Diagnostic evaluation of immune checkpoint inhibitor (CPI) colitis: The role of CT scan

821

Background: CPI therapy has expanded rapidly in recent years and represents a major advancement in the treatment of many cancers, including hepatocellular carcinoma, gastric cancer, and colon cancer. However, these therapies are associated with significant toxicities. CPI colitis is one of the most common toxicities and can be fatal, especially when not diagnosed and treated promptly. The current gold standard for diagnosis is endoscopy with biopsy, an invasive procedure that is resource- and time-intensive. CT has emerged as a possible alternative. The primary objective of this study is to identify the diagnostic performance of CT in the evaluation of CPI colitis. Methods: With IRB approval, we conducted a retrospective cohort study of patients who received CPI therapy between 2009-2019 across a single healthcare system. Patients were included if they underwent both abdominal CT and upper/lower endoscopy with biopsy within 72 hours of each other. We reviewed the electronic medical record to identify possible cases of colitis based on either CT or pathology. All cases were labeled as either true positive or false positive based on pathology. We examined clinical characteristics, including CTCAE grade and treatment received. We calculated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of CT for diagnosing CPI colitis when compared to the gold standard of tissue diagnosis. Results: Of the 4,474 patients screened, 141 met inclusion criteria. Average age was 63 years (23 – 91); 43% were male. Most common tumor types were melanoma (36%) and NSCLC (20%). Seventy-four percent of patients were treated with anti-PD-1/PD-L1 monotherapy. Forty percent had signs of colitis on CT scan and 59% had biopsy-proven CPI colitis. Sensitivity and specificity of CT were 51% and 74%, respectively. PPV of CT was 74% and NPV was 51%. Of those with confirmed CPI colitis, 78% had symptoms that were classified as grade 3 or above. Seventy-three percent received IV steroids and 38% received infliximab. Conclusions: CT demonstrates moderate specificity and PPV and remains an important diagnostic test but does not replace endoscopy/biopsy in the evaluation of CPI colitis.

Flu vaccination rate of patients with severe immune-related adverse events

e18234

Background: Infection with influenza in adults with cancer carries an increased risk of morbidity and mortality. Vaccination against seasonal influenza (Flu-V) can decrease the incidence of influenza, shorten its course, and reduce influenza-associated morbidity. Recent data has suggested that the administration of the Flu-V to patients on an ICI leads to an exaggerated inflammatory response and an increased risk of irAE. However, this trend was demonstrated in a small cohort of patients with lung cancer. Current recommendation for annual Flu-V in patients treated with ICI is unclear and literature about safety is limited. We compared rates of Flu-V for patients on ICI admitted with severe toxicity vs those patients on ICI who were admitted for reasons other than toxicity. We also evaluated rate of Flu-V among oncology patients who had received non-immunotherapy modalities. Methods: We retrospectively evaluated patients treated with ICI who were admitted to Massachusetts General Hospital from February 5, 2011- June 12, 2017. Patients received ipilimumab, pembrolizumab, nivolumab, atezolizumab, durvalumab, avelumab, or a combination in treatment of an advanced solid tumor malignancies including melanoma, NSCLC, SCCHN. Admissions due to irAE were confirmed by review of clinical, radiologic, and pathologic features. Flu-V status was determined by rigorous chart review. Nearest neighbor matching was used to create a control group of cancer patients treated with non-ICI modalities. Descriptive statistics compared rates and timing of Flu-V relative to admission. Statistical significance was determined using Fischer’s Exact Test, p < 0.05. Results: Of 540 patients on ICI, 28% were admitted for irAE, 72% had a non-irAE reason for admission. The rate of Flu-V in the flu season prior to admission for irAE group was lower than for non-irAE (18.5% vs 29.6%; p value = 0.01). There were no differences in vaccination rates within ≤30 days (2.7% vs 3.6%, p = 0.80), ≤90 days (4.0% vs 9.3%, p = 0.05), or ≤180 days of admission (11.9% vs 18.5%, p = 0.07). Flu-V rate overall in patients on ICI was 26.5%. In comparison, Flu-V rate in the nearest neighbor non-immunotherapy oncology patients was 67% (n = 101). Conclusions: Flu-V rates were much lower in patients treated with ICI compared to patients treated with non-ICI modalities. We did not see a higher rate of Flu-V in patients admitted with irAE compared to non-irAE which suggests that Flu-V and severe irAE may not be linked in clinical practice. Additional studies are needed, but Flu-V in patients on ICI holds potential to improve care.

