1. Introduction

The World Health Organization (WHO) declared a global pandemic due to the COVID-19 (C19) virus on 11 March 2020 [1]. On 15 March 2020, national lockdowns were enforced in the United States, leading to severe healthcare disruptions, including decreased routine well-woman visits, decreased cancer screenings, preoperative delays and cancellations of planned surgeries [1, 2, 3].

The C19 pandemic led to delayed screening and treatment across multiple cancer types. Cancer-related mortality increased during the pandemic, which had been steadily decreasing for the last two decades prior [4]. For example, in April 2020, breast cancer screening was reported to have decreased by 85%, mirrored by decreases in colon cancer screening by 75%, 74% for prostate cancer, and 56% for lung cancer [5]. As a result, during the early pandemic era, the number of mastectomies for breast cancer and colectomies for colon cancer had decreased from baseline in April to July 2020 [5]. Despite the decrease in screening observed, breast cancer and lung cancer stage of presentation did not differ significantly from before the C19 pandemic [6, 7].

To address resource limitations during the pandemic, professional societies, such as the American Society of Clinical Oncology (ASCO) and the American Society for Colposcopy and Cervical Pathology (ASCCP) recommended delaying routine cancer screening [2, 8]. Later, the ASCCP established guidelines to redirect screening recommendations for cervical cancer; individuals with low-grade screening tests may postpone follow-up diagnostic evaluation for 6–12 months, and individuals with high-grade screening tests should schedule follow-up diagnostic evaluation within 3 months or less [8]. Additionally, the American College of Obstetrics and Gynecology (ACOG) recommended that in-person visits to the Obstetrician-Gynecologist should be reserved for urgent concerns, which did not include cervical cancer screening [9]. These modifications have been correlated with delays in cancer screening on an international scale, with decreases in cervical cytology ranging from 50–80% [10, 11]. In the United States, an estimated 22 million cancer screening tests were delayed in the first year of the C19 pandemic [12].

Based on the impacts of the C19 pandemic on other cancer types and the changes in screening for cervical cancer, we aimed to investigate if there were increased numbers of advanced cervical cancer during the C19 pandemic compared to the pre-pandemic era, whether the pandemic was associated with differences in treatment modalities for cervical cancer, and whether there were increased number of treatment delays during the pandemic.

2. Materials and methods

This retrospective study analyzed the National Cancer Database (NCDB) 2021 on Cervical Cancer for patients with SCCC. The NCDB is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society, which contains de-identified data from more than 1500 CoC-accredited facilities across the United States and captures 72% of diagnosed malignancies in the United States annually [13, 14]. A Business Associate Agreement is established between the American College of Surgeons and the participating CoC-accredited hospitals for the use of data. The American College of Surgeons and the CoC have not verified the statistical analysis employed by this study and are not responsible for the conclusions drawn from these data. This database was selected for the source in this study as it yielded a large sample size of patients from a large geographical region and captures approximately 70% of cancer patients treated in the United States, allowing for increased statistical power and generalizability [15].

Patients diagnosed with SCCC in the pre-C19 era were defined as all patients diagnosed on 01 January 2018 to 31 December 2019, and patients diagnosed with SCCC in the C19 era were defined as all patients diagnosed on 01 January 2020 to 31 December 2021. The most current NCDB version at the time of analysis had data up to the year 2021, thus, the C19 cohort was limited to years 2020–2021. To reduce statistical bias, the pre-C19 cohort was also limited to 2 years (2018–2019).

The severity of SCCC was analyzed via stage distribution. The NCDB does not report the International Federation of Gynecology and Obstetrics (FIGO) 2018 stage distribution but does report American Joint Committee on Cancer (AJCC) 2017 tumor node metastasis (TNM) clinical stage distribution (S0–S4). Therefore, AJCC 2017 TNM clinical stage distribution was correlated to the FIGO 2018 stage distribution. Because the FIGO staging system changed in 2018, using the AJCC 2017 TNM clinical stage distribution also avoided any discrepancies in reporting, and ensured a standardized staging system across the years included in this study, if any lag in adoption of the FIGO staging system had occurred. Early-stage disease was defined as stages T1aN0M0–T2a2N0M0 by AJCC, which equates to FIGO 2018 Stage IA1–IIA2. Locally advanced disease was defined as T2BN(any)M0–T4N(any)M0 by AJCC, which equates to FIGO 2018 Stage IIB–IVA. Metastatic disease is defined as T4N(any)M1 by AJCC, which equates to FIGO 2018 Stage IVB. Stage 0 disease was excluded from analysis as this represents cervical carcinoma in situ rather than cervical cancer.

