e24184 Background: Immune checkpoint inhibitors (ICIs) are increasingly used in early-stage cancers in the peri-operative setting. Yet while severe immune related adverse events (irAEs) are well-characterized in the metastatic setting, less is known about the presentations and outcomes of severe irAEs, particularly those requiring hospitalization, in neoadjuvant and adjuvant settings. Methods: This retrospective cohort study included patients hospitalized at Massachusetts General Hospital (MGH) and diagnosed with severe irAEs related to neoadjuvant ICI (regardless of whether they ultimately underwent surgery) or adjuvant ICI from 2/2011 to 4/2025. Patients were identified via institutional pharmacy records linked to hospital admission databases. Chart review was performed to collect demographics and characterize irAEs. Results: From 2/2011 to 4/2025, 87 patients were admitted to MGH for severe irAEs in the setting of neoadjuvant or adjuvant ICI. Of those, 29 (33.3%) had hospitalizations attributable to irAEs in the neoadjuvant setting and 58 (66.7%) to irAEs in the adjuvant setting. The most common cancers were melanoma (37.9%), lung (25.3%), and breast (16.1%). Most patients had stage II (19.5%) or III (67.8%) cancers, and most (83.9%) received anti-PD(L)1 therapy. Among the study cohort of 87 patients admitted with severe irAEs, the most common irAEs were endocrine (27.6%), gastrointestinal (21.8%), liver (19.5%), and cardiac (19.5%), and 29.9% of patients experienced multiple severe irAEs. Most patients had grade 3 irAEs (86.2%), but there was also a subset with grade 4 (10.3%) or 5 (2.3%) irAEs. Regarding irAE timing, most patients presented within 6 months of ICI initiation (66.7%), though 21.8% presented between 6-12 months, and 11.5% over 12 months after ICI initiation. Among 71 patients with non-endocrine irAEs, 94.4% received systemic steroids and 53.5% second-line immunosuppression. Premature ICI discontinuation due to irAEs occurred in 80.5% of 82 patients with non-fatal hospitalizations. Of 26 neoadjuvant patients with non-fatal hospitalizations, the majority (84.6%) underwent surgical resection (decisions to forego surgery were unrelated to irAEs); of those, 27.3% experienced an irAE-associated surgical delay (median delay 21 days, IQR 11 – 45.3). Among 82 total patients with non-fatal hospitalizations, 18.3% were re-challenged with ICI, 7.3% peri-operatively and 12.2% upon cancer recurrence. Of the 82 patients with non-fatal hospitalizations, 62.2% developed chronic irAEs (i.e. irAEs persisting ≥12 weeks). Conclusions: Severe irAEs requiring hospitalization represent a consequential source of morbidity among patients receiving neoadjuvant or adjuvant ICI and carry risks for mortality, surgical delay, premature ICI discontinuation, and chronic toxicity. Characterizing severe irAEs in the curative intent setting is crucial not only for acute management, but also for survivorship.

Kaplan–Meier curves for survival outcomes in patients with MSS/pMMR mCRC receiving ICI-based therapies, stratified by liver metastasis status at the time of ICI initiation. A, PFS: Patients without liver metastases had significantly higher 12-month PFS rates (12.8% vs. 1.1%; P = 0.034). The median PFS was 2.1 months (95% CI, 1.59–2.62) in those with liver metastases and 2.5 months (95% CI, 2.22–2.71) in those without liver metastases (HR, 1.68; 95% CI, 1.13–2.5; P = 0.009). B, OS: Median OS was 11.53 months (95% CI, 3.01–20.06) in those without liver metastases and 6.17 months (95% CI, 2.87–9.46) in those with liver metastases (HR, 2.03; 95% CI, 1.35–3.06; P < 0.001).

Demographic and clinical characteristics of patients with MSS/pMMR mCRC by liver metastasis status.

