Introduction

Cancer is one of the leading causes of morbidity and mortality worldwide ¹. Cancer incidence has increased considerably in most countries ², and is projected to increase 47% by the year 2040, taking 2020 as the baseline ¹. Cancer causes a high mortality rate yearly, particularly in low- and middle-income countries. Also, deaths in these countries are expected to increase as a result of aging populations and declines in deaths from other causes ³.

The burden of cancer represents an important challenge for Latin American countries, including Colombia, which face an increase in incidence and mortality ⁴, in the context of insufficient population-level interventions to control this disease ⁵. In this sense, population-based cancer registries (PCRs) are fundamental inputs for designing cancer control strategies; PCRs allow the monitoring and evaluating the event over time in geographically defined populations ⁶. In Colombia, progress has recently been made in the quality of the PCRs, as evidenced by the inclusion of the PCRs of Bucaramanga, Cali, Manizales and Pasto in the most recent volume of Cancer Incidence in Five Continents (CI5-XI) ^7,8. Based on the information provided by the RPCs, it has been established, for example, that prostate, breast, stomach and cervical cancers are responsible for more than 50% of the burden of this disease in Colombia ⁹.

The Population-Based Cancer Registry of Manizales (PCRM) has operated since 2002, generating comparable, reliable and valid information on the epidemiological behavior of cancer in the city ¹⁰. Thus, in Manizales, patterns of incidence and mortality of cancer for the five years of 2003-2007 have been previously analyzed ¹¹, and a general incidence rate lower than that estimated for Colombia and a mortality rate higher than the national estimate stand out. In analyzing these rates, one element that has acquired relevance is the evaluation of changes in their trends ¹². This allows, among other things, to identify priority areas for intervention and evaluate the impact of cancer control programs.

It is necessary to analyze the cancer situation in Manizales with updated data and complementary statistical methods. This research aimed to describe trends in cancer incidence and mortality in Manizales between 2008 and 2017, based on data from the ' 'city's PCR.

Materials and Methods

Type of study

A population-based descriptive study on trends in cancer incidence and mortality in Manizales, Colombia, from 2008-2017.

The PCRM covers the urban and rural areas of Manizales. The city has 434,403 inhabitants, according to the 2018 National Population and Housing Census (CNPV) ¹³. The population structure of Manizales is of regressive type (Figure 1), with an aging index of 79 older adults for every 100 children under 15 years old ¹⁴.

Figure 1 Population structure by age and sex (population pyramid), Manizales, 2018. Source: own elaboration based on data from the National Population and Housing Census, DANE 2018. 

Results

Incidence and mortality by all cancer sites

A total of 11,188 incident cases were recorded, 56.7% (n= 6,348) in women. The median age at diagnosis was 66 years (IQR= 19 years) for men and 61 years (IQR= 23 years) for women. 0.8% of cases occurred in the pediatric population (under 15 years of age). Figure 2 summarizes the incidence rates and the distribution of cases by age group in men and women; as expected, it was observed that the older the age, the greater the cancer burden. It is noteworthy that in women,, the incidence rises at an earlier age than in men.

Figure 2 Cancer incidence rates and distribution of cases by age groups in (a) men and (b) women, Manizales, 2008-2017. Blue columns represent the number of cases, and red lines represent age-specific incidence rates. Note: cases with unknown ages were excluded from the graph. 

For the five-year period 2008-2012, there were 2,186 (41.6%) and 3,072 (58.4%) new cases in men and women, respectively. Between 2013 and 2017, 3,276 (55.2%) new cases occurred in women and 2,654 (44.8%) in men. The age-standardized incidence rate in both periods was slightly higher in females, with male-to-female incidence rate ratios of 0.89 (95% CI= 0.867-0.914) and 0.99 (95% CI= 0.985-0.995) for the five-year periods 2008-2012 and 2013-2017, respectively

Regarding mortality, between January 1, 2008, and December 31, 2017, 4,996 cancer deaths occurred in people residing in Manizales, 52.6% (n= 2,628), 2,260 in women in the first five-year period, and 2,736 in the second. During the five years 2008-2012, there were 1,219 deaths (53.9%) from cancer in women and 1,041 (46.1%) in men. Between 2013 and 2017, the sex distribution was 1,409 (51.5%) and 1,327 (48.5%) deaths in women and men, respectively. Of all deaths during the decade, 0.9% occurred in children under 15.

