Introduction

Colorectal cancer (CRC) is the third most prevalent malignancy and the second leading cause of cancer death worldwide [1]. The Asian population ranks first in terms of the number of incidences and mortality of CRC compared to other populations worldwide [2]. The trend of CRC is expected to increase across different populations in the world, both in the West, including the United States, Canada, and Australia, and in Eastern populations, such as China, and Korea [3].

In Indonesia, 34,189 new cases of CRC were estimated in 2020, ranking the fourth most prevalent cancer incidence in the country [4]. The CRC mortality rate is the fifth largest, reaching 6.7 deaths per 100,000 person-years [1]. In Yogyakarta Province, one of 34 provinces in Indonesia with the highest cancer prevalence according to the National Basic Health Survey [5], CRC incidence was recorded to be 7.40 per 100.000 population [6]. Colorectal cancer cases in Indonesia were dominated by males (54%), with a peak incidence at 50–54 years of age [7]. The local cases reported were also dominated by advanced tumors (30–40%) [6,8], all of which were associated with worse prognosis and decline in quality of life.

Colon and rectal cancer are frequently regarded collectively as colorectal cancer due to the assumption that they arise from organs that have similar morphological and molecular characteristics. However, the colon and rectum possess various differences in terms of their histological and anatomical origin, physiological properties, and genetic pathways, attributing to the different risks and survival among the two cancers [9,10]. The current estimate in the Global Cancer Observatory shows a higher incidence of colon compared to rectal cancer (11.4 vs. 7.6 per 100,000 population worldwide) [1].

Histological differences might explain the lower incidence of rectal cancer, including the better defense provided by the mucin-secreting goblet cells that present in higher quantities in the rectum, resulting in the lower cases reported on the respective site [10]. In addition, colon cancer has a flat morphology, making it more frequently diagnosed at a later stage and with poor differentiation [11]. Alteration in mismatch repair (MMR) genes in the germinal cells might result in hereditary nonpolyposis CRC (HNPCC), which is predominant in the distal colon and rectum, and involved familial adenomatous polyposis (FAP). When this occurs in somatic cells, this would result in microsatellite instability (MSI) that is responsible for the oncogenesis of sporadic CRC and is predominant in the proximal colon [10,11]. The different characteristics result in different response to treatment and prognosis. Patients with lower MSI, usually in cases of rectal cancer, have a better prognosis with adjuvant chemotherapy and targeted therapy [11]. Another study observed a worse prognosis of rectal cancer in the presence of complications compared to colon cancer [12].

The difference in risks between colon and rectal cancer is also influenced by other factors such as age, sex, and location of residence [13–15]. Distal colon and rectal cancer, with a higher proportion of chromosomal instability (CIN)-high tumors, are usually predominated by male sex and younger age. On the contrary, proximal colon cancer which is more frequently present with MSI-high tumors, are predominant in female and older age [11].

The disparity of risk among different sexes might be attributable to sexual dimorphism, due to the numerous mechanisms of protective effects of estrogens against colon and rectal cancer in females [14]. However, previous studies have suggested the existence of gender-specific risk of colon and rectal cancer due to the difference in anatomical characteristics, resulting in lower sensitivity of screening tools among females, hence worse presentation and prognosis [16]. According to the more recent consensus molecular subtypes (CMSs) stratification by the CRC Subtyping Consortium in 2015, females with right-sided colon cancer were more frequently diagnosed with CMS1 and were also associated with higher histological grade and worse prognosis after relapse. Conversely, male, who were more frequently diagnosed with left-sided MSI2 tumors, has notably better prognosis [13].

The risk of rectal cancer is suggested to be 1.5 higher in males, while the risk of colon cancer is 1.2 in females [9]. However, the incidence of CRC consistently demonstrated a higher burden among males compared to females, both in colon and rectal cancer [3,15]. At present, the worldwide incidence of colon cancer (10 per 100,000) almost doubles those of rectal cancer (5.6 per 100,000) among women (H. Sung et al., 2021). Moreover, the disparity among sexes in rectal cancer is more than twice that observed in colon cancer (male-to-female ratio = 1.75 vs. 1.31, respectively) [17].

