Incidence and Predictors of Respiratory Viral Infections By Multi-Plex PCR in Allogeneic Hematopoietic Cell Transplant (HCT) Recipients 50 Years and Older Including Geriatric Assessment (GA)

Introduction

Community respiratory viruses (CRV) are important agents of morbidity and mortality after HCT. Although several studies have included subsets of patients 50 years and older, no studies have specifically investigated the incidence of CRV infections in older patients nor the influence of health impairments as measured by Geriatric Assessment (GA).

Methods

We retrospectively reviewed patients undergoing allogeneic HCT, age 50 years or older at HCT, and who completed GA prior to HCT. The GA instruments and thresholds for prognostic marker analyses followed our prior published results (Muffly L et al, Haematologica 2014). We limited the cohort to transplant years 2009-2013 to coincide with the availability of a multiplex PCR for CRV covering respiratory syncytial virus (RSV), parainfluenza virus (PIV) 1-4, influenza A and B, Human metapneumovirus (HMPV), coronavirus (hCoV), entero/rhinovirus (human EV), and adenovirus (Adv). Testing was performed at the discretion of the treating physician as clinically indicated through either nasal swab and/or bronchoalveolar lavage. We analyzed cumulative incidence of CRV after HCT at day 100 and one year, outcomes, and univariate risk factors for infection.

Results

The baseline characteristics of the 121 evaluable patients included: median age 58 years (50 - 73); AML (39%); MDS (17%); not in remission/response (39%); haplo-cord (21%); myeloablative conditioning (16%); and T-cell depletion by ATG/alemtuzumab (95%). Thirty-three first-episode CRV infections occurred among 121 evaluable patients for a cumulative incidence of CRV by day 100 of 10.5% (95% CI: 5.5 – 17.2) and 1 year of 24.8% (CI: 16.9-33.4%). Multi-plex PCR identified the following CRV infections: influenza A or B (n=4, 12%), RSV (n=7, 21%), PIV (n=5, 15%), human EV (n=11, 33%), hCoV (n=1, 3%), Adv (n=3, 9%), and HMPV (n=2, 6%). Co-infections occurred in roughly half of the cases and commonly included cytomegalovirus, Pseudomonas aeruginosa, Aspergillus fumigatus and Clostridium difficile. Morbidity and mortality were restricted to those who developed lower respiratory tract infections (LRTI). The outcomes of LRTIs are presented in table 1. Viruses were grouped within the table according to characteristic disease course as previously described (Wolfromm et al, BBMT 2014). Twenty-two (67%) patients with CRV infection required hospitalization with a median length of stay of 11 days. Importantly, CRV directly contributed to death in 7 patients (32% of LRTIs). Of these, 5 patients had a bacterial co-infection and 6 of the 7 were receiving steroids at CRV onset. CRV infection showed no association with GA measures of high HCT-CI (p=.34), high CRP (p=.86), lower performance status (p=.66), impairments in instrumental activities of daily living (p=.96), frail (p=.83), low self-report physical function (p=1.0), and low self-report mental function (p=.51). The one-year incidence of CRV was non-significantly lower for albumin below 3.5 g/dL (11% vs. 28%, p=0.17), slow walk speed (11% vs 31%, p=.06) and age 60+ versus 50-59 (20% vs 28%, p=.36). The limited sample size precluded an analysis of CRV associated morbidity by GA.

Conclusions

The incidence of CRV infection of 25% among older allogeneic HCT recipients by multi-plex PCR appears similar or slightly lower than previously reported data for HCT in general. Health impairments by GA did not translate into heightened risk of CRV infection. Most LRTI related deaths occurred in patients receiving steroids at the onset of infection. Larger prospective studies will be needed to determine if patient health status influences CRV related morbidity in older adults.