Discussion

This study represents an update of previously published data ¹¹ on the epidemiological behavior of cancer in Manizales, including a more extended period of incidence and mortality and a statistical analysis of trends. Taking the five-year period 2003-2007 as a baseline ¹¹, a general increase in cancer incidence in the city was observed, a situation that is consistent with regional and national projections ^4,20. This may be partly due to the observed aging of the population of Manizales ¹³. However, when scrutinizing the data for the decade 2008-2017 discriminated by sex, important differences were observed: at the end of the study period, the overall incidence rate in women decreased and equaled that of men, from a significant decrease in the rates of cervical, stomach, oral cavity, lung, gallbladder and Hodgkin's lymphoma cancers. Likewise, in the period studied, there was a slight decrease in cancer mortality for women, at the expense of lower mortality in cervix, stomach, esophagus, and connective tissue cancers. In contrast, overall cancer incidence and mortality did not decrease in men. This, together with the finding of a higher concentration of cases diagnosed at older ages in men compared to women, indicates the need to strengthen cancer control strategies with a differential approach by sex, in particular by promoting the participation of men in screening and early detection programs.

Breast, prostate, stomach, thyroid, lung, colon, and cervical cancers contributed more than half of all cancer cases in Manizales, which is why we will discuss them in depth below.

Regarding breast cancer, its incidence in Manizales is lower than that estimated by GLOBOCAN for South America (42.3 Vs. 56.4, respectively). Likewise, the figures are slightly lower than those estimated for Colombia (42.3 vs. 48.3) ¹ and Cali (42.3 vs 48.0) ²¹, but higher than those for Pasto (42.3 vs 27.7) ²². The re-estimation of the incidence rate for the period 2008-2012 in Manizales is higher than that reported in CI5-XI, and is much closer to the figures reported by Bucaramanga and Cali for that period ⁸.

Despite the use of organized population-based screening for breast cancer in women since 2013, in this period, no significant achievements in the fight against this neoplasm are perceived: this remains the most frequent and deadly cancer in women. Contrary to what is expected for a tumor with an organized screening strategy, breast cancer mortality did not decrease between 2008 and 2017; even worse, it presented a slight increase compared to the mortality of 2003-2007 (10.3 Vs 8.3 respectively) ¹¹. The non-reduction of mortality from this cancer in Colombia has been attributed to the low coverage of screening in the population, problems in the quality of mammography, and low access to diagnosis and timely treatment ²³. Thus, it is a priority to address these problems and at the same time, strengthen health education, remove health system barriers, and investigate the particularities of this cancer in Colombian women.

Prostate cancer, the most frequent neoplasm in men, did not experience major changes in incidence and mortality during this period. However, regarding the 2003-2007 period, its incidence increased, possibly as a result of the screening activities initiated in 2013 in Colombia and the aging of the Manizales population ¹³. Interestingly, the incidence of this cancer in Manizales is lower than that reported in Cali (46.5 Vs 65.6) ²¹ and South America (46.5 Vs 62.5) ¹, a phenomenon possibly attributable to a lower proportion of Afro-descendant population in Manizales; However, other factors, such as those related to screening coverage and case collection by the PCRM, could be related to this lower incidence, which merits further investigation. On the other hand, in Manizales there was no evidence of a decrease in mortality from this cancer, unlike what was found in Cali over a longer study period ²⁴. Thus, it is a priority to consolidate screening activities, early detection and timely and comprehensive management of cases to achieve significant improvements in mortality from this cancer.

In Manizales, as in many regions of the world, a pronounced decrease in cases of stomach cancer has been observed in recent years, which has been attributed mainly to the decline in the prevalence of chronic Helicobacter pylori infection ²⁵. However, the incidence and mortality of gastric cancer remain high in this population, mainly in men, in whom the incidence rate is twice that of women, and the mortality rate is three times that found in women, a difference that was accentuated in the period 2013-2017 from the decline in cases and deaths in women. Thus, it is of interest for future studies to analyze the possible differences by sex in exposure to risk factors of various kinds that could explain the differences observed: the significant decrease in incidence in women but not in men. The significant decline in mortality from this cancer in women but not in men could be related to the existence of cultural factors linked to differential knowledge, attitudes and practices about the disease that could favor earlier diagnosis or better adherence to treatment in women. In addition, the adoption of screening programs based on digestive endoscopy, implemented in countries with a high frequency of the disease, could be considered by Colombian health authorities to impact mortality and improve survival ^26,27.