Age also contributes to other major non-modifiable risk factors for both colon and rectal cancer, with a significant increase in incidence and mortality after the age of 50. Around 60% of global CRC incidence and 50% of its mortality presented between the ages of 50 to 74 years [18,19]. Although the reason is yet fully understood, the proportion of early-onset CRC (EOCRC) is increasing and recorded higher, especially among Asian countries [19]. EOCRC is more frequently present at the rectal location, with a more advanced stage, and more unfavorable histopathology, worsening its prognosis [20,21]. In younger age, MSI occurrence in rectal cancer strongly indicates a mutation associated with HNPCC or Lynch syndrome that usually shows MSI-high status [11,22].

Several other modifiable lifestyle factors such as high BMI and physical inactivity are frequently associated with an increased risk of colon cancer, but not rectal cancer, emphasizing the difference in etiology between the two tumor sites [11]. Living in urban areas has been found to be associated with a higher risk of different types of cancer [23]. CRC, however, were among the cancers with a higher incidence in rural areas, owing to the lower coverage of screening in rural areas [23–25]. Another study suggests that the association of rurality on the incidence of CRC might also be associated with other factors such as sex, race, and tumor location [23].

At present, the difference in risk between colon and rectal cancer and its association with differences between sex, age, and rurality of residence has been studied in several different populations [16,26–28]. This, however, has not been widely studied in Indonesia. In the present study, we aim to examine the influence of sex, age, and different risks attributable to location of living to the risk of colon and rectal cancer and to describe the trend of CRC in the local population in Yogyakarta Province.

Materials and methods

Results

Characteristics of colon and rectal cancer patients

The present study performed an analysis of the demographic and clinical characteristics of a cohort of 1,295 CRC patients drawn from the Yogyakarta PBCR database, as detailed in Table 1. More cases in the cohort were present with rectal cancer (n = 893, 68.96%). A male predominance was present in the rectal cancer subgroup (n = 504, 56.44%). In the case of colon cancer, a near-equivalent sex distribution was observed, with 49.75% males (n = 200) and 50.25% females (n = 202). The mean age at presentation for the entire cohort was 57.66 years, with a slightly younger mean age noted in colon cancer patients (mean = 55.99±13.43) compared to those with rectal cancer (mean = 58.27±13.63, p = 0.06). The age of subjects ranged from 14 to 93 with most subjects diagnosed in the 50–59 age group, both in colon (n = 119, 29.60%) and rectal (n = 253, 28.33%) cancer. The proportion of young-onset cases (<50 years) or EOCRC were 25.41%, comprises of 121 cases in the colon (30.10%) and 208 cases (23.29%) in the rectal cancer group. In both groups, more than 68.57% of subjects resided in urban areas within the three districts.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of CRC patients in Yogyakarta population-based cancer registry (n = 1,295).

https://doi.org/10.1371/journal.pone.0301191.t001

Comparison of colon and rectal cancer by sex and age

We presented the result of the log-binomial regression analysis for colon and rectal cancer in Table 2. In this study, female patients exhibit 20% higher odds of colon cancer as compared to males (RR = 1.20, 95%CI = 1.02–1.44, p = 0.025). Moreover, within the stratified age group, subjects aged 30–39 years old have an 87% higher likelihood of colon cancer (RR = 1.87, 95%CI = 1.10–3.19, p = 0.021). Our finding observed no discernible disparity in the odds of developing colon cancer among those residing in urban compared to rural areas in this population (RR = 0.98, 95%CI = 0.82–1.17, p = 0.830).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Crude and adjusted relative risk of colon and rectal cancer among CRC cases in Yogyakarta.

https://doi.org/10.1371/journal.pone.0301191.t002

Our observation indicates an 8% lower likelihood of rectal cancer among female subjects as compared to the males (RR = 0.92, 95%CI = 0.85–0.99, p = 0.027). In contrast to our finding on the colon cancer subgroup, subjects aged 30–39 and 40–49, respectively, demonstrate a 25% and 18% decrease in the odds of rectal cancer (RR = 0.75, 95%CI = 0.60–0.93, p = 0.008 and RR = 0.82, 95%CI = 0.69–0.98, p = 0.027). Similar to their counterparts in the colon cancer subgroup, residing in urban areas also did not significantly elevate the odds of rectal cancer (RR = 1.01, 95%CI = 0.93–1.10, p = 0.831).

