Clinical Characteristics of COVID-19-Infected Cancer Patients in Pakistan: Differences Between Survivors and Non-Survivors

Background Cancer patients are considered as highly vulnerable individuals in the current COVID-19 pandemic. We studied the clinical characteristics of survivor and non-survivor COVID-19-infected cancer patients in Pakistan. Patients and Methods We did a retrospective study of 70 cancer patients with PCR-confirmed COVID-19 infection from Shaukat Khanum Memorial Cancer Hospital and Research Centre, Lahore and Peshawar, Pakistan between April 13 and July 09, 2020. These patients were discharged from the hospital or had died by July 09, 2020. Clinical, pathological and radiological characteristics were compared between survivors and non-survivors by fisher’s exact test and chi-square test. Univariable and multivariable logistic regression models were performed to explore the risk factors of mortality. Results Seventy cancer patients with SARS-CoV-2 infection were enrolled and the majority were males 38 (54.3%). 57 (81.4%) had solid tumors and 13 (18.6%) had hematological malignancies. Dyspnea (44 cases) was the most common symptom (62.9%). Complications were reported in 51 (72.9%) patients during the course of disease. 19 (27.1%) patients were admitted to an intensive care unit (ICU). A significant increase in the C-reactive protein level and neutrophil count was observed in the deceased patients as compared to the surviving patients. D-dimer values of ≥0.2 mg/L were significantly associated with mortality (P=0.01). We identified two independent risk factors associated with death, ICU admission (P=0.007) and D-dimer (P=0.003). Conclusion Pakistani cancer patients with COVID-19 infection reported poor prognosis. Intensive surveillance of clinicopathological characteristics of cancer patients infected with COVID-19 especially D-dimer values may play a pivotal role in the outcome of the disease.

Epidemiological data have indicated that cancer patients are more susceptible to the COVID-19 infection (7)(8)(9)(10)(11). SARS-CoV-2 infection affects the functioning of cardiovascular, renal, hepatic, pancreatic and gastrointestinal systems (8). Furthermore, the virus has the ability to invade and cause neural disease in the central and peripheral nervous system (8,12). There is an increased risk of opportunistic infections in cancer patients due to their impaired immunity, poor nutrition and post cancer therapy associated side effects (7). Liang et al. demonstrated that cancer patients with COVID-19 had poorer outcomes (13). Zhang et al. also reported that cancer patients with COVID-19 showed poor prognoses (14). Comprehensive data published by Yang and colleagues concluded that patients with cancer and COVID-19 are a high-risk population with an increased mortality rate (7).
Each year around 200,000 new cancer cases are being diagnosed in Pakistan (15) and approximately, 45,000 cancer patients come to the Shaukat Khanum Memorial Cancer Hospital and Research Centre (SKMCH & RC). SKMCH & RC is a state-of-the-art charitable hospital where 75% of patients diagnosed with cancer are treated free of charge (16). The hospital also accepts cancer patients from Afghanistan.
We analyzed the data from cancer patients with SARS-CoV-2 infection who were admitted to SKMCH & RC. Here, we report the systematically characterized clinical features of survivors and non-survivors COVID-19-infected cancer patients in Pakistan, in an attempt to define the nature of the aggravating factors caused by COVID-19 infection.

Study Design and Participants
The current study was conducted at SKMCH & RC Lahore and Peshawar, Pakistan. This is a retrospective cohort study. The first cases of SARS-CoV-2 were identified in Pakistan in February 2020 (3). Immediately after the COVID-19 outbreak, SKMCH & RC Lahore and Peshawar initiated the diagnosis and treatment of SARS-CoV-2 infected cancer patients. COVID-19 patients were diagnosed based on the guidelines provided by WHO (17). All enrolled patients had previously been diagnosed with cancer. We included all patients (aged ≥ 18 years) with pathological diagnoses of any malignant solid or hematological malignancies along with PCR-confirmation of SARS-CoV-2 infection. We aimed to investigate the clinical characteristics of survivors and non-survivors COVID-19-infected cancer patients in Pakistan. All patients fulfilling the eligibility criteria were included in the study. The study cutoff date was July 09, 2020. The institutional review board (IRB) of SKMCH & RC approved the current study (IRB-EX-20-04-20-01). IRB granted the waiver of the written informed consent from study participants.

