AbstractBackgroundPatients with cancer are particularly vulnerable to Clostridioides difficile infection (CDI). Guidelines recommend a two-step diagnostic algorithm to differentiate carriers from CDI; however, there are limited data for this approach while including other confounding risk factors for diarrhea such as radiation, cytotoxic chemotherapy, and adoptive cell based therapies.MethodsWe conducted a prospective, non-interventional, single center, cohort study of cancer patients with acute diarrhea and C. difficile, identified in stools by nucleic acid amplification tests (NAAT) and culture. Fecal toxin A/B was detected by enzyme immunoassay (EIA) and isolates were ribotyped using 16s rRNA fluorescent sequencing. Patients were followed for 90 days to compare outcomes according to malignancy type, infecting ribotype, and EIA status.ResultsWe followed 227 patients with a positive NAAT. Of these, 87% were hospitalized and 83% had an active malignancy. EIA was confirmed positive in 80/227 (35%) of patients. Those with EIA+ were older (60 ± 18 years vs 54 ± 19 years., P = .01), more likely to fail therapy [24/80 (30%) vs 26/147 (18%), P = .04] and experience recurrence [20/80 (25%) vs 21/147(14%), P < .05]. We found a low prevalence (22%) of ribotypes historically associated with poor outcomes (002, 018, 027, 56, F078-126, 244) but their presence were associated with treatment failure [17/50 (34%) vs 33/177 (19%), P = .02].ConclusionsWhen compared to cancer patients with fecal NAAT+/EIA−, patients with NAAT+/EIA+ CDI are less likely to respond to therapy and more likely to experience recurrence, particularly when due to ribotypes associated with poor outcomes.