In the multivariable analysis, sex remains significantly associated with the likelihood of colon cancer, in which female subjects demonstrated a 21% increase in odds, controlling for age and location of residence (ARR = 1.21, 95%CI = 1.03–1.42, p = 0.022). Subjects in the 30–39 age group have an 88% significant increase in the probability of developing colon cancer compared to those aged 80 and older (ARR = 1.88, 95%CI = 1.11–3.18, p = 0.020). Despite not being statistically significant, we observed higher odds of developing colon cancer in all age groups when compared to the oldest group, controlling for sex and location of residence.

A contrasting, and consistent, result was evident in the multivariable analysis result of the rectal cancer subgroup. Female subjects displayed 7% lower odds compared to males, controlling for age and residence (ARR = 0.93, 95%CI = 0.86–1.00, p = 0.043). Subjects in the 30–39 and 40–49 age groups displayed similarly reduced odds of developing rectal cancer when controlling for age and residence (RR = 0.75, 95%CI = 0.61–0.93, p = 0.009 and RR = 0.82, 95%CI = 0.69–0.97, p = 0.022).

The age-stratified trend of colon and rectal cancer

The analysis of dynamics on the annual change in the trend of the overall CRC as well as colon and rectal cancer is presented in Fig 1 for cases younger than 50 years (left), age group 50–64 (middle), and age group 65 and older (right). In subjects younger than 50, we observed a significant increase in the trend of CRC from 2008 to 2017 (annual percent change/APC = 7.38%, 95%CI = 2.93–13.03%). Similarly in colon cancer, the trend of incidence increases 8.15% annually on average (95%CI = 1.62–16.94%). In rectal cancer, however, the increasing trend from 2008–2017 (APC = 9.71%, 95%CI = 6.49–27.33%) was followed by a non-significant decrease from 2017–2019 (APC = -17.23%, 95%CI = -17.23–32.50%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

Average annual change of colorectal cancer (blue), colon cancer (red), and rectal cancer (green) incidence among subjects aged under 50 years old (left), 50–64 years old (middle), and 65 years old or older (right). CRC incidence demonstrated an upward trend across the three different age groups. The increase of rectal cancer is higher among the under 50 and 50–64 age group, compared to colon cancer. The increase of colon and rectal cancer however is similar in the 65 and older age group.

https://doi.org/10.1371/journal.pone.0301191.g001

A notable significant increase in CRC incidence was apparent in cases aged 50–64 (Fig 1 (middle), APC = 16.93%, 95%CI = 13.39–22.84%), from 4.79 per 100,000 population in 2008 to 23.61 per 100,000 population in 2019. In the same age group, a similar trend was observed for colon (APC = 15.30%, 95%CI = 7.75–29.11%) and rectal cancer (APC = 17.08%, 95%CI = 13.02–23.32%). Despite not being as high as the 50–64 age group, we also observe a similar increase in the trend of CRC in the 65 and older group (APC = 11.38%, 95%CI = 7.64–16.76%), from 9.22 per 100,000 population in 2008 to 24.73 per 100,000 population in 2019. In this age group, the increase in trend was very similar between those in the colon (APC = 10.90%, 95%CI = 1.63–24.82%) and rectal cancer (APC = 10.95%, 95%CI = 4.09–21.26%).

Discussion

In the present study, we observed a higher proportion of rectal cancer (68.96%) in the studied population. This finding is contradictory to the global data showing the higher incidence cases of colon compared to rectal cancer [1]. Although this might be attributable to the high proportion of cases with unspecified location (ICD-10 C18.8–9) excluded from the study (18.9%), our result is similar to the proportion of colon and rectal cancer in previous studies coming from low- and middle-income countries [32,33]. This proportion is also similar to Vietnam (59.3%) and North Korea (51.7%) which reported a higher number of incidence rectal cases compared to colon [1].

