Impact of the COVID-19 Pandemic on Colorectal Cancer Patients at a Major Brazilian Cancer Center: An Increase in Metastatic Patients upon Presentation?

Colorectal cancer (CRC) is one of the most prevalent types of cancer worldwide, and one of the major causes of cancer-related deaths1 2—and this is not different in Brazil.3 4 In the state of Bahia, Hospital Aristides Maltez (HAM) is responsible for most of the oncological care.5 According to the Brazilian National Cancer Institute (INCA), 1,480 new CRC cases were estimated in Bahia in 2020.3 A report from the Bahia League Against Cancer (LBCC, from the Portuguese Liga Bahiana Contra o Câncer) shows that 538 new cases of CRC were treated at HAM in 2019 (corresponding to 36% of the total estimated for the whole state in 2020), with similar numbers in the 2 previous years.5

The years 2020 and 2021 were peculiar due to coronavirus disease 2019 (COVID-19) in several aspects. Tackling the pandemic was prioritized. Therefore, cancer patients, including those with colorectal malignancies, may have been affected by oncological care delays. In an article published in 2020, it was recommended that resection of CRC occurred as soon as possible, depending on the availability of resources and the epidemics situation at the local level, as delaying resection could have a negative impact on survival.6 Another study, done before the COVID-19 pandemic, demonstrated a negative impact on survival when the time between CRC diagnosis and its resection was increased.7 In line with this, a systematic review published in 2020, which took into account studies from 2005 to 2020, demonstrated that a delay of 30 to 40 days in primary colon cancer resection was associated with shorter survival.8 This systematic review also showed that, in rectal cancer cases, a time interval above 7 to 8 weeks between neoadjuvant treatment and surgery resulted in decreased survival.8

The main objective of the present research is to evaluate the impact of the COVID-19 pandemic in CRC patients care at HAM. The secondary objective is to describe the characteristics of the patients treated at HAM with this malignancy.

Data from electronic medical records of HAM (a reference cancer hospital in Salvador, Bahia, Brazil) were retrospective accessed. A list of adult patients with CRC was generated using the coding system (codes C18, C19, C20) of the International Classification of Diseases, 10th edition9 (ICD-10). It was restricted to 2 periods of time: first year of COVID-19 epidemic in the state of Bahia (COVID period)10 and the 12 months prior to the pandemic (pre-COVID period). The COVID period’s list had 469 people, and the pre-COVID’s one had 443.

Sample sizes were calculated separately using the tool “sample size for % frequency in a population (random samples),” available at the OpenEpi (open source) website.11 Anticipating a frequency of 20% of metastatic disease12 and with absolute precision of 5%, the sample size was 162 (COVID period) and 159 (pre-COVID period), with 95% confidence interval.

Sample sizes were also calculated considering both groups in the same analysis, using the tool “Sample Size: Cross-sectional, Cohort and Randomized Clinical Trials,” also available on the OpenEpi website.11 Considering a bilateral confidence level of 95%, power of 80%, with a “non-exposed-to-exposed rate” of 0.94 (443/469), an estimated percentage of metastatic disease in the “non-exposed group” of 20%,12 and an odds ratio of 2, the results of sample size calculation were 177 people for the COVID group and 166 people for the pre-COVID group.

Research subjects’ selection was aleatory. The aleatory number generation tool “Gerador de números aleatórios”13 was used for this purpose.

Data was collected by a team of six people, using the RedCap software (Vanderbilt University)14 15 authorized to Fiocruz Bahia. Data extraction form is available in the Supplementary Appendix 1 (online only).

Adults that had their first appointment at HAM specifically due to CRC (either suspected or confirmed), with adenocarcinoma or carcinoma not otherwise specified histological types, at one of the following 2 periods of time: from March 18, 2020, to March 17, 2021,10 or from March 1, 2019, to February 28, 2020.

Subjects that were misclassified with one of CRC’s international codes; subjects that already concluded one step of oncological treatment at another institution, as oncological surgery, radiotherapy or systemic treatment (with the exception of those who previously received just a non-specialized surgical treatment, as for bowel obstruction).

