In the February 2020 issue of The Oncologist, Corty and colleagues highlight the need to explore how the human microbiome and antibiotics influence the clinical effects of chemotherapy, focusing on gemcitabine-associated toxicities [1]. A previous study reported that gemcitabine is metabolized by cytidine deaminase (CDDL)–expressing bacteria, reducing antitumor activity [2]. Conversely, antibiotics might increase the effective dose of gemcitabine. Retrospective data from the MPACT trial [3] suggest that patients with pancreatic ductal adenocarcinoma (PDAC) treated with gemcitabine and concomitant antibiotics have an increased risk of hematologic and gastrointestinal adverse events [1].

Patients with PDAC have dismal prognosis, and identification of factors that can predict therapy response or toxicity is warranted. Hence, we value these findings, but we believe that some key points should be carefully discussed.

Various antibiotics are known for their gastrointestinal side effects, especially diarrhea, and may also cause hematological abnormalities. However, this study did not differentiate between specific antibiotic toxicity and antibiotic-enhanced gemcitabine toxicity [1]. To overcome these limitations, future trials including information on administered antibiotics are essential.

Moreover, due to the retrospective design and observational nature of the study, the authors could not identify or quantify tumor-related bacteria and related proteins. Interestingly, previous studies observed a diverse microbial community in patients who underwent pancreatic duct instrumentation [2, 4]. More than 15% of the patients in the gemcitabine arm of the MPACT trial underwent biliary stenting [3]. Future studies are needed to evaluate the intra- and extratumor bacterial presence and its relationship to medical procedures, as well as the potential of microbiome studies in liquid and solid biopsies to guide chemotherapy and antibacterial treatments.

Another critical point is the potential “patient bias,” because patients who required antibiotics suffered from an infection that could be caused by concomitant pathologies and poor clinical conditions. Moreover, during the MPACT trial, chemotherapy-related neutropenic fever and pneumonitis were common adverse events [3]. These can both result in hematologic abnormalities and require antibiotics.

The authors mention that an increased dose of gemcitabine could enhance clinical efficacy. However, adverse events seem a poor indicator of gemcitabine efficacy in the MPACT trial as well as previous trials [3]. Most recently, Riquelme and collaborators showed that PDAC microbial diversity was significantly higher in long-term survivors [5]. However, to evaluate clinical efficacy in patients with metastasis, prospective studies are required that determine objective endpoints such as reduction or stabilization of tumor size. Preclinical data on the latter are also urgently needed. Although Geller and collaborators performed bacterial analysis in human PDAC samples, their in vivo experiments were performed in colorectal cancer models [2].

Finally, further studies should evaluate the effects in patients (and preclinical models) treated with gemcitabine and nab-paclitaxel, because this combination might be differentially affected by antibiotics and cause different adverse events.

In conclusion, we are indebted to Corty and collaborators, because their study constitutes the first attempt to correlate clinical data on toxicity to the preclinical findings on the contribution of CDDL-expressing bacteria to gemcitabine resistance [1]. Beside more preclinical investigations exploring underlying molecular mechanisms, we look forward to additional studies on the effects of antibiotics in patients treated with gemcitabine and nab-paclitaxel, within prospective trials with well-annotated clinical data.

The authors indicated no financial relationships.