Editorial Commentary: Whole-Genome Sequencing of Clostridium difficile: Exquisitely Sensitive but Not Yet Optimally Applied

Clinical Infectious Diseases EDITORIAL COMMENTARY Whole-Genome Sequencing of Clostridium difficile: Exquisitely Sensitive but Not Yet Optimally Applied Dale N. Gerding1,2 1 Hines Veterans Affairs Hospital, Hines, and 2Loyola University Chicago Stritch School of Medicine, Maywood, Illinois (See the Major Article by Kumar et al on pages 746–52.) Keywords. Clostridium difficile infection; whole-genome sequencing; molecular typing; epidemiology; strain type. Received 8 December 2015; accepted 9 December 2015; published online 18 December 2015. Correspondence: D. N. Gerding, Research Service, Hines VA Hospital, 5000 S 5th Ave, Bldg 1, Rm 347, Hines, IL 60141 (). Clinical Infectious Diseases® 2016;62(6):753–4 Published by Oxford University Press for the Infectious Diseases Society of America 2015. This work is written by (a) US Government employee(s) and is in the public domain in the US. DOI: 10.1093/cid/civ1037 despite the fact that they analyzed different parts of the bacterial genome [5]. To date, WGS and similar studies have focused on elucidating CDI transmission events based on convenient cultured isolates obtained from patients diagnosed with CDI. An initial study utilized multilocus sequence typing (MLST), in which a selected number of housekeeping gene loci (7 in the case of C. difficile) are amplified and sequenced to yield sequence types (STs), which are grouped by evolutionary relationships into clades [2, 6]. MLST classification of C. difficile clades may be an accurate proxy for WGS analysis; however, the vast majority of STs are clustered in a single clade, clade 1, with only 1 or 2 STs in each of the remaining 4 clades. To make epidemiological transmission connections, these studies have employed hypothesized definitions of ward-based contacts between cases, representing “potential” transmission events [2]. Links were made when 2 CDI cases shared time on a ward [2, 7]. If CDI cases did not share time on a ward, indirect transmission via the contaminated ward environment was postulated up to 28 days after CDI patient discharge [1]. Furthermore, a putative “minimum infectious period” was defined as the time between the first sample from the potential donor and ward contact with the recipient up to 8 weeks maximum, and a putative “incubation period” as the time between this ward contact and the first sample in the recipient up to 12 weeks maximum [1, 2]. It should be noted that neither of these definitions should be construed as actual infectious periods or infection incubation periods, as there were no available data to identify actual transmission dates. Given the relative “newness” of WGS and the convenient availability of isolates from CDI patients for typing, the available WGS studies represent a good “first cut” at elucidating molecular relationships in the epidemiology of CDI in healthcare environments. To their credit, these studies have demonstrated a very high rate of introduction of new WGS types to the healthcare system, suggesting a very large pool of C. difficile strains being introduced by patients to these institutions [1]. The study was conducted at a time when healthcare-associated CDI rates were low, no dominant WGS types were identified, and an unexpectedly low rate of putative transmissions (38%) was found for these CDI patients [1]. Given the relatively low rates of CDI in these institutions at the time, perhaps this should not be surprising; it has been shown using REA typing that during periods of low CDI rates a multiplicity of strains are found causing CDI, whereas during periods of high CDI rates a dominant strain can be identified, presumably as a result of frequent healthcare transmissions [8]. In addition, a study of WGS typing of initial and recurrent CDI isolates has shown that there is a predominance of the same strain by WGS causing recurrence and meeting the definition of a relapse of infection [9]. In this issue of Clinical Infectious Diseases appears another report of the use of WGS that utilized the same aforementioned EDITORIAL COMMENTARY • CID 2016:62 (15 March) • 753 The development of exquisitely sensitive whole-genome sequencing (WGS) of Clostridium difficile has ushered in the opportunity to untangle the complex transmission epidemiology of C. difficile infection (CDI) that has heretofore been only partially available [1–3]. Few existing typing systems for C. difficile have the sensitivity of WGS to precisely identify specific molecular changes that constitute identification of multiple strains within a given typing group such as the epidemic NAP1/BI/027 group. A prior study of 42 NAP1/BI/027 isolates and 7 typing methods showed that only multilocus variablenumber tandem-repeat analysis (MLVA) and restriction endonuclease analysis (REA) were sufficiently discriminatory to distinguish among strains from different NAP1/BI/027 outbreaks [4]. Of these 2 methods, only MLVA could be used to construct a dendrogram of related strain types, as REA provides no information regarding evolution or genetic relatedness of REA groups or types within groups [4]. Subsequently, WGS and MLVA have been compared using the same set of C. difficile isolates and were confirmed as concordant in 58 of 61 (95%) investigations, ST1 isolates and an absence of any specimens from asymptomatic colonized patients. In my view, this study and previous WGS studies are also limited by the current hypothetical definitions for transmission that may or may not be indicative of actual transmissions. Now that exquisitely sensitive WGS typing is available, it is time to broaden the collection of isolates and improve the epidemiology to partner with the molecular technology. Current WGS studies leave too many unknowns regarding sources of transmission such as asymptomatic colonized patients and the environment, or even hands of healthcare workers. One small study has attempted to identify asymptomatic carriers and use WGS typing to identify transmissions, but results were inconclusive [3]. Prior studies using less sensitive typing methods have been designed to repeatedly culture stool or rectal swabs of patients while in hospital or on a specific hospital ward to identify asymptomatic stool colonization by C. difficile as well as cases of CDI [13–15]. In addition, environmental cultures can be taken periodically to assess contamination, although this can prove challenging if multiple colonies are obtained as all will need to be typed to assess which strains are in the environment. Nonetheless, it is time for more-rigorous epidemiologic studies to pair with the exquisite sensitivity of WGS. I look forward to the outcome of such studies to enhance our knowledge of the potentially very complex transmission epidemiology of CDI. Notes Financial support. The author is supported by grants from the US Department of Veterans Affairs Research Service. Potential conflict of interest. D. N. G. holds patents for the treatment and prevention of C. difficile infection licensed to ViroPharma (...truncated)