Data Collection
The information about the demographic data, clinicopathological characteristics, laboratory findings and chest X-ray examinations were acquired from pathology and radiology reports from our hospital medical records' system. Data about gender, age, comorbidities (diabetes, hypertension, cardiac disease, chronic pulmonary disease, chronic kidney disease and chronic liver disease), cancer history (cancer type, stage and treatment), vital signs (temperature, heart rate, respiratory rate, blood oxygen saturation), body mass index and symptoms (dyspnea, cough, fever, muscle ache, sore throat, diarrhea, vomiting, chills, abdominal pain and headache) and pathology lab tests (white blood cells, neutrophils, lymphocytes, platelets, red blood cells, hemoglobin, creatinine, blood urea nitrogen, lactate dehydrogenase, lactate, pH, PCO2, PO2, HCO3, O2 saturation, alanine transaminase (ALT), aspartate transaminase (AST), albumin, globulin, gamma-glutamyl transpeptidase (GGT), alkaline phosphatase, C-reactive protein, ferritin and D-dimer) were all obtained at the time of diagnosis of COVID-19 in cancer patients. According to the TNM staging system, cancers were defined as stages I, II, III and IV. We also collected the information about treatments for cancer patients with COVID-19 (intravenous antibiotics, antiviral therapy, immunotherapy, anthelmintic agent, montelukast, hydroxychloroquine, zinc/ vitamin-C, continuous renal replacement therapy, oxygen therapy and mechanical ventilation), complications and outcomes during admission to hospital.

Statistical Analysis
We hypothesized that differences exist in demographic, clinical, and laboratory characteristics, treatments, and cancer history between survivors and non-survivors of COVID-19 with cancer. Quantitative variables were presented as medians/mean (range: minimum-maximum/standard deviation), and qualitative variables were presented by frequencies and percentages. The independent t-test, Mann-Whitney U test, Chi-square test, and Fisher's exact test were applied to analyze the differences between groups according to the type of data. Kaplan-Meier analysis (logrank test) was used to check the survival difference. Risk factors associated with death and their odds ratios (ORs) were analyzed by the univariable logistic regression model. Variable selection for multivariable analysis was based on significance from the univariable logistic regression analysis (p<0·05). We used IBM SPSS Statistics 20.0 software for statistical analysis. The tests we used were all two-sided with less than 5% type I error. The differences between groups were considered to be significant when the P-value was less than 0·05.
Overall survival was compared among the D-dimer categories, <0.2 and ≥0.2. There was a statistically significant difference (P=0.01) between the two groups ( Figure 1B). The median survival time in D-dimer was lowest in ≥0.2 category (11 days with confidence interval 8-13) compared to D-dimer <0.2 category (26 days with confidence interval 13-39). Overall survival of D-dimer ≥0.2 category was significantly associated with mortality (P-value 0.01) as shown in Figure 1B.
In a univariable logistic regression model, dyspnea, respiratory rate, heart rate, ICU admission, creatinine, neutrophil, albumin, blood urea nitrogen, C-reactive protein and D-dimer values were associated with death. In addition, SpO2 and lactate were marginally statistically significant with respect to death in the univariable logistic regression model ( Table 5). In a multivariable logistics regression model, we identified two variables as independent risk factors of death, ICU admission (odds ratio (OR), 21

