Abstract
Background:
Early cancer detection is crucial for improving patient prognosis and survival. However, in Latin America, the implementation of screening programs faces economic and logistical barriers.
Design and methods:
This study evaluates the utility of an early detection program in a private clinic in Peru. A prospective study was conducted at the clinic’s Oncology Center between 2017 and 2019. The sample comprised 31,057 patients who underwent routine examinations, including mammograms, biopsies, and blood tests. Patients with a previous diagnosis of primary cancer were excluded. Data collected were analyzed using descriptive statistics and the Chi-squared test to assess the effectiveness of the early detection service.
Results:
The annual prevalence of cancer diagnosis among apparently healthy patients was 0.7%. Breast cancer was the most frequently diagnosed type (40.7%), with a significant difference observed between cases with positive (50.0%) and negative (2.3%) screening results. Most cases were in early clinical stages (stages 0 and 1, 58.4%), with a homogeneous distribution across groups (p = 0.570).
Conclusions:
The preventive service proved to be highly effective in early detection of cancer, especially in patients with no apparent symptoms. However, limitations were identified, such as the incidence of false negatives in some tests.
Keywords: early detection, breast cancer, preventive services, clinical stages, early diagnosis
Prevention relevance statement
Our study evaluates a comprehensive cancer screening program Peru, highlighting its effectiveness in early-stage detection. High-quality screening improves prognosis, reduces mortality, and supports less invasive treatments. Findings emphasize the necessity of universal access policies to screening programs, offering critical insights for cancer prevention and public health strategies in diverse settings.
Introduction
Cancer remains a leading cause of morbidity and mortality worldwide, imposing a significant burden on public health systems and economies. It accounts for 22.8% of deaths from non-communicable diseases globally. 1 According to GLOBOCAN 2020, approximately 19.3 million new cancer cases were reported worldwide, with projections estimating to 35 million by 2050.2–4 The lifetime risk of developing cancer is estimated at 1 in 5 individuals, with a mortality rate of 1 in 9 males and 1 in 12 females, respectively. 4 In 2020, approximately 1 million women died due to oncological diseases, with breast and cervical cancers accounting for 50% of these deaths. 5 Among both sexes, lung cancer remains the leading cause of cancer-related mortality worldwide.1,2
In Peru, the main causes of mortality correspond to infectious diseases, cardiovascular diseases, and cancer. 6 Regarding the last one, approximately 70,000 new diagnoses were recorded in 2020, highlighting its growing impact on public health. The most frequently diagnosed cancers among men were prostate (26.6%), stomach (10.2%), and colorectal (6.9%), whereas among women, breast (18.5%), cervical (11.5%), and colorectal (6.4%) cancers were predominant.3,4 Despite this high burden, cancer screening coverage remains suboptimal, leading to late-stage diagnoses and worse clinical outcomes. Peru has approximately 34 million inhabitants, of which 30% reside in Metropolitan Lima. The age distribution shows that about 25% of the population is over 50 years old, 7 and 76.8% of the Metropolitan Lima population has health insurance. 8 Various social and behavioral determinants influence the incidence of cancer in Peru. Among them, tobacco use and infections by human papillomavirus or Helicobacter pylori stand out. 9
In Peru, the Ministry of Health has implemented national programs for the early detection of cervical, breast, prostate, and colorectal cancers, which are offered free of charge in public health facilities.10,11 However, the uptake of these services remains limited. Cervical cancer screening coverage among women aged 30–49 reached 25.1% in 2023, while colorectal cancer screening coverage reached 15.2% in the same year. 11
Early cancer detection through screening programs is essential to improving prognosis and survival rates by enabling timely intervention and curative treatments. 12 Commonly employed screening tools include mammography for breast cancer, prostate-specific antigen (PSA) for prostate cancer, fecal occult blood testing and colonoscopy for colorectal cancer, and Papanicolaou (PAP) and human papillomavirus (HPV) tests for cervical cancer.13,14 However, the effectiveness of these methods varies. For instance, mammography sensitivity ranges from 63% to 87%, 15 PSA testing demonstrates a sensitivity of 20% in stages A1 and 98% in stages D2, 16 while colonoscopy exhibits a sensitivity of approximately 95%. 17 Although these methods enhance early detection rates, their specificity and the occurrence of false-positive and false-negative results impact clinical decision-making and patient outcomes.
Early cancer detection systems are essential to improve prognosis and increase survival rates, allows the use of less invasive and more effective treatments, reducing mortality. 14 Limited access to screening programs in certain regions and inadequate coverage of these programs in vulnerable populations are two factors contributing to this problem. 12 In Peru, these barriers are further exacerbated by regional inequalities in healthcare infrastructure, particularly in underserved populations.
This study aims to evaluate the diagnostic performance and feasibility of a comprehensive cancer screening program in a private referral clinic in Peru. The findings will provide real-world data on early detection rates and inform public health policies regarding cancer prevention.
Several studies have demonstrated that well-structured cancer screening programs significantly reduce cancer-related mortality. For example, a meta-analysis showed that mammography screening reduced breast cancer mortality by 20% among women invited to participate. 18 Similarly, widespread PAP testing has led to an 80% reduction in cervical cancer incidence in countries with established national screening programs. 19 However, while screening programs improve early detection rates, challenges related to test accessibility, accuracy, and follow-up care remain critical concerns.12,20
This study aims to evaluate the diagnostic performance and feasibility of a comprehensive cancer screening program implemented in a private referral clinic in Peru. By analyzing real-world data, this study seeks to provide insights into early cancer detection rates, screening test performance, and clinical outcomes. These findings will contribute to the optimization of screening strategies and inform public health policies aimed at improving cancer prevention and early diagnosis.
