Improving infection reporting in hematology treatment trials

TO THE EDITOR:

We read with great interest the proposal for harmonizing the reporting of infections during treatment with bispecific antibodies in multiple myeloma by Ludwig et al,¹ published recently in Blood Advances.

The authors raised salient points about the importance of accounting for differences in the duration of follow-up and drug exposure between trials to reduce variation in reporting and proposed an approach to harmonize data reporting.¹ Combining this approach with reporting of all episodes of infection over the treatment period rather than a single, highest-grade infection episode resulted in a more holistic characterization of rates and temporal distribution of infections, which was not previously appreciated.¹ 

We agree with the critical need to reduce variations in the reporting of infection adverse events in trials of new therapies for hematologic malignancies. However, the drive to improve and standardize the reporting of infection adverse events should go beyond accounting for treatment or follow-up duration and inclusion of multiple episodes of infection. We and others have highlighted the ongoing deficiencies with the reporting of infection adverse events in clinical trials involving new-generation hematologic treatments.^2-4 Less than 50% of hematology clinical trials report infections grouped by type or source with the focus on reporting only clinical syndromes of pneumonia, sepsis, and neutropenic fever.² Less than 10% of trials report the type of pathogen responsible for infections, which negatively affects knowledge advancement and targeting of preventive and prophylaxis strategies.² On systematic reviews, no trials report the definition used by their investigators to diagnose or classify infection, raising concern about significant intrastudy and interstudy heterogeneity.² 

In a recent systematic review and meta-analysis of infection events in trials of bispecific antibody therapy for multiple myeloma, 21 of 89 publications were excluded early due to nonreporting of infection adverse events.⁵ Of the 16 studies included, 94% reported rates of severe infection but with limited reporting of further details. Only 44% reported some detail pertaining to clinical infective syndromes such as pneumonia or sepsis, whereas 56% of studies inconsistently reported the pathogens encountered, electing to report only opportunistic infections or infections of special interest to the investigators. Two of those 9 studies only reported on severe acute respiratory syndrome coronavirus-2 infections. Half of the studies did not report all the causative pathogens for fatal (grade 5) infections.⁵ Similarly, low rates of reporting of clinical syndromes and etiology of infections were noted in a systematic review of bispecific therapy for lymphoma, highlighting room for significant improvement.⁶ 

As part of a global effort, 7 major specialist societies involved in care and research of immunocompromised patients have recently published a consensus position statement outlining a comprehensive approach to improve reporting of infection adverse events across all studies involving immunocompromised patients, including hematology patients.⁴ There was recognition of the challenging balance between curating and reporting an ideal infection data set and the resources available to implement this in a busy clinical and research environment. Clinically useful and consistent data for analysis are the primary goal. As such, a core reporting data set consisting of category, site, severity, organism, and end points was developed as a minimum standard for the reporting of infection events across all study types. For clinical trials, additional data reporting is recommended, which includes clear reporting of infection-related exclusion criteria (eg, chronic infections), systematic grouping, and categories of infection, whereas the use of definitions used to diagnose and classify infections should be well described or cited.⁴ Critically, use of prophylaxis or preventive strategies such as immunoglobulin replacement in the trial protocol, which will have a significant impact on reported rates and patterns of infection events, should be clearly outlined. These recommendations are intended to encourage more complete reporting of data that are already available to trial investigators; data either prespecified in trial protocols, routinely collected as part of trial management, or reflect existing clinical practice by participating sites. There is recognition that reporting of adverse events is driven by regulatory requirements and ongoing engagement with regulatory agencies to streamline, and standardized reporting of infection adverse events is required to reduce the resources required by trial investigators.

This comprehensive approach to standardize the reporting of infection events in studies involving patients with hematologic malignancy is a vital first step in improving our collective research efforts to reduce the burden of infection in a high-risk patient group. There is need for all clinicians with a research interest in the prevention and management of infections in at-risk patient groups, including patients with hematologic malignancy, to work collaboratively to report these events consistently and cohesively to advance research and patient care in this field.

Contribution: B.W.T. conceived and wrote the manuscript with data input from G.K.R.; and M.M., N.J.M., and M.A.S. contributed to the editing of the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Benjamin W. Teh, Department of Infectious Diseases, Peter MacCallum Cancer Centre, 305 Grattan St, Melbourne 3000 VIC, Australia; email: ben.teh@petermac.org.