Clinical outcomes of patients with stage IV cancer receiving immune checkpoint inhibitors in the inpatient setting

6634

Background: Immune checkpoint inhibitors (ICI) represent a major leap in the treatment of many cancers. Use has rapidly expanded in recent years, yet it is unknown whether hospitalized patients, who are often sicker than those who were studied in clinical trials, derive benefit from ICI. The primary objectives of this study were to characterize the clinical features and outcomes of inpatients receiving ICI at a single institution, and to identify predictors of survival. Methods: After IRB approval, we conducted a retrospective chart review of inpatients with Stage IV solid tumors receiving ICI between 2015 – 2018 at a tertiary care referral hospital. Patients receiving ICI on clinical trial were excluded. We examined the clinical characteristics, readmission rate, and post-discharge survival. We then conducted a Cox multivariable regression analysis to identify predictors of post-discharge survival. Results: A total of 103 patients with Stage IV solid tumors were treated with ICI as inpatients between 2015 – 2018. Average age was 57 years (range = 26 to 85); 57% were male; 27% had ECOG performance status (PS) 3-4; average Charlson Comorbidity Index score was 8.3. Most common tumor types were melanoma (35%) and lung (22%). Seventy-six percent began ICI as an inpatient and 24% received ICI as continuation of outpatient therapy. Seventeen percent experienced an immunotherapy related adverse event, most commonly colitis and pneumonitis. The 30 day readmission rate was 41%. The median post-discharge survival was 31 days; 47% of patients died during admission or within 30 days of discharge; 14% survived more than 6 months. Factors predictive of shorter post-discharge survival were PS of 3-4 relative to PS 0-2 (HR 2.0, p < 0.004), and lung cancer (HR 2.0, p < 0.024) and other tumor types (HR 2.1, p < 0.004) relative to melanoma. Conclusions: While the majority of inpatients receiving ICI died during admission or within 30 days of discharge, a subset of patients with stage IV disease were alive at 6 months. Tumor type and ECOG PS predict post-discharge survival and may be used to identify inpatients more likely to benefit from ICI. These novel findings, which are unique to a single institution, require additional validation.

Factors associated with severity of immune checkpoint inhibitor gastroenterocolitis requiring hospitalization in melanoma patients

81

Background: Immune checkpoint inhibitors (ICIs) have resulted in a subset of patients experiencing durable responses in solid and liquid malignancies. ICIs can be associated with adverse events, including gastroenterocolitis (GEC), that require immunosuppression and even hospitalization. Currently, no data determine the severity of patients hospitalized for ICI-related GEC. Our primary objective was to identify candidate surrogate endpoints that may predict the severity of ICI-related GEC; our secondary objective was to identify predictors of glucocorticoid (GC) response. Methods: In a retrospective cohort study, we identified melanoma patients who developed ICI-related GEC requiring hospitalization at Massachusetts General Hospital between 6/1/11 and 12/31/17. We extracted clinical, laboratory, radiographic, and pathological data; linear regression was used to estimate trends. Additional subgroup analyses were performed. Results: 69/1842 (3.7%) melanoma patients treated with ICI developed GEC requiring hospitalization (total 98 admissions). Mean age was 64 +/- 13; 42 (61%) were male. Readmission rate was 21/69 (30.4%); 6/21 (28.6%) required multiple readmissions. 90/98 (92%) were confirmed by histopathologic examination. 26/69 (37.7%) responded to GCs; 43/69 (62.3%) required second-line immunosuppression (e.g. TNFi) and/or operative intervention. ECOG PS (at initial ICI administration) was associated with response of GEC to GCs (p = 0.04). Lymphocyte count (p = 0.03), % lymphocyte count (p = 0.02), and LDH (p = 0.004) on admission were independently associated with GEC requiring second-line immunosuppression. Conclusions: Admissions for ICI-related GEC are infrequent, but they are associated with a high readmission rate and the need for second-line immunosuppression. We show that high ECOG PS at time of ICI administration may be a predictor of GC response in ICI-related GEC requiring hospitalization. We also propose that low % lymphocyte count and high LDH may serve as clinical biomarkers of severity in ICI-related GEC requiring hospitalization. Our findings suggest the need for further research and validation of these proposed biomarkers.

Posttraumatic stress disorder among black Vietnam veterans

Because of racism in the military and racial and social upheaval in the United States during the Vietnam War years, as well as limited opportunities for blacks in the postwar period, black veterans of the Vietnam War often harbor conflicting feelings about their wartime experiences and have difficulty rationalizing brutality against the Vietnamese. As a result, black veterans suffer from posttraumatic stress disorder (PTSD) at a higher rate than white veterans. Diagnosis and treatment of PTSD in black veterans is complicated by the tendency to misdiagnose black patients, by the varied manifestations of PTSD, and by patients' frequent alcohol and drug abuse and medical, legal, personality, and vocational problems. The author presents his and others' recommendations about ways to treat black veterans with PTSD.

Dyspituitarism, with Hydrocephalus

Images

Juvenile Tabes Dorsalis with Negative Wassermann Reaction in Blood and Cerebro-spinal Fluid

Images