COVID-19 mortality in cancer patients: a report from a tertiary cancer centre in India

Introduction

The ongoing pandemic of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused unprecedented health and societal crises across the globe. However, this catastrophe has victimized the cancer patients the most, adversely impacting diagnosis and treatment in about 55% of cases worldwide (World Health Organisation, 2020). The number of patients visiting and accessing oncological services has considerably reduced and the collateral damage can have an adverse impact on cancer outcomes (Guven et al., 2020; Sud et al., 2020). It is generally assumed that patients with cancer are at a higher risk of contracting COVID-19 due to their immunosuppressive state, and side effects associated with anticancer therapies like leukopenia and disruption of the barrier to infections (ElGohary et al., 2020; Ganatra, Hammond & Nohria, 2020; Liang et al., 2020). Additionally, it is also postulated that the severity and resulting mortality is amplified in cancer patients with COVID-19 due to their elderly and the immunocompromised state further worsened by cancer treatment. Multiple studies from different geographical locations show that the case fatality rate (CFR) of the SARS-CoV-2 infected cancer patients varies from 3.7% to 61.5% (He et al., 2020; Kalinsky et al., 2020; Saini et al., 2020). Despite recent efforts, no clear consensus has been reached for the relation of mortality to demographics, cancer type, stage, and underlying comorbidities; and it may depend on the epidemiology and prevalent oncology practices (Miyashita et al., 2020; Oh, 2020). Various studies regarding the risk of treating cancer patients during the pandemic have shown contradictory results (Dai et al., 2020; Lee et al., 2020; Liang et al., 2020; Pinato et al., 2020). Larger datasets acquired globally from different sources and geographic locations and rationally analyzed has the best potential to provide an effective risk-benefit calculus to assist oncologists to optimize the management and use anticancer treatments to restore outcomes to the pre-COVID-19 era.

We aim to describe the clinical and demographic characteristics and COVID-19 outcomes in a cohort of patients with cancer and symptomatic COVID-19. An attempt has also been made to assess the adverse effect of cytotoxic and other novel therapies on mortality in cancer patients with COVID-19.

Materials and Methods

Results

A total of 3,101 cancer patients were treated at the indoor facility of the center. One thousand and eighty-eight patients developed signs, symptoms, and/or radiological features suspicious of COVID-19. Of these, 186 tested positive for COVID-19 and formed the study cohort. The infection rate of COVID-19 among all cancer patients treated at the center was ~6% (186/3,101) and 17.1% (186/1,088) of the symptomatic and tested cancer patients. The clinical features are shown in Table 1. Most patients had solid malignancies (82.26%); gastrointestinal cancer (21.51%) was the most common cancer type, and about 17.74% of cases presented with hematological malignancies. More than a quarter of cases (26.88%) were metastatic. Eighty-six patients (46.24%) had at least a single comorbidity; hypertension (24.19%) and diabetes (18.28%) were the most common. About 60% of cases were on active cancer treatment and had received cancer-directed treatment within a month before the onset of COVID-19 symptoms. The majority of patients were on chemotherapy (37%).

The chief presenting symptoms of SARS-CoV-2 infection were fever (123/186, 66.13%), fatigue (26/186, 13.98%) and respiratory distress (25/186, 13.44%). Around 5% of patients were presented with severe disease. The COVID-19 associated fatality rate in the cohort was 14.52% (27/186) (median follow-up duration: 63 days). Compared to hospitalized cancer patients without COVID-19, the CFR in cancer patients with COVID-19 was significantly higher (27/186, 14.32% vs. 40/2,915, 1.37%; P < 0.0001). The CFR for hematological malignancies tended to be higher than the CFR for solid malignancies, however, the difference did not reach statistical significance (7/33, 21.21% vs. 20/153, 13.07%, P = 0.274).

Severe COVID-19 infected cancer cases were managed by treatments including corticosteroids, hydroxychloroquine, remdesivir, tocilizumab and convalescent plasma therapy. Patients with mild or moderate disease severity were given symptomatic treatment. Hydroxychloroquine and/or dexamethasone were administered in moderate disease severity cases. Assisted ventilation was given to 12 patients (6.45%), however, all of these patients eventually developed COVID-19 related complications like pneumonitis and associated respiratory failure, septic shock, or sudden cardiac arrest and succumbed to the disease.

Next, we explored the differences between the cancer patients who died and those who survived the SARS-CoV-2 infection (Table 1). There was no significant difference between the survivors and the non-survivors with respect to age, gender, type of malignancy, and cancer spread. Also, no significant effect on mortality was noted for the patients who had received anticancer therapy within the past month. Deceased patients displayed significantly higher rates of comorbidity compared to the cancer patients who survived (66.67% vs. 42.77%, P = 0.035); patients with greater than one comorbidity had significantly inferior outcome than those with single or no comorbidity (P = 0.004). Importantly, patients with diabetes experienced significantly more deaths than patients without diabetes (P = 0.013) (Table 1). Further, we observed that cancer patients presented with moderate or severe COVID-19 symptoms had significantly higher mortality than those presented with mild symptoms (P < 0.00001).