The primary aim of this study is to investigate if an upstaging of SCCC disease severity occurred during the C19 pandemic. The secondary aim of this study was to evaluate changes in time to treatment and any changes in treatment utilization during the C19 pandemic. Age and Charlson Comorbidity Index scores were compared between the pre-C19 and C19 cohorts. Next, analysis compared the stage distribution between the pre-C19 and C19 cohorts. The mean number of days from diagnosis to treatment in the pre-C19 and C19 eras were compared using a student t-test; similarly, the difference between the mean number of days from diagnosis to surgery in the pre-C19 and C19 eras were compared. Patients were excluded from analysis if they had no recorded days from diagnosis to treatment or diagnosis to surgery. Next, the rate of patients who received surgical management between the pre-C19 and C19 cohorts were compared, as well as the rate of patients who received chemotherapy and radiation between the two cohorts. Finally, the stage distributions in the C19 cohort between patients who tested positive and negative for C19 virus infection were compared. All categorical data was analyzed using a chi-squared test. A p-value less than 0.05 was considered significant. Statistics were performed with Python software using the SciPy library.

3. Results

The analysis compared 11,455 patients with SCCC during the pre-C19 era to 10,155 patients with SCCC during the C19 era. There were no significant differences in Charlson Comorbidity Index scores (p = 0.23) or age (p = 0.12) between the two cohorts (Fig. 1). Patients with early-stage disease comprised 44.1% of patients with SCCC in the pre-C19 cohort and 38.1% in the C19 cohort (p < 0.001). Patients with locally advanced disease included 42.8% of patients with SCCC in the pre-C19 cohort and 47.8% of patients with SCCC in the C19 cohort (p < 0.001). Patients with metastatic disease made up 13.0% of patients with SCCC in the pre-C19 cohort and 14.1% of patients with SCCC in the C19 cohort (p < 0.001) (Fig. 2). Overall, there was a significant decrease in early-stage disease, while there was a significant increase in locally advanced and metastatic disease.

Fig. 1.Pre-COVID-19 (Pre-C19) versus COVID-19 (C19) Charlson Comorbidity Index scores and age. (a) A chi-squared test compared Charlson Comorbidity Index scores between the two cohorts. (b) A student t-test compared the mean age between the pre-C19 cohort and the C19 cohort. No statistical differences were found between the cohorts in Charlson Comorbidity Index scores (p = 0.23) or mean age (p = 0.12).

Fig. 2.Pre-COVID-19 (Pre-C19) versus COVID-19 (C19) stage distribution. Analysis compared the stage distribution between the pre-C19 cohort and the C19 cohort using a chi-squared test. Analysis found less early-stage disease (FIGO IA1–IIA2) (44.1% pre-C19 vs. 38.1% C19), more locally advanced disease (FIGO IIB–IVA) (42.8% pre-C19 vs. 47.8% C19), and more metastatic disease (FIGO IVB) (13.0% pre-C19 vs. 14.1% C19) (p < 0.001) in the C19 cohort. SCCC: squamous cell cervical cancer; FIGO: International Federation of Gynecology and Obstetrics.

Mean time to treatment (p = 0.10) and surgery (p = 0.29) did not differ significantly between the two time periods (Fig. 3). The rate of SCCC patients receiving surgical management during the pre-C19 era was 34.1%, which was higher than that in the C19 era, at 31.8% (p < 0.001). The rate of SCCC patients receiving chemotherapy during the pre-C19 era was similar to that in the C19 era, at 68.8% and 69.9%, respectively, (p = 0.09). Similarly, the rate of SCCC patients receiving radiation treatment during the pre-C19 era was similar to that in the C19 era, at 29.9% and 28.8%, respectively (p = 0.10) (Fig. 4). Of the SCCC patients in the C19 era, the C19 positive rate was 4.29%; no statistical difference was found in cancer stage between patients who tested positive and those who tested negative for the C19 virus (p = 0.25).

Fig. 3.Pre-COVID-19 (Pre-C19) versus COVID-19 (C19) mean time to treatment and surgery. (a) A student t-test compared mean time to treatment between the pre-C19 cohort and the C19 cohort, which was not significant (p = 0.10). (b) A student t-test compared mean time to surgery between the cohorts, which was not significant (p = 0.29).

