The outcome of gynecologic cancer patients with Covid-19 infection: A systematic review and meta-analysis [version 1; peer review: awaiting peer review]

Background: Cancer is a comorbidity that leads to progressive worsening of coronavirus disease 2019 (Covid-19) with increased mortality. This is a systematic review and meta-analysis to yield evidence of adverse outcomes of Covid-19 in gynecologic cancer. Methods: Searches through PubMed, Google Scholar, ScienceDirect, and medRxiv to find articles on the outcome of gynecologic cancer with Covid-19 (24 July 2021–19 February 2022). The Newcastle-Ottawa Scale tool was used to evaluate the quality of included studies. Pooled odds ratio (OR), 95% confidence interval (CI) and random-effects model were presented. Results: We accepted 51 studies (a total of 1991 gynecologic cancer patients with Covid-19). Covid-19 infection cases were lower in gynecologic cancer vs hematologic cancer (OR 0.71, CI 0.56-0.90, p 0.005). Severe Covid-19 infection and death were lower in gynecologic cancer vs lung and hematologic cancer (OR 0.36, CI 0.16-0.80, p 0.01), (OR 0.52, CI 0.44-0.62, p <0.0001), (OR 0.26, CI 0.10-0.67 p 0.005), (OR 0.63, CI 0.47-0.83, p 0.001) respectively. Increased Covid death was seen in gynecologic cancer vs population with breast cancer, nonCovid cancer, and non-cancer Covid (OR 1.50, CI 1.20-1.88, p 0.0004), (OR 11.83, CI 8.20-17.07, p <0.0001), (OR 2.98, CI 2.23-3.98, p <0.0001) respectively. Conclusion: Gynecologic cancer has higher Covid-19 adverse outcomes compared to non-cancer, breast cancer, non-metastatic, and Covid-19 negative population. Gynecologic cancer has fewer Open Peer Review Approval Status AWAITING PEER REVIEW Any reports and responses or comments on the article can be found at the end of the article. Page 1 of 22 F1000Research 2022, 11:525 Last updated: 18 MAY 2022


Introduction
The Covid-19 pandemic has changed the way health care providers around the world manage care provided to their patients. The pandemic has also proven to shift the attitude of standard practice and procedure between providers and patients, for example, to reduce gynecologic cancer patients visiting the hospital as possible because the risk of getting infected with Covid-19 is increased regarding their comorbidities. 1 Despite this circumstance, gynecologic cancer patients are still often required to perform routine hospital visits for treatments or other medical procedures under guidance made by gynecological cancer societies during the Covid-19 pandemic. 2 The cancer incidence and mortality are still increasing around the world. According to Global Cancer Statistic: 2020 for gynecologic cancer, there are 604.127, 417.367, 313.959, 45.240, and 17.908 new cases of cancer of the cervix uteri, corpus uteri, ovary, vulva, and vagina respectively. 3 Most concerns are coming from these patients about how they may proceed to seek or continue their cancer treatment and surveillance during the Covid-19 pandemic. 4 Studies are showing various results on increased mortality and severity among cancer patients infected with Covid-19. Systematic review and meta-analysis studying the outcome of cancer patients with Covid-19 show 2.1-4% proportion of cancer patients among those infected with Covid-19, additionally compared to non-cancer with Covid-19 greater amount of mortality and severity are observed in cancer population with Covid-19. [5][6][7] However studies and data on the outcome of gynecologic cancer patients with Covid-19 are still lacking. Several SARS-CoV-2 variants of concern listed by WHO (World Health Organization) pose challenges in mitigating the pandemic as these variants often increase transmission rate and severity. 8 The world has been experiencing a wave of active case surges by these variants and on 26 November 2021 the WHO designated the variant Omicron (B.1.1.529) as an addition to the list. 9 Thus we attempt to review the literature and quantify the effect of the SARS-Cov-2/ Covid-19 infection among gynecologic cancer patients to assess whether the risk of infection, hospitalization, severity, and mortality are increased than non-gynecologic cancer population.

Methods
We conducted this systematic review and meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses/PRISMA statement. 10,82 This study and its protocol were registered to PROSPERO (CRD42021256557).

Eligibility criteria
We took into consideration of studies with observational cohort studies, case-control, cross-sectional, case report, and case series designs that evaluate the outcome of gynecologic cancer patients infected with Covid-19 from the year 2019. Each study ought to report Covid-19 associated infection, hospital admission, mortality, severity, or admission to the intensive care unit (ICU); a summary of eligible studies and its extracted outcome of interest were managed in the Microsoft Excel spreadsheet provided in the Underlying data. 82 We exclude studies other than the English language, reviews or guidelines, and inconceivable results of the sought outcome.
Database and literature search Study articles were systematically searched through PubMed/Medline, ScienceDirect, Google Scholar, and medRxiv. Relevant articles had been screened from 24 July 2021 to 19 February 2022. Reference searches from retrieved articles citation lists were identified if any were needed. Boolean operators technique used for Pubmed/Medline search with ("COVID-19" or "2019-nCoV" or "SARS-CoV" or SARSCOV2 or 2019-nCov or "2019 coronavirus" or covid19) AND (gynecology or gynaecology) AND (tumor or malignancy or cancer) AND (outcomes or outcome) AND (gyn* tum* or gyn *malign* or gyn* cancer) AND (cancer surgery or oncolog* surger*) AND (brachytherapy or radiotherapy). We used "Gynecologic cancer AND Covid-19" with Google Scholar, Science Direct, and medRxiv. Two authors separately handled the literature search. Findings were accumulated and stored in Mendeley and Zotero for management and automated duplicate identification. Thorough stepwise screening from title and abstract was then conducted to determine possible article inclusion. Potentially eligible studies were then evaluated for in-depth full-text review. Each author would consult senior authors to resolve any differences found during the literature's selection process.
Data extraction and quality assessment The data was extracted independently by two authors and stored them in the Microsoft Excel spreadsheet. Data was then discussed for an agreement. Name of authors, year of publication, country, type of studies, study period, number of patients, comparators, and target conditions was collected. The NOS/Newcastle-Ottawa Scale was used by authors to assess the quality of the cohort and case-control study, and The Joanna Briggs Institute (JBI) Critical Appraisal Checklist for an analytical cross-sectional study. 11 The assessment was performed by two authors and the results were discussed with the first author.

