to lower doses. Our data suggest that the MEK inhibitor trametinib is an effective treatment for ECD and RDD, including those without the BRAFV600E mutation.

Abstract Infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can result in a cytokine storm that is associated with poor outcomes in patients. The activation of a NOD-like receptor pyrin domain containing 3 (NLRP3) inflammasome following SARS-CoV-2 infection has been shown to play a major role in inflammatory immune responses. In this study, we show that open reading frame 8 (ORF8) protein is abundantly secreted as a glycoprotein in vitro. The glycosylated ORF8 stimulates human CD14+/CD16+ monocytes to cause upregulation of proinflammatory cytokine production and cell surface marker expression within 24 hours. The data suggests that stimulation of human monocytes by ORF8 is not receptor mediated. Rather, ORF8 is likely phagocytosed by human monocytes. ORF8 then stimulates the monocytes by independently binding to the NACHT and LRR domains of the NLRP3 protein. Pharmacologic inhibition of NLRP3 significantly diminished the production of pro-inflammatory cytokines from ORF8-stimulated human monocytes. Finally, this study shows that the ORF8 protein is also secreted in patients with newly diagnosed coronavirus disease 2019 (COVID-19). Levels of ORF8 in the blood of patients diagnosed with COVID-19 correlated with disease mortality and trajectory of disease. ORF8 stimulation of monocytes causes pro-inflammatory cytokine production that leads to the development of severe COVID-19.