Our data shows higher odds of colon cancer and a lower likelihood of rectal cancer among females. This is also supported by previous data suggesting a 1.2-fold risk of colon cancer and 0.5 risk of rectal cancer among females [9]. Another study also highlights the elevated risk of colon cancer among the female sex, especially in the proximal colon [34]. This difference could be explained by the presence of higher insulin susceptibility in the colon. Changes in the expression of insulin receptors trigger abnormal cell proliferation that may induce malignancy [35]. The high concentration of estrogen in females further promotes this mechanism, resulting in a higher risk of colon cancer in women [34,36].

We observed 30% of patients with colon cancer and 23% with rectal cancer diagnosed before the age of 50. This finding is consistent with a previous observation performed in the Tanzania population [32]. This is also in line with our finding, showing an increase in the overall trend of CRC and colon cancer, and a decline in the trend of rectal cancer in the last three years among the younger group. This overall proportion of EOCRC cases in this study (25.4%) is comparably higher than previously reported in the Western population which ranged less than 10% as reported from Milan, Italia and Florida, USA [19,27]. This proportion was also higher than reported from neighboring Southeast Asian countries, ie. 14.5% in Northern Malaysia and 5.7–13.4% in Singapore, as well as the 30–40% proportion previously reported from pathology and hospital-based data from Indonesia, demonstrating the higher incidence of EOCRC in the Asian population [8,37–40].

The increasing risk of CRC among the younger population has been observed consistently among different populations [27,41–44]. The higher frequency of Lynch syndrome in the local population possibly contributes to the higher EOCRC compared to the Western population. The proportion of cases with probable Lynch syndrome as the underlying cause in the local population is reported to be 6-fold of those reported in western countries [45]. Another previous study from Indonesia also reported the proportion of Lynch syndrome among EOCRC almost two-fold of those with older age of onset [42].

The risk of developing EOCRC was associated with lifestyle factors such as low physical activity, sedentary behavior, smoking and high processed meat consumption [46–49]. Despite the changing profile of gut-microbiota along with ageing, higher presentation of microbiota is associated with both the initiation, promotion, and progression of CRC oncogenesis among younger cases [50]. Although not observed in the present study, the association of age on the risk of colon and rectal cancer is stronger among women, who have higher odds of presenting with tumor characteristics associated with worse prognostic [34].

Although not further observed in the present study, caution needs to be taken when interpreting the gap of incidence between EOCRC cases and the older age groups. It has been observed that EOCRC might have longer symptom duration, time to diagnosis and treatment as compared to older cases, especially in cases with less advanced stages [51]. The fact that CRC screening is currently only recommended to those with known familial syndrome and familial history [52], might undermine the actual burden among younger age due to low diagnostic coverage.

The present study is the first to use population-based data to evaluate the risk of colon and rectal cancer in the local population, which can provide stronger evidence of the actual disease burden as compared to hospital-based data as previous reports from Indonesia. This study provides evidence of the high and increasing burden of young-onset CRC in the local population, suggesting the need to also focus on the younger population in an effort to the control cancer burden. The inclusion of only data from the Yogyakarta Cancer Registry in this study warrants further exploration utilizing data from a wider scope of population to capture the condition of other provinces in Indonesia that possess variations in lifestyle, diet, as well as environmental risk factors.

Conclusions

Our findings highlight the difference in relative risk of colon and rectal cancer, especially in the younger age group. Some of these patterns need more detailed exploration to understand other risk factors that might influence this trend. The fact of higher risk of young onset colon cancer, especially in between age 30–49 years old, may serve as recommendation increase effort and awareness for screening especially towards young population. A nationwide investigation may provide better and more comprehensive data for calling for actions in various levels of authority in performing CRC screening programs and further promotion of healthcare services to provide better management for patients with CRC.

Supporting information

Acknowledgments

The authors thank Yogyakarta population-based cancer registry and Dr. Sardjito hospital-based cancer registry for providing data used in the present analysis.