Regimens of immune checkpoint inhibitor-based therapies

Immune checkpoint inhibitors (ICI) have limited efficacy in microsatellite-stable (MSS) metastatic colorectal cancer (mCRC), potentially because of immunosuppressive mechanisms associated with liver metastases. The impact of liver metastases on survival outcomes was evaluated in a retrospective cohort of patients with MSS mCRC treated with ICI-based therapies at Mass General Brigham between January 2015 and December 2022. Patients were stratified by liver metastasis status at ICI initiation. The primary endpoint was progression-free survival (PFS); the secondary endpoint was overall survival (OS). A total of 132 patients were included, of whom 93 (70.5%) had liver metastases at ICI initiation. Most patients in both groups had received ≥2 prior lines of therapy. No significant differences were observed between groups for RAS/BRAF mutation status or tumor mutational burden. Patients without liver metastases demonstrated higher clinical benefit rates (46.2% vs. 16.1%; P = 0.001), longer median PFS (2.5 vs. 2.1 months; HR, 1.68; P = 0.009), and higher 12-month PFS rates (12.8% vs. 1.1%; P = 0.034). The median OS was also prolonged in patients without liver metastases (11.5 vs. 6.2 months; HR, 2.03; P < 0.001). No history of liver metastases was an independent favorable risk factor for PFS and OS in univariable and multivariable analyses. These findings indicate that liver metastases are associated with inferior survival outcomes in patients with MSS mCRC treated with ICI-based therapies, supporting the immunosuppressive role of liver metastases and underscoring the importance of stratifying patients by liver metastasis status to guide patient selection and optimize therapeutic strategies. SIGNIFICANCE:The limited efficacy of ICIs in MSS mCRC remains a major challenge. The association between liver metastases and inferior outcomes supports the liver's immunosuppressive role and suggests that liver metastasis status may guide patient selection and treatment optimization.

Subsequent treatments after immune checkpoint inhibitor-based therapy

Univariable and multivariable Cox regression analysis for OS.

Kaplan-Meier curves for survival outcomes in MSS/pMMR mCRC patients receiving ICI-based therapies, stratified by liver metastases status at baseline

Details of molecular characteristics and tumor mutational burden by timing of treatment

Univariable and multivariable Cox regression analysis for PFS.

2658 Background: Treatment of severe irAEs with multiple immunosuppressive therapies (ISTs) decreases the morbidity and mortality of these conditions. Nevertheless, the rates of and risk factors for infectious complications in this population are not known. Methods: We conducted a retrospective study of patients (pts) who received an immune checkpoint inhibitor (ICI) and experienced ≥1 irAE requiring treatment with corticosteroids along with at least two lines of steroid-sparing ISTs administered either concurrently or within 90 days of each other. We annotated all infections from ICI start until 90 days after IST. Opportunistic infections (OIs) were defined as herpesvirus (CMV, EBV, VZV, and HSV) and invasive fungal infections. Infection density was reported as number of infections per 1000 patient-days and graded as mild (not requiring treatment), moderate (requiring oral treatment), severe (requiring hospitalization or parenteral treatment), life threatening, or fatal. Risk factors were identified using univariable and multivariable Cox regression analysis adjusting for age at ICI initiation, sex, ICI regimen, ISTs, and steroid dose. Results: 175 pts (52% male, mean age: 66) with 238 irAEs and 417 ISTs were analyzed with a median follow-up of 367 days. The associated ICI regimens included αPD-1 (n = 81, 46%) and αPD-1/αCTLA-4 (n = 67, 38%). The most common irAEs were colitis (n = 67, 38%), hepatitis (n = 42, 24%), and myocarditis (n = 26, 15%). The most frequently used ISTs were mycophenolate mofetil (n = 92, 53%), infliximab (n = 77, 44%), and vedolizumab (n = 54, 31%). 103 pts (59%) developed 223 infections (median 2/pt, range: 1-8). 93 pts had 187 non-OIs. Of the 87 pts (50%) who had OI testing, 29 (33%) had 36 OIs, most commonly EBV DNAemia (n = 14, 16%) and CMV reactivation (n = 12, 14%). 6 pts had >1 OI, including 1 pt with CMV, EBV, and HSV. OI density significantly increased after starting ISTs for irAEs, but non-OI density was unchanged (Table 1). αCD20 use was associated with increased non-OI risk (HR: 10.58, 95% CI: 3.64-30.72, p < 0.001), while there was a trend towards increased non-OI risk with a max prednisone dose >100mg (HR: 1.74, 95% CI: 0.96-3.13, p = 0.07). In contrast, a max prednisone dose >100mg was associated with increased OI risk (HR: 2.89, 95% CI: 1.21-6.90, p = 0.017). 58 pts (33%) had severe or life-threatening infections, of whom 16 (9%) had OIs. 8 pts (5%) had fatal infections in this population. Conclusions: Use of multiple ISTs for severe irAEs is associated with increased rates of opportunistic infections as well as a 5% infection-related mortality rate. Patients requiring multiple lines of ISTs must be closely monitored for infectious complications, and prophylaxis should be considered when appropriate. Mean OI and non-OI density per 1000 patient-days while on ICI alone vs ISTs. ICI ISTs p-value OI 0.03 1.70 0.002 Non-OI 10.70 8.98 ns

