Impact of admission viral load on respiratory outcomes in hospitalized SARS-CoV-2 infected patients with cancer and without cancer: A 2-, 4- and 6-months follow-up prospective study

BACKGROUND: This prospective follow-up study aimed to determine the temporal changes in respiratory outcomes over 6 months period in patients with and without cancer hospitalized for severe COVID-19 and to determine the associated risk factors based on admission viral load. METHODS: All adult patients hospitalized with a confirmed diagnosis of severe SARS-CoV-2 infection were investigated using rRT-PCR on nasopharyngeal swab specimens. Patients were divided into three arbitrary groups according to their cycle threshold (C(T)) values obtained at admission as high (C(T) < 25.0), medium (C(T) between 25.0 and 30.0), and low (C(T) > 30.0) viral load. Patients had pulmonary function tests, chest high-resolution computed tomography (HRCT), and a 6-minute walking time distance measured at each follow-up visit. RESULTS: This follow-up study had a total of 112 participants, of which 75 were cancer-free and 37 had active cancer. Overall, 29.5% had a low viral load, compared to 48.2% who had a high viral load, and 22.3% had a medium viral load. For patients who did not have cancer, the mean age was 57.3 (SD 15.4) and for those who had cancer, it was 62.3 (SD 18.4). Most patients had overall better temporal changes in pulmonary function and tolerance, as well as exercise capacity, even though severe and chronic respiratory abnormalities persisted in a fraction of the patients. In patients without cancer who had a high viral load, we have seen a substantial reduction in diffusion capacity of the lungs for carbon monoxide (DLCO) predicted value with a median of 65 (IQR 63–70) while in patients with cancer, it was 60 (IQR 56–67) at 2 months. At 4 and 6 months, the predicted DLCO values for patients without cancer were 65 (IQR 61–70), whereas the predicted DLCO values for patients with active cancer were 62 (IQR 60–67) and 67 (59−73). Importantly, radiological abnormalities persisted in 22 (29%) non-cancer patients and 16 (43%) cancer patients. Multivariate regression analysis showed an increased odds ratio of impaired HRCT associated with a high viral load of 3.04 (95% CI:1.68–6.14; p < 0.001) for patients without cancer and 5.07 (95% CI: 4.04–10.8; p < 0.0001) for patients with cancer. The CT pneumonia score at hospitalization was 2.25 (95% CI:1.76–3.08; p = 0.041) and 2.85 (95% CI:1.89–5.14; p = 0.031) for non-cancer and cancer patients respectively. CONCLUSIONS: The evidence of persistent pulmonary abnormalities and radiographic changes was found in both patient groups who had high viral load at hospital admission and suggesting that SARS-CoV-2 viral load might serve as a useful indicator to predict the development of respiratory complications in patients with COVID-19.