In this edition of “How I will treat my next patient,” I examine two recently published efforts to enlighten our sensitivity to the seriousness of immune-related adverse events (IrAEs) in patients on immune checkpoint inhibitors (ICIs) and the effect of delays in initiating systemic adjuvant therapy on the long-term outcomes of patients with resected pancreatic cancer.
IrAEs requiring hospitalization
Investigators led byof the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University, Baltimore, reviewed a 6-month audit of inpatient oncology admissions of solid-tumor patients who had ever received ICIs and ascertained the prevalence of hospitalization for management of IrAEs (J Oncol Pract. 2019 Aug 6. ). To determine that an IrAE had occurred they required: consensus among two oncologists, clinical improvement with immune-directed therapy, exclusion of alternative diagnoses or pathologic confirmation of an IrAE, or chronic management of an IrAE for more than 6 months.
The bottom line: They found a cumulative incidence of a confirmed IrAEs among hospitalized ICI-treated solid tumor patients of 23%. As expected, the majority (65%) were grade 3-4 in severity. In total, 91% required multidisciplinary management, and 65% improved or resolved. But 87% of patients never received an ICI again.
Patients with preexisting autoimmune disease (25% of patients, although they included hypothyroidism in that group) were not more vulnerable to an IrAE with ICI therapy (odds ratio, 1.0; 95% confidence interval, 0.3-4.0). Not unsurprisingly, the median age was higher for ICI-treated patients who were admitted for IrAEs than for those not admitted (68 years vs. 59 years; OR, 5.4; 95% CI, 1.6-17.8), and more admitted patients had received combination ICIs than single agents (OR, 6.8; 95% CI, 2.0-23.2).
The median time from beginning ICIs to an IrAE-related hospitalization was 64 days, and the median number of ICI doses was one, with a wide range for both days and doses. The authors were quick to comment that this is a small, academic, single-institution survey over a brief period of time and that the generalizability of the results is uncertain.
What this means in practice
This publication changes very little for most practicing oncologists, but it does reinforce that ICI therapy can cause unpredictable, severe IrAEs. Clinical markers for selecting patients at highest risk are imperfect. As with chemotherapy, the patients we worry about the most – older individuals and patients treated with drug combinations – are, in fact, the ones we should be worrying about the most.
In view of the potential severity and impact of IrAEs, research efforts should place equal priority on identifying biomarkers of toxicity, such as tumor mutation burden, and biomarkers of efficacy (JAMA Oncol. 2019 Aug 22.