Survival Among Veterans Receiving Steroids for Immune-Related Adverse Events After Immune Checkpoint Inhibitor Therapy

Abstract

Importance: Systemic steroids are commonly used to manage immune-related adverse events (irAEs), but it remains unclear whether they may undermine immune checkpoint inhibitor (ICI) therapy outcomes. Few studies have assessed the impact of steroid timing and its association with continuation or cessation of ICI therapy.

Objective: To characterize how systemic steroids and steroid timing for irAEs are associated with survival in patients receiving ICI therapy.

Design, setting, and participants: This multicenter retrospective cohort study encompassed veterans receiving ICI for cancer between January 1, 2010, and December 31, 2021. Data analysis was conducted September 8, 2023.

Exposures: Identifiable primary diagnosis of cancer. Patients were categorized into 3 cohorts: those receiving no steroids, systemic steroids for irAEs, and steroids for non-irAE-associated reasons. All eligible patients received 1 or more doses of an ICI (atezolizumab, avelumab, cemiplimab, durvalumab, ipilimumab, nivolumab, or pembrolizumab). Eligible patients in the steroid group received at least 1 dose (intravenous, intramuscular, or oral) of dexamethasone, hydrocortisone, methylprednisolone, prednisone, or prednisolone. Steroid use at baseline for palliation or infusion prophylaxis or delivered as a single dose was deemed to be non-irAE associated. All other patterns of steroid use were assumed to be for irAEs.

Main outcomes and measures: The primary outcome was overall survival, with a 5-year follow-up after ICI initiation. Kaplan-Meier survival analyses were performed with pairwise log-rank tests to determine significance. Risk was modeled with Cox proportional hazard regression.

Results: The cohort consisted of 20 163 veterans receiving ICI therapy including 12 221 patients (mean [SD] age, 69.5 [8.0] years; 11 830 male patients [96.8%]; 9394 White patients [76.9%]) who received systemic steroids during ICI treatment and 7942 patients (mean [SD] age, 70.3 [8.5] years; 7747 male patients [97.5%]; 6085 White patients [76.6%]) who did not. Patients with an irAE diagnosis had significantly improved overall survival (OS) compared with those without (median [IQR] OS, 17.4 [6.6 to 48.5] months vs 10.5 [3.5 to 36.8] months; adjusted hazard ratio, 0.84; 95% CI, 0.81-0.84; P < .001). For patients with irAEs, systemic steroids for irAEs were associated with significantly improved survival compared with those who received steroids for non-irAE-related reasons or no steroid treatment (median [IQR] OS, 21.3 [9.3 to 58.2] months vs 13.6 [5.5 to 33.7] months vs 15.8 [4.9 to not reached] months; P <.001). However, among those who received steroids for irAEs, early steroid use (<2 months after ICI initiation) was associated with reduced relative survival benefit vs later steroid use, regardless of ICI continuation or cessation following steroid initiation (median [IQR] OS after ICI cessation 4.4 [1.9 to 19.5] months vs 16.0 [8.0 to 42.2] months; median [IQR] OS after ICI continuation, 16.0 [7.1 to not reached] months vs 29.2 [16.5 to 53.5] months; P <.001).

Conclusions and relevance: This study suggests that steroids for irAE management may not abrogate irAE-associated survival benefits. However, early steroid administration within 2 months of ICI initiation is associated with shorter survival despite continuation of ICI therapy.

Figure 1.. Patient Enrollment Flowchart

The figure shows patients prescribed immune checkpoint inhibitors (ICIs) and the allocation of patients who did or did not use steroids. Percentages represent proportion of patients found to have an International Classification of Diseases, Ninth Revision (ICD-9) or International Classification of Diseases, Tenth Revision (ICD-10) code for an immune-related adverse event (irAE). Patients with uncertain ICI treatment status included those without documented ICI dates, those participating in research studies and potentially receiving placebos, and those who may have only undergone immunohistochemistry staining. The non–irAE steroid subgroups are not mutually exclusive, and some patients in these subgroups nevertheless had an ICD-9 or ICD-10 irAE indication.

Figure 2.. Kaplan-Meier Curve Demonstrating Overall Survival in Patients With and Without Immune-Related Adverse Event (irAE)–Related Indications

Survival probability is shown for patients with (+irAE) and without (-irAE) International Classification of Diseases, Ninth Revision or International Classification of Diseases, Tenth Revision codes for irAEs. The median (IQR) OS was 10.5 (3.5-36.8) months for those without irAE codes and 17.4 (6.6-48.5) months for those with irAE codes (adjusted hazard ratio, 0.84; 95% CI, 0.81-0.84; P < .001).

Figure 3.. Kaplan-Meier (KM) Curve Showing Association of Steroids With Overall Survival in the Full Veterans Cohort

Survival probability is shown for overall steroid (S) and nonsteroid (NS) groups (A) and for the following subgroups: patients in the nonsteroid group who nevertheless had an International Classification of Diseases, Ninth Revision (ICD-9) or International Classification of Diseases, Tenth Revision (ICD-10) codes for immune-related adverse events (irAEs; NS:+irAE), patients in the nonsteroid group who did not had an ICD-9 or ICD-10 code for irAEs (NS:-irAE), patients receiving steroids for irAEs who had an ICD-9 or ICD-10 code for irAEs (S:irAE+irAE), patients receiving steroids for irAEs who did not have had an ICD-9 or ICD-10 code for irAEs (S:irAE-irAE), patients receiving steroids for non–irAE reasons who nevertheless had an ICD-9 or ICD-10 code for irAEs (S:other+irAE), and patients receiving steroids for non–irAE reasons who did not have an ICD-9 or ICD-10 code for irAEs (S:other-irAE) (B) In Panel A, the adjusted hazard ratio was 0.88 (95% CI, 0.83-0.93; P < .001). The adjusted significance threshold was P < .003.

Figure 4.. Kaplan-Meier Curve Demonstrating Association of Steroid Timing and Immune Checkpoint Inhibitor (ICI) Continuation Status With Overall Survival (OS) in Patients With an Immune-Related Adverse Event (irAE) Pattern of Steroid Use

Early steroid use was defined as steroids administered less than 2 months after ICI initiation, whereas late steroid use was defined as steroids administered 2 or more months after ICI initiation. ICI status was defined as continuation or cessation of ICI treatment after steroid initiation. Solid lines reflect the hazard ratio whereas the shaded areas reflect the 95% CIs. The median (IQR) OS was 4.4 (1.9 to 19.5) months for early steroids and ICI cessation, 16.0 (7.1 to not reached) months for early steroids and ICI continuation, 16.0 (8.0 to 42.2) months for late steroids and ICI cessation, and 29.2 (16.5 to 53.5) months for late steroids and ICI continuation. The adjusted significance threshold was P < .008.