The incidence and mortality rates for cervical cancer showed a reduction of close to 50%, taking the period 2003-2007 as the baseline. This disease went from being the leading cause of death from cancer in women for 2003-2007 ¹¹ to the sixth for 2013-2017. This downward trend has also been observed in many Latin American countries ²⁸, as well as in other regions of the world ¹; However, its incidence is still above the elimination limit defined by the WHO (4 cases per 100,000 women). Such achievements could be related to improvements in the socio-economic conditions of the population, prevention and treatment of sexually transmitted infections and, above all, screening and timely management of pre-neoplastic lesions. The vaccination already implemented against HPV will further reduce incidence and mortality in the upcoming years.

Colon and rectal cancer have remained an important cause of morbidity and mortality in Manizales, without major variations in the studied period. In contrast to what has been evidenced in the world ¹, women presented incidence rates slightly higher than men in the first five years of the study period, although not statistically significant. Promoting healthy lifestyle is a crucial element for the control of this disease since a sedentary lifestyle, overweight and obesity, excessive alcohol consumption, smoking and consumption of red meat and/or processed meats are recognized risk factors for the development of this cancer ²⁹. It is expected that the screening activities recently adopted in Colombia may have an impact in the coming years, mainly in terms of mortality.

Lung cancer continues to be a very frequent and lethal neoplasm, particularly in men: for 2013-2017 the incidence and mortality rates in men were twice those of women. Smoking, responsible for 80% of lung cancer deaths in Colombia ³⁰, has a high prevalence in the country, especially in men ³¹. This situation may be related to the differential affectation by sex observed in this research. This problem makes it necessary to strengthen preventive actions aimed at controlling smoking with a differential approach by sex, some of which are contained in the WHO MPOWER strategy ³².

Previous local research had already emphasized the high frequency of thyroid cancer, particularly in women ³³. In the present investigation, the incidence rate of this cancer increased and doubled the rates observed in Pasto (22) and Cali ²¹. This increase, also observed in the world ¹, has been attributed, in part, to overdiagnosis of this cancer due to greater availability of specialized personnel and an improvement in health technologies, specifically, ultrasound; this could result in interventions of limited benefit for patients ³⁴. In Manizales, carcinogen emissions from industries have also been linked to a possible real increase in cancer incidence ^35,36. However, despite the marked increase in incidence, mortality from this cancer remains low.

Lip, oral cavity, and pharyngeal cancer, another phenomenon related to smoking, also showed differential effects by sex, similar to those observed with lung cancer: while incidence and mortality increased in men, the incidence decreased in women. The increase in the incidence of CNS neoplasms is mainly due to the recent uptake of benign tumors by the PCRM.

The information presented here is expected to contribute to decision-making in clinical settings, to expanding knowledge of cancer epidemiology in Colombia and Latin America, and, above all, to prioritizing population-based interventions and the definition of public health policies for cancer control.

One of the limitations of this study is the considerable number of DCO cases for some types of cancer, probably explained by the late access to death data and the denial of access to data consultation by some sources consulted by the PCRM, such as IPS and diagnostic centers. These results call for additional efforts by the PCRM to improve this quality indicator through a "tracing back" strategy ³⁷. Notably, for the five-year period 2013-2017 the percentage of DCO cases is lower, reflecting an improvement in the quality of the PCRM processes and establishing a favorable outlook concerning this topic.

On the other hand, the time considered for this study (10 years) may be still too short to detect significant changes in the rates of some cancers. In addition, changes in data quality indicators, specifically the decrease in DCO cases in the last period, may reflect an improvement in the completeness of the PCRM, which should lead to a cautious interpretation of trend results. DCO cases generally have less detail on tumor variables, so some common locations in DCO cases, such as lung, liver, or unspecified sites, may correspond to primary tumors in other locations. For example, concerning lung cancer in women, the significant decrease in the trend could be explained by a better specificity in topographic coding resulting from the reduction in the percentage of DCO cases between periods; that is, an overestimation of incidence could have been observed in the first period for some cancer sites that often correspond to DCO cases, especially for neoplasms with high lethality ³⁷.

It is a priority to strengthen research on this topic, as there are still several gaps in knowledge in our context. In this sense, it is essential to work collaboratively with the different actors that are part of the health system because their human and documentary resources are essential for developing PCR research activities in Colombia.

Conclusions

The aging population of Manizales has contributed to an increased cancer burden in the city.

In terms of incidence and mortality, progress in the fight against cancer is still discrete and has been concentrated mainly in women, making it imperative to strengthen a differential approach by sex when planning strategies for cancer control.

It is a priority to consolidate screening actions for certain cancers, reinforce strategies to promote healthy lifestyles, advance in the effective control of smoking, emphasize early detection and timely management of cases, remove heath system access barriers, and encourage local cancer research.