DISCUSSION
Cancer patients are considered as highly vulnerable individuals in the COVID-19 pandemic (7-11). These individuals are at high risk to develop serious illness once diagnosed with COVID-19 infection. The mortality rate is high as well in these individuals (7). In addition to the few studies recently reported from China, this is the first retrospective study conducted in Pakistan to illustrate the comparative analysis of clinical characteristics, treatment, radiological and laboratory findings of survivor and non-survivor COVID-19-infected cancer patients. In particular, we identified two variables as independent risk factors of death: ICU admission and D-dimer. SARS-CoV-2 infection may lead to acute respiratory distress syndrome (18). Dyspnea (62.9%), cough (57.1%) and fever (34.3%) are the most common symptoms as described in previous studies (7,19). Liang et al. observed that old age and comorbidities were the major risk factors (13). This was not observed in our study and our results were similar to the study published by Yang et al. (7). We further observed abnormal respiratory rate, heart rate and blood oxygen saturation in the COVID-19-infected cancer patients. A retrospective study reported a link between hypoxemia (SpO2 <90%) and mortality in COVID-19 patients (20). We found a significant difference in SpO2 in survivors and non-survivors patients. Yang et al. observed higher respiratory rates and lower levels of blood oxygen saturation among non-survivors (7). Our results are in keeping with this study, as we also found a significant difference in respiratory rates in survivors and non-survivors COVID-19 infected cancer patients. In addition to previously reported data, D-dimers are usually elevated in patients with COVID-19 (21). Yao et al. observed an association of increased D-dimer level (> 2.14 mg/L) with in-hospital mortality (21). Tian and colleagues showed that Ddimer might play a predictive role for COVID-19 severity particularly in cancer patients (22). Yang et al. also observed   (24). We also observed higher neutrophil counts and C-reactive protein levels in non-survivor vs survivor cancer patients infected with COVID-19. There is a possibility that proinflammatory neutrophils and C-reactive protein induce "cytokine storm" related to endothelialitis caused by COVID-19 infected cancer patients (23,24). Serum lactate is used to assess the disease severity, therapeutic response and disease outcomes (25). Evaluating the lactate level has clinical significance in the management of critical cases of severe sepsis (26). Previously published data have shown that hyperlactatemia may be a predictor of death (27). In our results, we found a significant increase in the lactate levels of non-survivor COVID-19 infected cancer patients. As COVID-19 disease severity changes rapidly, it is beneficial to identify the parameters of severity at initial stage. Several studies also identified higher levels of creatinine (7, 28), blood urea nitrogen (7,28), aspartate transaminase (29,30), and ferritin (31), that are associated with the severity of COVID-19 infection. We also found elevated levels of creatinine, blood urea nitrogen, aspartate transaminase, and ferritin in non-survivor COVID-19 infected cancer patients.
Xu et al. reported bilateral lung lesions in 53 (59%) of 90 patients (32). We found bilateral lung lesions in 70% of our data set, which was higher than reported previously (32). The involvement of the lungs suggests that cancer patients are more prone to severe pulmonary disease after COVID-19infection. Furthermore, we observed that non-survivors were significantly more likely to receive oxygen therapy, require mechanical ventilation, and be referred to the ICU.  Complications such as ARDS, acute renal failure, septic shock, coagulopathy, secondary infections and arrhythmia were common in these patients. Our results are in keeping with the data published previously (7). Yang et al. reported that COVID-19 cancer patients receiving chemotherapy within 4 weeks before the onset of symptoms were more prone to death during hospital admission (7). In contrast, Lee et al. conducted a study on COVID-19 cancer patients and reported that chemotherapy had no effect on patient mortality (33). We found similar results with our limited retrospective cohort of COVID-19 infected cancer patients who received chemotherapy within 4 weeks before the onset of symptoms, which made no significant difference in mortality as compared to those who did not receive it. Clearly, a long-term follow up of the COVID-19 infected survivor cancer patients is required to identify the impact of the disease. The current study is a retrospective cohort, in which we included only those cancer patients who were admitted to our hospital from April 13 to July 9, 2020. The sample size of our  study was limited because we could not include all cancer patients who tested positive for COVID-19infection since the clinical and pathological data was available for admitted cancer patients only. In addition, we have not evaluated the alteration in hematological parameters during hospitalization. Another limitation of our study is that we could not include non-cancer patients since our hospital is a specialized cancer care hospital.
In conclusion, we provide comprehensive information of the clinical characteristics, laboratory data and outcomes of COVID-19infected cancer patients from Pakistan. This study highlights that elevated D-dimer levels and ICU admission as the two major independent risk factors for early surveillance of disease progression. Furthermore, we need to be more vigilant to the levels of C-reactive protein and neutrophils, once cancer patients are infected with COVID-19. The data presented in this study will certainly serve as a useful addition to the already available knowledge and to decide whether an anti-viral treatment is required or not. The risk factors might help clinicians to determine who are the high-risk cancer patients at the early onset of the infection.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Shaukat Khanum Memorial Cancer Hospital and Research Centre -Institutional Review Board (IRB). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.