Fig. 4.Pre-COVID-19 (Pre-C19) versus COVID-19 (C19) treatment modalities. Treatments received in the pre-C19 and C19 cohorts were compared using chi-squared analyses. There were significantly fewer surgeries during C19 (34.1% pre-C19 vs. 31.8% C19, p < 0.001). There was an increase in chemotherapy during C19 (68.8% pre-C19 vs. 69.9% C19), but this was not statistically significant (p = 0.09). There was a decrease in radiation during C19 (29.9% pre-C19 vs. 28.8% C19), but this was not statistically significant (p = 0.10). ns: not significant.

4. Discussion

This study demonstrates a statistically significant increase in locally advanced and metastatic disease during the C19 era. This could be attributed to several factors including the evolution of screening practices and treatment methodologies. In 2020, the American Cancer Society (ACS) updated the cervical cancer screening guidelines. This update emphasizes a shift to primary screening via human papillomavirus (HPV) testing every 5 years in individuals aged 25–65; when primary HPV testing is unavailable, ACS recommends utilizing cytology and hrHPV-DNA co-testing every 5 years or cytology alone every 3 years [16]. Per the United States Preventive Services Task Force (USPSTF) guidelines from 2018, for primary screening, individuals ages 21–29 should receive cytology alone every 3 years, and those ages 30–65 should receive cytology alone every 3 years, with co-testing or a primary HPV test every 5 years [16]. The 2020 guideline to increase inter-screening intervals may have also potentially impacted the overall upstaging in SCCC observed in this study; barriers to the adoption of screening guideline changes, through failure of clinician adoption of such changes, logistical challenges, or other reasons, may have led to a delay in the impact on cervical cancer disease severity.

Additionally, clinical guidelines for SCCC management differ by staging. Early-stage SCCC is usually curable with surgery alone [17]. Locally advanced SCCC is generally regarded as treatable and/or curable with surgical management and concurrent chemoradiation with intention to reduce local and distant disease recurrence [17]. During the time period investigated in this study (2018–2021), treatment for metastatic SCCC was largely considered palliative, with patients unlikely to achieve any increase in overall survival with chemotherapy or chemoradiation [18]. However, in October 2021, the U.S Food and Drug Administration (FDA) approved pembrolizumab for use in cervical cancer treatment and has since changed treatment options for Stage IVB (FIGO 2018 stage distribution) disease; yet, this change likely did not confound the data from this study as this time period overlapped less than 3 months out of the 48 months included [19].

Despite key obstacles that may have arisen during the C19 era, this study did not observe any statistically significant difference in mean time to treatment and surgery from the pre-C19 and C19 eras. Of note, however, this study describes a statistically significant decrease in the rate of surgery during the C19 era. This largely aligns with the observed decrease in early-stage disease that is managed with surgery alone and the observed increase in more advanced disease that would require other treatment modalities for curative or palliative intent. Han et al. [20] found a 23% reduction in diagnosis of Stage I cervical cancer in 2020 from 2019 during their nationwide assessment using the NCDB from 01 January 2018, to 31 December 2020; incidence data for C19 was not available during this study’s analysis and thus it is unknown if this represents an overall decrease in patients or rather a decrease in proportion of diagnosed disease [20]. Komatsu et al. [21] also identified a similar trend in Japan using the TNM stage; the study showed the number of T2 and T3 cases of cervical cancer increased in 2021 compared with 2018–2020, and the number of T1 cases decreased. Wickenheisser et al. [22] pursued a multi-site analysis, which found the proportion of patients diagnosed with Stage I cervical cancer decreased in the “pandemic & recovery” cohort (defined as 01 April 2020–31 December 2021), and those diagnosed with Stage II–IV disease increased. The same analysis found that radical hysterectomy as primary management for cervical cancer occurred less frequently during the “pandemic & recovery” cohort and was more likely to be delayed for as long as 6 weeks after diagnosis [22]. Chapman et al. [23] identified that in the year 2020, patients were less likely to receive primary surgical management and more likely to receive radiation or systemic therapy as initial treatment.