Immune checkpoint inhibitors (ICIs) to treat cancer have revolutionized immuno-oncology. However, therapy is limited by frequent immune-related adverse events, including immune-mediated liver injury from checkpoint inhibitors (ILICI), which occurs in ∼1-17% of patients on anti-PD-1 and/or CTLA-4 therapy. While ILICI is often mild, it can cause severe hepatic dysfunction, delay cancer therapy, and require immunosuppressive treatments that may compromise anti-tumor immune responses. ILICI shares clinical and histologic features with autoimmune hepatitis (AIH). Therefore, the study of ILICI and AIH is important for understanding how immune tolerance is lost across a spectrum of inflammatory liver diseases. To characterize the cellular and molecular underpinnings of ILICI, we used single-cell and single-nuclei RNA sequencing with paired antigen receptor sequencing to characterize ∼300, 000 cells from paired liver and blood specimens from 23 patients (9 ILICI, 4 AIH, 3 controls on ICB, 7 controls not on ICB) and ∼150, 000 peripheral blood cells from an additional 28 control patients on ICB. ILICI was defined by a hepatocellular or cholestatic rise in liver function tests and centrilobular or panlobular inflammatory liver injury. Controls had drug-induced liver injury, hepatic steatosis, or venous outflow obstruction. Both AIH and ILICI patients showed marked expansion of clonally-related liver CXCR6Hi GZMKHi tissue-resident memory CD8 T cells expressing IFNG and GZMB and multiple CD4 T cell populations including cytotoxic GZMAHi effectors, CXCL13Hi effectors, and T regulatory cells. Compared to patients with AIH, those with ILICI showed a marked expansion of intravascular CX3CR1Hi CD8 T cells and a relative depletion of tissue dendritic cells while only patients with AIH had expanded plasma B cells. Analysis of matched blood samples from the same patient cohort revealed how cellular and transcriptional signatures in the liver were mirrored in circulating immune cells. Further analysis revealed that patients with ILICI and AIH had a marked interferon stimulated gene signature in a subset of hepatocytes, increased cholangiocyte progenitors, and increased periportal liver endothelial cells, which suggests significant liver parenchymal remodeling. Lastly, we identified a CXCL10-expressing cholangiocyte population that was expanded in both AIH and ILICI and was most pronounced in an individual with checkpoint inhibitor-associated cholangitis. In defining the cellular and transcriptional programs that are altered in ILICI and AIH, we have identified novel immune perturbations implicating PD-1 and CTLA-4 in liver immune tolerance. These discoveries offer promising avenues for the development of targeted diagnostics and therapeutics in the management of liver-related toxicities associated with immunotherapy. Molly Fisher Thomas, Thomas Chan, Neal Smith, Marc Sherman, Swetha Ramesh, Alice Tirard, John McGuire, Mazen Nasrallah, Kasidet Manakongtreecheep, Jessica Tantivit, Leyre Zubiri, Dejan Juric, Ryan Sullivan, Nir Hacohen, Genevieve Boland, Georg Lauer, Kerry Reynolds, Alexandra-Chloé Villani. Single-cell dissection of checkpoint inhibitor hepatitis reveals insights into immune tolerance in the human liver [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5812.

