Retrospective Analysis of Seasonal Respiratory Viral Infections (SRVI) in Hematology, Lymphoma and Oncology Patients

Introduction: Community acquired SRVI increase hospital referrals, hospitalization and ICU admissions resulting in high morbidity and mortality during winter season. As there are no defined preventative or treatment measures for most of the SRVI, there is increasing burden on health service resources during SRVI season. This analysis was carried out to evaluate the SRVI incidence, risk factors, impact on mortality and changes in incidence trends over 9 year period in immunocompromised cancer patients.

Methods: 2906 cancer patients (haematology: n=1098, 37.8%, lymphoma: n=643, 22.1%, other cancers: n= 1156, 40.1%) treated from January 2006 to January 2015 who had respiratory virus PCR were evaluated. Patients with haematological cancers included ALL (n=137), AML (n=338), Myeloma (n=396), CLL (n=131) or other cancers (n=96) [median age: 50 yr., 5-87, Male: 692, Female: 406). Common solid tumour diagnosis included cancer of Breast (n=280, 24.0%), GI tract (n=207, 17.8%), Lung (n=190, 16.3%), Genitourinary (n=180, 15.5%) or other sites (n=299, 25.7%), [Males: 461, Females: 695; median age: 55 yr., range: 6-89]. Patients with haematological malignancies were younger than patients with other cancers (median age: 50 yr. vs. 55 yr., p<0.001). 804 patients (27.7%) had stem cell transplant. Incidence was compared to the seasonal incidence of SRVI reported by NHS England. All patients with respiratory symptoms who had viral PCR requested on throat and nose swab were included. A total of 10,025 samples were evaluated.

Results: In patients with malignancy, the season for ParaFlu, Rhinovirus, Metapneumovirus and FluA lasted longer than in general population (average: 2 months, started early and ended later). Incidence of RSV (6.2%, 4.9%, 1.6%, p=0.001), Adenovirus (1.3%, 1.7%, 0.33%, p=0.004), Rhinovirus (16.6%, 19.9%, 8.5%, p=0.001) and ParaFlu (7.4%, 6.3%, 2.6%, p=0.057) was higher in hematology and lymphoma patients. Incidence of PCP was higher in oncology patients (15.1%, 7.2%, 9.6%, p=0.001). Incidence of PCP was higher with increasing age (5.8% age< 50, 12.2% age>50 yr., p=0.001). Rhinovirus (18.7% age<25 yr., 12.3% age >75 yr., p=0.001) and ParaFlu (8.1% age <25 yr. vs. 6.1% age >25 yr., p=0.02) was higher in younger patients. Stem cell transplant increased risk of RSV (6.8% vs. 3.5%, p=0.001), Adenovirus (1.7% vs. 0.6%, p=0.001) and ParaFlu (8.1% vs. 0.23%, p=0.001) but risk of PCP (7.7% vs. 11.8%, p=0.0001) was lower. Risk of positive PCR for any respiratory virus was higher with increasing age, hematological cancers, and use of stem cell transplant. Surprisingly, diagnosis of CLL and Myeloma did not increase SIRV risk. Thirty-day mortality was higher in patients who had SRVI (p=0.041). Mortality was higher in patients with solid tumours (p<0.0001), RSV infection (p<0.001), FluA (p0.02), PCP (p<0.001), non-SCT patients (p<0.0001) and older age. Except for increasing incidence of PCP in Oncology patients no annual variations in the incidence of specific pathogens was seen.

Conclusion: This is one of the first reports that compares incidence of SRVI in patients with cancer to that in general population. The analysis of SRVI using PCR based diagnosis demonstrates that incidence of SRVI in cancer patients show different trends than in general population. SRVI season lasts longer and RSV, FluA and PCP contribute to 30-day mortality. Increasing PCP incidence in patients with solid tumours raises the questions about need to use of PCP prophylaxis in all these cases.