Meta-analysis outcome
The main outcome of interest was Covid-19 mortality and severity. Covid-19 severity is defined as either ICU admission, acute respiratory distress syndrome (ARDS), or need for mechanical ventilation. Covid-19 infection and hospitalization were decided as secondary outcomes.

Data analysis & synthesis
We performed data analysis mainly using Review Manager 5.4.1 (RevMan 5.4.1) by Cochrane collaboration. 12 If needed, additional synthesis was then performed with STATA-16. We synthesized the dichotomous outcome from each study with an odds ratio (OR). The random-effects model (DerSimonian and Laird) was used to present pooled OR with 95% CI (confidence interval) and the result of overall effect (p). We addressed the presence of heterogeneity with I 2 as 0% to 40%: might not be important; 30% to 60%: may represent moderate heterogeneity; 50% to 90%: may represent substantial heterogeneity; 75% to 100%: considerable heterogeneity according to the Cochrane Handbook for Systematic Reviews of Interventions. We performed subgroup analysis by age, gender, other comorbidities status, cancer type, cancer stage, presence of metastatic disease, and active cancer treatment. Sensitivity analysis was performed by dividing multi-center/ single-center studies and removing/including the latest study period if concerns were raised of patients population duplication thus we could present robust pooled evidence. 13

Comorbidities
Cancer is a comorbidity, aside from which we tried to analyze other comorbidities (hypertension, diabetes, cardiovascular disease, pulmonary disease, renal disease, liver disease, immune disease, metabolic-endocrine disease) present within the cancer population. Among those with comorbidities, gynecologic cancer patients had fewer adverse Covid-19 outcomes than other cancer populations according to four studies (OR 0.31, CI 0.12-082, p 0.02, I 2 0%) Figure 11. 20,23,24,59 Data from five studies showed there was no significant adverse Covid-19 outcome between gynecologic cancer patients with comorbidities against no comorbidities (OR 2.34, CI 0.59-9.79, p 0.24, I 2 79%) Figure S25. 15,21,23,24,35 Gynecologic cancer patients without comorbidities against other cancer patients with comorbidities had no significant difference in   adverse Covid-19 outcomes, according to three studies (OR 0.29, CI 0.04-2.22, p 0.23, I 2 56%) Figure S26. 23,24,59 Gynecologic cancer patients with comorbidities against other cancer patients without comorbidities also showed no significant difference in adverse Covid-19 outcomes, according to four studies (OR 0.61, CI 0.22-1.72, p 0.35, I 2 0%) Figure S27. 20,23,24,59 Sensitivity analysis We performed sensitivity analysis by reproducing each outcome synthesis to pre-specified single center to multi-center studies, furthermore excluding overlapped study periods associated with its study centers, thus only one center with the most recent study period was included in Table S1. After exclusion of three studies, a difference of significance was found in severe Covid-19 between gynecologic cancer and cancer men population (OR 0.47, CI 0.19-1.17, p 0.10, I 2 0%) 24,31,52 Aside from that, the remainder of the calculated OR from reproducing each outcome synthesis by exclusion were within good accordance.

Publication bias
We found no publication bias within our included studies though at first, we identified an asymmetrical funnel plot; it was caused solely by heterogeneity nonetheless (Figures S28-31 ). 66 Our finding shows gynecologic cancer with metastatic disease has an increased Covid-19 death compared to those whose cancer is localized (OR 1.53, CI 1.06-2.21, p 0.02, I 2 0%), most studies also report identical outcomes to ours. 65 78 Our analyzed population comprises cancer as the main comorbidity, however with comorbidities other than cancer, our study shows that the gynecologic cancer population with additional comorbidities has fewer adverse events than other cancer with comorbidities (OR 0.31, CI 0.12-082, p 0.02, I 2 0%). Other meta-analyses prove that men have increased Covid-19 severity and mortality. 78 We hope these findings will be useful among gynecologist-oncologists in cancer centers or tertiary cancer referral centers who provide care to gynecologic cancer patients during the ongoing Covid-19 pandemic.
In several data syntheses with the statistically nonsignificant value, we analyze few data regarding severity, hospitalization, age, cancer stage/metastatic status, other comorbidities aside from cancer, and cancer treatment type due to limited data, however those aforementioned are well represented and distributed through other synthesis based on the patient's characteristics available in Table 1.