BACKGROUND:Immune checkpoint inhibitor-associated myocarditis (ICIM) carries high rates of morbidity and death, but clinical outcomes vary widely. Little is known about the clinical variables associated with long-term survival. METHODS:In this case-control study, patients diagnosed with ICIM at Massachusetts General Hospital between 2016 and 2022 were stratified into 3 groups based on length of survival after ICIM diagnosis: short-term (<30 days), intermediate-term (30-365 days), and long-term (>365 days). Baseline characteristics, immune checkpoint inhibitor regimens, laboratory values, ECG parameters, and ICIM treatments were analyzed to identify predictors of long-term survival. RESULTS:Among 35 patients with ICIM (median follow-up time, 8.3 months), there were 9 (25.7%) in the short-term survival group, 13 (37.1%) in the intermediate-term survival group, and 13 (37.1%) in the long-term survival group. Those in the short-term survival group were older (median age, 82 versus 68 for intermediate-term and 75 for long-term; P=0.003). Using logistic regression, long-term survival was associated with an interval from immune checkpoint inhibitor initiation to ICIM diagnosis ≥75 days (odds ratio, 5.4; P=0.043) and a troponin T decrement ≥42% by day 8 after immunosuppression initiation (odds ratio, 5.5; P=0.042). Using multivariate Cox regression modeling, troponin T≤1000 ng/L (hazard ratio [HR], 4.0; P=0.007) and neutrophil/lymphocyte ratio ≤4.4 (HR, 7.9; P < 0.001) were independently associated with longer survival. CONCLUSIONS:Time to onset of ICIM, multiple clinical tests, and responsiveness to immunosuppressive therapy were associated with long-term survival after ICIM. Consideration of these variables may help with risk stratification and immunosuppressive therapy individualization.

INTRODUCTION:Immune checkpoint inhibitor-associated acute interstitial nephritis presents significant clinical challenges. There are no reliable non-invasive biomarkers and kidney biopsy remains the gold standard for diagnosis. Prior studies have shown that urinary C-X-C-motif ligand 9 (CXCL9) is upregulated in patients with acute interstitial nephritis. However, its utility, specifically in patients with cancer treated with immune checkpoint inhibitors, is not well-understood. METHODS:We used proteomics followed by sandwich immunoassay to analyze urinary proteins among a multicenter cohort of prospectively enrolled participants with and without immune checkpoint inhibitor-associated acute interstitial nephritis. RESULTS:Among 79 participants receiving immune checkpoint inhibitors, proteomics identified urine CXCL9 as the top-performing urinary biomarker differentiating 38 patients with biopsy-proven acute interstitial nephritis from other forms of acute kidney injury. We validated these results using immunoassay in an expanded cohort of 116 patients, observing higher CXCL9 levels in immune checkpoint inhibitor-associated acute interstitial nephritis compared to several control groups. Urinary CXCL9 was strongly associated with immune checkpoint inhibitor-associated acute interstitial nephritis, with a receiver operating characteristic curve of 0.84, interquartile range [0.74, 0.93] when compared to other forms of acute kidney injury, and an even higher discrimination when compared with all control groups (0.90, [0.83-0.96]). CONCLUSIONS:Urinary CXCL9 demonstrated high discrimination for differentiating acute interstitial nephritis from other forms of acute kidney injury in participants on immune checkpoint inhibitor therapy. Our findings demonstrate the significant potential of this biomarker for non-invasive diagnosis of immune checkpoint inhibitor-associated acute interstitial nephritis.

