The COVID-19 pandemic significantly impacted healthcare systems worldwide, particularly impacting cancer care. Italy, one of the first Western countries severely affected, experienced disruptions in cancer screening, diagnosis, treatment, and emergency surgery. Preoperative delays increased, potentially leading to avoidable cancer-related deaths. This study aimed to evaluate the impact of the COVID-19 pandemic on the surgical treatment of colorectal, gastroesophageal, and pancreatic cancers in Italy, comparing outcomes from 2019 (pre-pandemic) and 2020 (pandemic year). The study sought to determine if the pandemic altered the stage at which these cancers were resected and to assess its effects on perioperative and postoperative outcomes. Understanding the pandemic's consequences on cancer care is crucial for future pandemic preparedness and improving patient outcomes.

Existing literature suggests that the COVID-19 pandemic led to disruptions in cancer care, including delays in diagnosis and treatment. Studies have reported increased preoperative delays for oncological surgeries and projected an increase in cancer-related deaths due to these delays. The extent of these disruptions in Italy and their impact on various gastrointestinal cancers have been partially investigated. This study contributes to a more comprehensive understanding of the impact of COVID-19 on gastrointestinal cancer care in Italy, focusing on specific perioperative and short-term oncological outcomes not thoroughly investigated in previous studies.

This multicentric retrospective cohort study involved 62 Italian surgical divisions and included adult patients who underwent surgery for colorectal, gastroesophageal, and pancreatic cancers between January 2019 and December 2020. Ethical approval was obtained, and the study was registered at ClinicalTrial.gov (NCT04686747). Informed consent was waived due to the anonymized nature of the data. Data on demographics (age, sex, BMI, ASA score, CCI score), preoperative outcomes (urgent diagnosis, neoadjuvant treatments, time from diagnosis to surgery), perioperative outcomes (urgent resection, resectability, minimally invasive surgery, length of hospital stay, postoperative complications), and early oncological outcomes (adjuvant therapy, lymph node retrieval, positive nodes) were collected and compared between 2019 and 2020. Tumors were categorized into early and advanced stages based on the AJCC Cancer Staging Manual, 8th edition. Statistical analysis included Student's t-test, Cochran-Mantel-Haenszel test, and chi-square test. Missing data were handled using listwise exclusion. The study also analyzed data by quarter to better understand the pandemic's temporal impact.

A total of 8250 patients were included (4370 in 2019, 3880 in 2020). The rate of patients with advanced-stage disease at surgery was not significantly different between 2019 and 2020 (49% vs 51%, P=0.25). However, analysis by quarter revealed a non-significant trend towards higher rates of advanced-stage cancer resections in the second half of 2020 compared to the same period in 2019 (P=0.05). Patients diagnosed in 2020 had a significantly higher Charlson Comorbidity Index (CCI) score (5.38 ± 2.08 vs 5.28 ± 2.22, P=0.04). The pandemic led to a significant increase in urgent diagnoses (24.2% vs 20.3%, P<0.001), neoadjuvant treatments (23.9% vs 19.5%, P<0.001), and urgent colorectal cancer resections (9.3% vs 7.4%, P=0.005). The mean time from diagnosis to surgery increased significantly in 2020 (64.2 days vs 56.8 days, P<0.001). Overall surgical activity decreased by 5.2% in 2020 (P<0.001), most significantly during the first two quarters. The conversion rate to open surgery decreased significantly in 2020 (7.2% vs 9.2%, P=0.01), and the mean length of hospital stay was shorter (11.6 days vs 12.1 days, P=0.04). Interestingly, the rate of positive lymph nodes increased significantly in 2020 (9% vs 7%, P<0.001), while other pathological outcomes remained largely unchanged.

While the pandemic did not significantly increase the proportion of patients with advanced-stage gastrointestinal cancers at the time of surgery, several perioperative and postoperative outcomes were negatively impacted. The increased CCI score, urgent diagnoses, neoadjuvant treatments, and shorter length of stay suggest that patients presenting with cancer in 2020 were potentially sicker, potentially reflecting delayed diagnosis and management. The increased rate of positive lymph nodes warrants further investigation. This finding could be related to the increased proportion of urgent surgeries or the delayed time between diagnosis and surgery, which could have allowed for disease progression. The reduced surgical activity during the early pandemic phase is consistent with other studies and reflects the prioritization of COVID-19 patients and resource limitations. Although the rate of minimally invasive surgery did not significantly change, this may be confounded by a shift in patient selection for MIS in 2020 and should be further examined.

The COVID-AGICT study demonstrates that while the pandemic did not notably affect the stage at which gastrointestinal cancers were resected, it substantially altered perioperative and postoperative outcomes. The increased rate of positive lymph nodes, alongside other observed changes, highlights the need for proactive strategies to mitigate the negative impacts of future healthcare crises on cancer care. Maintaining access to timely diagnosis, treatment, and screening programs during future pandemics is vital to preserving patient outcomes.

This study's retrospective design and reliance on existing data are limitations. The study did not account for patients who did not undergo surgery in 2020, potentially affecting epidemiological interpretations. Data granularity was limited, and the high proportion of colorectal cancer patients (69%) might have influenced the results. The relatively short study period might not fully capture long-term oncological effects. Future studies should utilize prospective designs and investigate longer-term outcomes.