Viral Load in Hematopoietic Cell Transplant Recipients Infected with Human Rhinovirus Correlates with Burden of Symptoms

Background: Human rhinovirus (HRV), the virus predominantly associated with the common cold, is the most common virus detected from respiratory specimens in hematopoietic cell transplant (HCT) recipients, largely due to the development of new sensitive molecular viral detection techniques. In immunocompetent children, HRV viral load from nasal specimens correlates with disease severity, namely the development of pneumonia, hospital admission, and oxygen requirement. Prospective studies describing the relationship between viral load and symptoms in immunocompromised hosts, however, are limited.

Methods: We conducted a prospective surveillance study of allogeneic HCT recipients, in which upper respiratory samples and symptom surveys were collected weekly for the first 100 days following HCT, then every 3 months or with new respiratory symptoms thereafter for 1 year. Symptom surveys included assessment of 11 respiratory symptoms including rhinorrhea, sinus congestion, post-nasal drip, dyspnea, cough, wheezing, sputum production, pharyngitis, sneezing, watery eyes and ear pain. Respiratory samples were tested by real-time RT-PCR targeting the highly conserved 5′ noncoding region. RT-PCR cycle threshold (CT) values were used as a proxy measure of viral load. Only patients with symptom survey results within 3 days of initial sample collection were considered eligible for analysis.

Results: Of 471 patients included in the study, 153 (32%) were positive for HRV at least once during the study duration; 128 patients had a reported viral load at the initial positive sample and had a symptom survey completed within 3 days of the initial positive sample. Comparison of median CT values of initial positive samples of patients with no symptoms, 1-2 symptoms, and 3 or greater symptoms revealed the presence of fewer symptoms was associated with higher CT values (i.e. lower viral loads, p = 0.036, Figure 1). Patients whose initial CT value was high (≥35) were more likely to be asymptomatic at initial positive test result than patients with lower initial CT values (25-30 and <25) (48% vs. 26% vs. 17%, p = 0.064, Figure 2A) and were more likely to remain asymptomatic throughout the viral shedding episode (27% vs. 7% vs. 9%, p = 0.033, Figure 2B).

Conclusions: No or few symptoms at the time of initial viral positivity was associated with lower viral loads (higher CT value), whereas the presence of an increased number of symptoms (≥3) was associated with a higher viral load. Lower viral load at initial surveillance testing correlated with an increased likelihood of being asymptomatic initially and remaining asymptomatic throughout the viral shedding episode. These results provide initial support for the use of viral load as a clinical endpoint in interventional trials.