Immune-related adverse events (irAEs) can cause severe morbidity and mortality, and they impair treatment with immune checkpoint inhibitors (ICI). Risk factors for irAEs are not well understood.We observed cases of patients having tumor deposits in their liver and lung during a workup of irAEs, which led us to hypothesize that the presence of tumor in an organ would increase the odds of developing severe irAEs in that organ. We then performed a retrospective cohort study that included patients who received an ICI for the treatment of cancer and were hospitalized between February 2011 and November 2021 at the Massachusetts General Hospital.We reviewed 384 patients hospitalized with concern for any irAE. A clinical diagnosis of ICI-related hepatitis occurred in 18% of patients with liver tumor deposits versus 8% of those without (OR 2.23, 95% CI (1.10 to 4.43), p=0.02). ICI-related pneumonitis occurred in 10% of patients with lung tumor deposits versus 4.4% of those without (OR 2.45, 95% CI (1.06 to 6.36), p=0.047). A combined analysis for liver and lung lesions demonstrated that the presence of tumor deposits in an organ increased the odds of having an irAE in that organ by over twofold (OR 2.31, 95% CI (1.34 to 3.99), p=0.002).Our results suggest that the presence of tumor deposits may represent a novel risk factor for severe irAEs in that organ.

Importance:The use of immune checkpoint inhibitors (ICIs) is increasing. Little is known about the frequency of late-onset immune-related adverse events (irAEs) and the patient-specific risk factors associated with their development. Objectives:To assess the incidence of persistent or de novo late-onset irAEs requiring hospitalization and identify patient factors associated with risk of late-onset irAEs. Design, Setting, and Participants:This retrospective observational cohort study conducted from January 2011 to October 2022 included patients who received ICIs and were hospitalized with irAEs at an academic medical center. Exclusion criteria included ICI therapy outside of the hospital system and no irAE diagnosis during admission. Data were analyzed from November 15, 2022, to January 8, 2025. Exposure:Late-onset irAEs. Main Outcomes and Measures:The main study outcomes were (1) incidence of irAE hospitalization at 0 to 6 months (early), more than 6 to 12 months (intermediate), and more than 12 months (late) after ICI initiation and (2) patient factors associated with risk of late-onset irAEs. Results:Among the 795 patients hospitalized with irAEs, the median age was 67.3 years (IQR, 58.3-74.8 years); 476 (59.9%) were male. Most patients (n = 517 [65.0%]) received anti-programmed death ligand 1 (PD-L1) and anti-programmed cell death 1 monotherapy, with the most common indications being melanoma (n = 335 [42.1%]) and lung cancer (n = 167 [21.0%]). The median time from start of ICI therapy to hospital admission was 2.7 months (IQR, 1.2-6.1 months), with 14.7% of patients (n = 117 of 795) presenting 6 to 12 months after initial ICI exposure and 10.8% of patients (86 of 795) presenting more than 12 months after initial exposure. The irAEs most likely to present late included those involving the kidney (10 of 32 [31.3%]) and hematologic (5 of 23 [21.7%]) organ systems. In univariate analysis, ICI type was significantly associated with the timing of hospital admission for irAEs; of the 517 patients receiving anti-PD-L1-based therapy, 13.5% (n = 70) presented late compared with 5.4% (9 of 167) receiving dual therapy with anti-cytotoxic T-lymphocyte-associated protein 4 (P < .001). Patients receiving perioperative ICI therapy were significantly more likely to be admitted at the intermediate interval (16 of 68 [23.5%]) compared with those with metastatic disease (87 of 678 [12.8%]) (P = .03). Timing of irAE was also significantly associated with active ICI exposure; among the patients presenting late, 7.4% (48 of 651) had received ICI therapy within the last 60 days compared with 26.4% (38 of 144) who had not had recent ICI exposure (P < .001). Conclusions and Relevance:The findings of this retrospective observational cohort study suggest that late irAEs are possible, with a subset of patients presenting years after the start of ICI therapy. Clinicians must remain vigilant for irAEs regardless of elapsed time from ICI therapy, especially as patients live longer and ICIs become more widely used. Future investigations are needed to better understand the risk factors for late-onset irAEs and the distinct immunologic pathways that underlie such events.