Descriptive analysis methods were used, as absolute numbers, frequencies, proportions, measures of central tendency and dispersion. This part of the statistical analysis was done with the same system used for data collection, the RedCap14 15 authorized to Fiocruz Bahia. Comparative analysis between the groups was done with the STATA software (StataCorp LLC), version 18.0.16 For this, the Pearson’s Chi-squared test was applied.

Subjects’ confidential data leakage. Measures to avoid it were taken, such as identification of research subjects with a new coding system, and use of a special data storage system, the RedCap.14 15 Access to it was protected by an individual password for each member of the research team.

Research protocol was previously submitted to (and approved by) both ethics committees of the main institution, Fiocruz Bahia, and the coparticipant institution, HAM. The Brazilian Certificate of Presentation for Ethical Appreciation (CAAE) number is 66752823.5.0000.0040 (available for consultation at the “Plataforma Brasil” website17).

Sociodemographic and health data are shown in Tables 1 and 2.

Table 1

	Pre-COVID group	COVID group
Number of subjects	152	170
Age (years)
Median  Percentile 25–75	63 54.00–73.25	63 53.00–75.75
Gender
Female  Male	78 (51.3%) 74 (48.7%)	93 (54.7%) 77 (45.3%)
Skin color or ethnicity
Yellow  White  Native American  Brown  Black  Other	0 11 (7.2%) 0 129 (84.9%) 7 (4.6%) 3 (3.3%)	0 0 0 144 (84.7%) 26 (15.3%) 0
Marital status
Single  Married  Stable union  Separated  Divorced  Widow(er)  No data	80 (52.6%) 53 (34.2%) 0 1 (0.7%) 8 (5.3%) 11 (7.2%) 0	63 (37.1%) 74 (43.5%) 0 0 12 (7.1%) 21 (12.4%) 0
Place of birth
Brazil – Bahia  Brazil – Other states  Brazil – Not specified  Other countries	139 (92.1%) 11 (7.3%) 2 (1.4%) 0	158 (92.9%) 12 (7.1%) 0 0
Location of residency
Brazil – Bahia	152 (100%)	170 (100%)
Regional Health Nucleus (Bahia)19 of residency
Eastern*  Others  No data	78 (51.31%) 74 (48.68%) 0	88 (51.76%) 82 (48.23%) 0
Education
Analphabet  Elementary school  High school  Bachelor’s degree  Masters or Doctorate degree  No data	8 (5.30%) 85 (55.92%) 51 (33.55%) 6 (3.94%) 0 2 (1.31%)	10 (5.88%) 92 (54.11%) 59 (34.70%) 9 (5.29%) 0 0

Abbreviation: COVID, coronavirus disease 2019.

Notes: Results presented in absolute numbers (with respective percentages in parentheses). *Eastern Regional Health Nucleus (of Bahia), made up of 47 municipalities, including Salvador.19

Table 2

	Pre-COVID group	COVID group
Primary tumor site
Cecum  Appendix  Ascending colon  Hepatic angle  Transverse colon  Splenic angle  Descending colon  Sigmoid colon  Colon, with invasive lesion  Colon, unspecified  Rectosigmoid junction  Rectum  Colorectal, not specified	6 (3.9%) 0 9 (5.9%) 4 (2.6%) 7 (4.6%) 6 (3.9%) 2 (1.3%) 17 (11.2%) 3 (2.0%) 14 (9.2%) 18 (11.8%) 77 (50.7%) 2 (1.3%)	9 (5.3%) 0 16 (9.4%) 4 (2.4%) 11 (6.5% 1 (0.6%) 3 (1.8%) 25 (14.7%) 4 (2.4%) 13 (7.6%) 24 (14.1%) 71 (41.8%) 0
Histological type
Adenocarcinoma  Carcinoma, not specified	149 (98.0%) 3 (2.0%)	166 (97.6%) 4 (2.4%)
Cancer stage
Metastatic  Non-metastatic  No data	41 (27.0%) 107 (70.4%) 4 (2.6%)	58 (34.1%) 107 (62.9%) 5 (2.9%)
Cancer stage
Metastatic  Non-metastatic  Note: “No data” excluded	41 (27.7%) 107 (72.3%)	58 (35.15%) 107 (64.85%)
Metastatic, but curable
Yes  No  No data	7 (17.1%) 24 (58.5%) 10 (24.4%)	4 (6.9%) 43 (74.1%) 11 (19%)
Stage, if non-metastatic
I  II  III  No data	7 (6.5%) 47 (43.9%) 40 (37.4%) 13 (12.1%)	3 (2.8%) 40 (37.4%) 37 (34.6%) 27 (25.2%)
Cancer treatment before HAM
No  Yes, surgery	113 (74.3%) 40 (26.3%)	131 (77.1%) 39 (22.9%)
Life status in the last record
Alive  Deceased	122 (80.3%) 30 (19.7%)	137 (80.6%) 33 (19.4%)