Additionally, the C19 positive rate in SCCC was only 4.29%. Our data indicates that once patients presented for care with a diagnosed cancer, the C19 pandemic did not result in delays in care. Although the C19 virus did not likely impact SCCC disease severity directly, the C19 pandemic itself rather was more likely responsible for this observed upstaging. Patients often attributed delayed care during the pandemic to logistical concerns with scheduling appointments with their providers [24]. Not only did healthcare providers decrease availability for patient services, but patients also voluntarily avoided seeking medical care. A study conducted in June 2020 found that an estimated 31.5% of patients avoided routine healthcare early in the pandemic period due to concerns about C19 [25]. These concerns included the fear of becoming infected with the C19 virus and fear of further adding to the already overwhelmed healthcare system [24]. This, as a result, changed “health-seeking behaviors” such that patients chose to avoid seeking health care providers in instances when they previously would have before the pandemic [24].

Limitations of this study mainly include the limitations of the available data on the NCDB. This version of the database did not include data regarding Vital Status, thirty-day mortality, and ninety-day mortality for patients diagnosed with cervical cancer during or after 2020; thus, this study was unable to analyze any data regarding either progression-free survival or overall survival for these cohorts. Further studies should be done to investigate these outcomes as this data becomes available for the years 2020 and 2021. Also, this dataset only contains data from patients treated at CoC-accredited facilities and, therefore, cannot encompass comprehensive national data from other non-participating institutions. Additionally, as this study was an observational retrospective study, it is not able to establish causation between the upstaging and decrease in cervical cancer screening during the C19 era. This study did not investigate factors including the patient’s geographic location, type of treatment center, socioeconomic status, and race; these factors could have played a role in the disease outcomes observed, and further studies may investigate how these social determinants of health influenced SCCC staging before, during, and after the C19 pandemic.

Strengths of this study include the amount of time that was evaluated, which, therefore, demonstrated a more thorough investigation of the future impacts of the pandemic on cancer outcomes than previous studies. This study also included both time to surgery and time to treatment in general, which further illustrates the factors that may or may not have led to the outcomes observed. Understanding these factors that likely did impact cervical cancer disease severity, including prolonged time between cervical cancer screening, and those that did not impact disease severity, including time to treatment and surgery, during this C19 pandemic era will help guide treatment and screening recommendations if another similar catastrophic global event arose in the future.

5. Conclusions

In conclusion, despite no significant differences in time to treatment and surgery, there was an increased rate of advanced stages at diagnosis of SCCC during C19. There was an observed decrease in surgery during C19, which possibly was related to the decreased number of patients with early-stage disease that could be effectively treated with primary surgery with curative intent. The increased rate of more advanced SCCC was not related to individuals with SCCC testing positive for the C19 virus. Thus, this is likely linked to the decrease in access to routine cervical cancer screening tests, diagnostic colposcopies, and healthcare visits for non-emergencies due to recommendations from large professional associations to postpone evaluations during C19. Further research is awaited as more data emerges from the C19 era regarding the impact of the C19 pandemic on SCCC stage distribution, management, and patient outcomes.

Recommendations: These results suggest the importance of continued routine health maintenance screening in the prevention of SCCC and may suggest a need to continue to recommend routine cervical cancer screening should another situation similar to the C19 pandemic arise in the future.

Availability of data and materials

The datasets generated and/or analyzed during the current study were requested and accessed from the National Cancer Database (NCDB) repository. These data may be available to others upon request and approval from the NCDB. More information about the NCDB is available at: https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/.

Author contributions

CJ, CB, SS and DY—designed the research study. CJ, CB and SS—analyzed the data and performed statistical analyses. DY and BR—provided guidance on study design. BR and DM—provided professional expertise. CJ and CB—wrote the manuscript. IR and AD—edited the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

This study was reported to the Augusta University (AU) Institutional Review Board and reviewed by AU Committee C, which determined this project does not meet the definition of human subject research under the purview of the IRB according to federal regulations. The study’s IRB reference number is 2155794-2.

Acknowledgment

This work was made possible due to the support of the Surgical Oncology Clinical Outcomes Research Group (SOCORG) at the Medical College of Georgia. Specifically, the group acquired the NCDB for use and IRB approval for the study. We especially thank Danny Yakoub, Bunja Rungruang, Daniel Milgrom, Steven Colquhoun, and Alicia Arnold for providing resources and guidance on this project. We would also like to thank the department of Gynecologic Oncology at the Medical College of Georgia for financial support in publication of this manuscript.

Funding

This research received no external funding.

Conflict of interest

The authors declare no conflict of interest.