The univariate logistic regression analysis for death has been shown in Table 2. The mortality risk was statistically significant for the presence of any comorbidity (OR = 2.68, P = 0.025), multiple vs. single morbidity (OR = 3.01, P = 0.047), cardiovascular disease (OR = 3.77, P = 0.046) and diabetes (OR = 3.31, P = 0.009). The odds of death were significantly higher in patients presented with severe COVID-19 infection symptoms compared to mild/moderately symptomatic patients (OR = 27.48, P < 0.001). Patients who were on active cancer treatment during 1 month before contracting COVID-19 did not have an increased risk of death (P = 0.460). Also, the treatment modalities including surgery, chemotherapy, radiotherapy, targeted therapy, and immunotherapy did not confer an increased risk of death in univariate analyses. There was a significant difference in the median age of the patients who received chemotherapy compared to those who did not (49 years vs. 54 years, P = 0.005). So we further examined whether anticancer therapies could influence mortality in the cohort by adjusting for age and comorbidity in the multivariate logistic regression analysis (Table 3). In comparison to the patients who were not on these treatments, there were no significant increase in risk of death with chemotherapy (OR = 1.63, 95% CI [0.64–4.15], P = 0.301), radiotherapy (OR = 0.19, 95% CI [0.02–1.59], P = 0.126), targeted therapy (OR = 2.70, 95% CI [0.42–17.37], P = 0.296), immunotherapy (OR = 0.40, 95% CI [0.04–3.56], P = 0.413), or surgery (OR = 0.87, 95% CI [0.25–2.94], P = 0.819) (Table 3).

Discussion

The COVID-19 pandemic has raised several fears. One of these has been the increased risk of cancer patients to contract COVID-19. In the current study, we found the incidence of COVID-19 to be ~6.0% in our hospitalized cancer patients. The national incidence of COVID-19 in an unselected cohort is 0.32% and 0.29% as per the European Centre for Disease Prevention and Control (Fig. 1A) (Government of India, 2020; Roser et al., 2020). The incidence of COVID-19 observed in hospitalized cancer patients is expected to be higher than in the general population. This on the one hand can be ascribed to patients’ factors, visitations, and admission to a health care facility with a high risk of contracting infection; may also be partly because of lower case detection rates in the general population in India due to poor access to health care facilities and low testing rates. The reported incidence in other high COVID-19 incidence nations like Brazil and the USA, and other middle and low incidence countries have been shown in Fig. 1B.

The average CFR of COVID-19 is 1.9% in the unselected Indian population as per the national database (Arogyasetu Application), 1.84% by Johns Hopkins COVID-19 tracker, and 1.7% as per the European Centre for Disease Prevention and Control (Fig. 1C) (Government of India, 2020; John Hopkins University and Medicine Coronavirus Resource Center, 2020; Roser et al., 2020). The 14.52% COVID-19 related CFR observed for hospitalized cancer patients in the present study is similar to the data reported by Ma et al. (2020) from Renmin Hospital of Wuhan University (CFR-13.5%), and is higher than a recent report from Tata Memorial Hospital, Mumbai, India (CFR-6.5%) (Ma et al., 2020; Ramaswamy et al., 2020). The previous studies from the European continent have also shown a far greater CFR in cancer patients with COVID-19. In the UK Coronavirus Cancer Monitoring Project (UKCCMP), a CFR of 30.6% was observed, where 319 of the 1,044 cancer patients with COVID-19 died with 92.5% had their death attributed directly to COVID-19 (Lee et al., 2020). In another observational study by Pinato et al. (2020) of 890 cancer patients diagnosed with SARS-CoV-2, mortality was found to be 33.6%. Similarly, high rates were noticed in a New York Hospital System where a CFR of 28% was observed (Mehta et al., 2020). A large systematic review of 52 studies by a group led by Kamal S. Saini involving more than 18,000 cases of cancer with COVID-19, the probability of death was 25.6% (95% CI [22.0–29.9]; I² = 48.9%) (Saini et al., 2020). Among the hospitalized cancer patients, we observed that the CFR was about 10.6 times higher in the COVID-19 infected patients than those without COVID-19 (14.52% vs. 1.37%, P < 0.0001) indicating that COVID-19 infection significantly increased the risk of death in the cohort.

Unlike the previous retrospective studies from China, we did not find any significant association between recent cancer treatments and mortality (Dai et al., 2020; Liang et al., 2020). Our observation is in line with two large cohort studies conducted by Lee et al. (2020) on 800 COVID-19 positive cancer patients from UKCCMP, and by Robilotti et al. (2020) on 423 symptomatic COVID-19 cancer patients at Memorial Sloan Kettering Cancer Centre in New York (Lee et al., 2020; Robilotti et al., 2020). The recent multi-center study by Pinato et al. (2020) in the European cancer patients and the Covid-19 and Cancer Consortium (CCC19) database study also strengthens the notion that cancer treatment is not associated with mortality (Kuderer et al., 2020; Pinato et al., 2020). Similar to these studies we found that it is the underlying comorbid conditions and COVID-19 disease severity that are associated with adverse outcomes. Consistent with the CCC19 study we observed a higher burden of pre-existing comorbidities (>1) to be associated with increased mortality (Kuderer et al., 2020). Cardiovascular diseases and diabetes posed a higher risk of death in our cohort. It should be noted that comorbid conditions like diabetes frequently co-occur with hypertension or coronary artery disease in patients and can further weaken the immune response escalating the risk of death due to COVID-19 (Guan et al., 2020; Naqvi et al., 2017).

Conclusion

Our study highlights the high rates of COVID-19 in cancer patients, with a CFR of 14.52%. Recent anticancer therapies did not have a significant effect on mortality in the cohort. Pre-existing comorbidities especially diabetes, presence of more than one comorbidity and severe COVID-19 presenting symptoms were significantly linked with COVID-19 related deaths in the cohort.

Supplemental Information