Abbreviations: COVID, coronavirus disease 2019; HAM, Hospital Aristides Maltez.

Note: Results presented in absolute numbers (with respective percentages in parentheses).

As demonstrated in Table 2, there was an increase of 26.89% of metastatic disease at presentation from pre-COVID to COVID group (from 27.7–35.15%, after excluding subjects that were not possible to determine if metastatic or not at admission). Despite this numerical increase, the Pearson’s Chi-squared test result was 2.00, Pr 0.15 (without statistical significance).

We found that there was a numerical increase in the percentage of metastatic CRC patients at first presentation, during the first year of the COVID period, in comparison to the 12 previous months (but without statistical significance).

The pre-COVID group’s characteristics represent the usual CRC population that seeks cancer care at HAM. In this group, genders were well balanced, with a median age of 63 years old, which is in line with international data.18 There was a predominance of brown skin color and low level of education. Although the Hospital is located at the main city of the state, in the far east, there was a similar representation of patients that came from its Regional Health Nucleus (Eastern) and others. This demonstrates the scope of the hospital, which serves people from all parts of Bahia (which is divided in 9 regional health nuclei).19 The high percentage of people with rectal cancer (half of the sample) calls for attention. This may be due to the fact that radiotherapy, which is frequently used during the treatment of this part of the bowel,20 is available at HAM, but not at other health centers across the state. Representation of metastatic CRC in the pre-COVID group was higher than international data—27% in this versus 20% in USA data.12 This may represent a lost opportunity to offer curative treatment, as, in most cases, CRCs are no longer curable when metastatic.20

The group of patients who came to HAM and had anatomopathological diagnosis established during the first year of the COVID-19 epidemic in Bahia (COVID group)10 presented similar epidemiological data in relation to the previous one. Nonetheless, there was an increase of 26.89% of people presenting at the metastatic stage (from 27.7% in the pre-COVID to 35.15% in the COVID group)—although the Pearson’s Chi-squared test did not show statistical significance for this percentage increment. There are some hypotheses that may explain this finding. The sample size was inferior to the one that was calculated, when taking in consideration both groups together, which may have underpowered the study for this evaluation (of the difference of percentages between groups). Another important point is that the representation of metastatic patients since the beginning, in the control group (pre-COVID), was higher than expected (27.7%, instead of 20% according to international data12). This might have influenced the results too. If this higher percentage was used to calculate the sample size, it would be larger. That said, there is still the possibility that the current result (without statistical significance) reflects the reality.

A similar study done at another Brazilian hospital, which is a reference center in cancer treatment in the state of São Paulo, found an increase in the proportion of new cases of advanced CRC, between March and July 2020, compared with the same period in 2019.21 This result reached statistical significance.

According to another study, that evaluated CRC screening programs in 29 countries, there was a decrease of CRC screening during the COVID-19 pandemic.22 Moreover, this research states that, if measures to correct this problem are not implemented, this could lead to an increase in cases of CRC, as well as greater mortality from this malignancy.

The results of those two other studies21 22 are in line with the findings of this present research, which highlights the importance of cancer care during special periods, such as the COVID-19 pandemic, as well as in the subsequent years.