Our cohort included 61 consecutive patients admitted with ICI-myocarditis. Mean age was 68±13 years, 34% were female and median follow-up was 113 [47-304] days. All patients were promptly hospitalized upon suspicion of ICI-myocarditis with diagnosis confirmed via endomyocardial biopsy or cardiac MRI. Each patient was serially assessed for circulating biomarker (CK/cTnI/cTnT) of myotoxicity; multiple biomarkers were assessed within the first 3 days (median number[IQR] of tests/patient for CK: 3[2-4], cTnI: 3[1-4], cTnT: 4[3-5]). The clinical and demographic characteristics of this cohort are displayed in Table-2. Most patients had definite ICI-myocarditis (49/61, 80%) as determined by a diagnostic level of certainty; the rest had probable myocarditis.12 A total of 64% of patients received anti-PD1 monotherapy, 21% anti-PDL1 monotherapy and 15% received a combination of anti-PD1 and anti-CTLA4. Most patients suffered from non-small lung cancer (39%) or malignant melanoma (21%). Most patients were symptomatic at initial presentation (44/61; 72%) contrasting with a small subset being identified asymptomatically as part of systematic screening strategy (17/61; 28%). A total of 24/61 (39%) patients developed at least one MACE (MACE events detailed in Table-2). Overall mortality and ICI cardio-myotoxicity related death occurred in 26/61 (43%) and 10/61 (16%) patients, respectively. Cardio-myotoxicity related deaths occurred earlier after ICI-myocarditis diagnosis versus non-related deaths (11[9-29] vs. 132[85-344] days, p<0.0001). Causes of death are detailed in Table-2.
Time course of cTnT, TnI and CK in patients with ICI-myocarditis
At index admission, cTnT, cTnI and CK were increased in most patients: 51/52 (98%), 28/34 (82%, p=0.009 vs. cTnT), 33/48 (69%, p<0.0001 vs. cTnT), respectively. Within 72h after admission, maximum blood concentrations expressed as multiples of URL were higher for cTnT (median=28[10-64]) compared to cTnI (median=12[5-59]; p=0.03 vs. cTnT) and CK (median=6[1-23]; p<0.0001 vs. cTnT). These biomarkers were serially tested during hospitalization and time-dependent concentration changes in the different time periods following presentation are displayed in Figure-1. Peak values were observed for cTnT on day 7[3-15], for cTnI on day 4[2-9] (p=0.02 vs. cTnT, paired Wilcoxon-test), and for CK on day 1[1-6] (p=0.05 vs. cTnT, paired Wilcoxon-test) after initial ICI-myocarditis diagnosis. Figure-1 shows a rapid decline in maximal (Figure-1A), minimal (Figure-1B) and median (Figure-1C) CK and cTnI levels during the early phase of the ICI-myocarditis (weeks) in contrast to a more prolonged elevation of TnT lasting several months. Minimal circulating levels of cTnT, cTnI and CK were below URL between day 15-30 after ICI-myocarditis diagnosis in 2%, 66% and 83% of cases; and in 11%, 84%, 98% between day 31-90 (p<0.0001 at all times, more extended follow-up data are shown in Figure-1E), respectively. In patients in which measured biomarkers normalized during the follow-up, the median time to first value below URL was longer for cTnT (55[39-141]days), compared to cTnI (23[9-34]days), and CK (9[5-15]days). Maximum discrepancy between ratio of cTnT/URL over cTnI/URL (maximal ratio= 19.88 [5.7-77.6]) during follow-up was identified between day 15-30 after diagnosis (Figure-1D).
Predictors of MACE in ICI-myocarditis
We next investigated the prognostic value of cTnT, cTnI and CK at index admission and during the course of their surveillance. Characteristics of ICI-myocarditis patients with MACE compared to patients without MACE during follow-up are shown in Table-2. The maximal cTnT/URL value measured within 72h upon diagnosis performed best in predicting MACE (AUC=0.82) during follow-up compared to CK/URL (AUC=0.74) and cTnI/URL (AUC=0.67) (Figure-2A). An analysis of deviance showed consistency and significant superior performance of models including cTnT/URL with no improvement by addition of CK/URL, age, sex, and cTnI/URL (Figure-2B). Consistently, the addition of cTnT/URL to models containing cTnI alone, CK alone, and both cTnI/CK significantly increased their performance (p=0.0003, p=0.0004, and p=0.003, respectively). Using best threshold detection based on ROC analysis, we found a maximal cTnT/URL value within ≤72 hours of admission for ICI-myocarditis above >32 to be the best predictor for MACE during the first 100 days (71% vs. 13%, p<0.0001, Cox regression hazard-ratio: 9.4, 95% confidence interval: 3.1,28.3, Figure-2C). Results were similar when missing data were imputed (see methods, Figures-3A,B). MACE in the 4/31 patients with cTnT/URL<32, were non-fatal and occurred after hospitalization discharge except for one ventricular tachycardia diagnosed at the time of admission for ICI-myocarditis. Notably, cTnT/URL values (58[28-187]) were abnormal in all patients before the occurrence of first MACE (22/22 patients) while cTnI/URL (14[2-38]) and CK/URL (4[1-13]) values were normal in 3/21 (14%) and 6/24 (25%) of patients, respectively (p<0.0001 Fisher’s exact test) (Figure-4B). The time interval from last blood sample to occurrence of first MACE were 1[1-5], 3[1-4], 1[1-6] days before MACE for cTnT, cTnI and CK, respectively. Selected kinetic changes of declining or normalizing CK and cTnI despite persistently high or even increasing cTnT levels at the time of first MACE in 4 ICI-myocarditis patients are shown in Figure-4A. These cases highlight the discrepant prognostic information of kinetic changes of cTnI/CK versus cTnT in ICI-myocarditis (Table-2).
External validation of cTnT, cTnI diagnostic value in patients with ICI-myocarditis
The external validation cohort comprise 35 patients from an international registry (cases described in the Franco-German cohort were not included) who had both cTnI and cTnT measurements within 72 hours of admission (Figures-5A & B). While 66% patients (23/35) with ICI-myocarditis had an increased cTnI/URL on admission, the respective percentage was 91% (32/35) for cTnT/URL (p=0.003, Figure-5E). This discrepancy also persisted for peak troponin values (cTnI/URL>1 in 21/30 (70%) vs. TnT/URL>1 in 29/30 (97%) (p= 0.001, Figure-5F). When this external validation cohort was queried for cases with either cTnI or cTnT available; initial cTnT was increased in 115/126 (91%) vs. 127/153 (83%) for cTnI (p=0.04, Figure-5C).