Materials and methods
Study design
This was an observational, retrospective, and longitudinal study designed to evaluate the frequency and characteristics of cancer detection in asymptomatic patients undergoing cancer screenings at a private referral clinic in Peru from 2017 to 2019.
Population
Patients were selected based on predefined inclusion and exclusion criteria, without randomization or stratification. All individuals aged ≥18 years with private health insurance who attended the oncology clinic for preventive cancer screening during the study period were included. Although the program did not formally define a minimum age for breast cancer screening, international recommendations were considered during its implementation. Consequently, mammography and breast ultrasound were preferentially offered to women aged 40 years and older.
This consecutive selection approach ensured the study captured real-world clinical data and minimized biases associated with selective sampling. Patients with a history of primary tumors diagnosed before screening were excluded. Each participant underwent annual comprehensive preventive examinations, with an additional 1-year follow-up in cases where clinical, laboratory, or imaging abnormalities were detected.
The oncology clinic, located in Lima, Peru, serves a nationwide patient population, with approximately 75.4% residing in Lima and 24.6% from other regions. 21
A formal sample size analysis was not initially conducted due to the observational nature of the study and the availability of complete electronic records for all eligible. However, a post hoc power analysis was performed to assess the adequacy of the sample size for detecting significant associations between cancer detection rates and screening outcomes. This ensures the robustness of the study conclusions.
Screening program and population selection
The comprehensive cancer screening program was launched in 2014 as part of the institution’s preventive health strategy aimed at the asymptomatic adult population. Patients are invited to participate in the program upon registration with the clinic and during routine preventive visits. There is no target age group beyond the minimum age of 18 years, and a family history of cancer is not required for inclusion.
In Peru, approximately 10% of the population is covered by private health insurance, and about half of them have specific oncological coverage that allows access to this program. Uninsured patients were not included, as access to the program is restricted to insured individuals. Furthermore, an estimated 30% of insured patients do not attend the clinic or participate in the program, reflecting a participation gap.
Screening results and diagnostic follow-up
Patients attending the Cancer Prevention and Control Service underwent medical examinations tailored to their age, sex, and personal preferences:
Women: physical examination, complete blood count, fecal occult blood test (Thevenon test guaiac-based, qualitative method), chest X-ray, gynecological examination, PAP, endoscopy (≥40 years), colonoscopy (≥50 years), breast ultrasound, and mammography.
Men: physical examination, complete blood count, fecal occult blood test (Thevenon test guaiac-based, qualitative method), chest X-ray, urologic examination, endoscopy (≥ 40 years), colonoscopy (≥ 50 years), and PSA test.
The screening program included a standardized set of tests aimed at the early detection of specific cancer types prioritized according to international guidelines. These included: breast cancer (via mammography and ultrasound in women ≥ 40 years), cervical cancer (via gynecological exam and PAP smear), colorectal cancer (via colonoscopy in individuals ≥ 50 years), prostate cancer (via digital rectal exam and PSA test), and gastric cancer (via endoscopy in individuals ≥ 40 years). Endoscopy was performed exclusively for gastric cancer screening in asymptomatic individuals and was not intended to detect hepatobiliary malignancies. The selection of these tests was based on their proven utility in population studies and recommendations issued by organizations such as the U.S. Preventive Services Task Force and American Cancer Society. 22
Besides, the complete blood count—which was the only routine laboratory test included—additional tests such as thyroid, liver, or renal function panels, as well as abdominal, pelvic, or neck ultrasounds, were not part of the standard protocol. These were ordered only in selected cases at the physician’s discretion, based on physical findings or personal/family history of cancer. Due to their selective application, these tests were not consistently documented in electronic health records. Consequently, incidental cases of cancers other than the targeted ones may occasionally be detected and reported within the screening programs themselves.
A positive screening was defined as the presence of any suspicious abnormality in the screening package that triggered clinical referral for diagnostic confirmation. A negative screening referred to the absence of abnormal findings and therefore no need for immediate clinical follow-up.
A post-screening cancer diagnosis was defined as any cancer detected within 4 years after participation in the program, regardless of cancer type or recommended screening frequency. This fixed follow-up window was uniformly applied, given that most patients underwent only a single evaluation during the study period and specific test timing was not systematically recorded. Thus, we did not establish cancer-specific observation windows (e.g. 1–2 years for mammography or 5–10 years for colonoscopy), which we acknowledge as a study limitation.
Patients with positive screening results underwent further evaluation by a medical oncologist within 1 year to confirm the diagnosis. Diagnostic follow-up included confirmatory tests such as biopsies, additional imaging, or molecular testing, depending on the suspected malignancy. The standardized follow-up protocol ensured consistency in evaluating abnormal findings.
To assess the diagnostic performance of the screening program, both sensitivity and specificity were analyzed, along with false-positive and false-negative rates. This comprehensive approach allowed for a more nuanced evaluation of the effectiveness and clinical utility of the screening tools.
Demographic data, medical history, body mass index, risk factors, and screening results were systematically recorded in a structured database. Mortality data were obtained by cross-referencing the clinic’s system with the Peruvian National Mortality Registry, ensuring accuracy in survival outcomes. The study included a 4-year follow-up period to evaluate long-term prognostic implications.
Due to the full implementation of electronic medical records since 2014, no participants were excluded due to incomplete data. Furthermore, the structured follow-up system ensured that all participants with abnormal findings completed their diagnostic evaluations. Only 0.1% of patients were lost to follow-up at 3 years, and 0.22% after 4 years, rates that were considered negligible. The study incorporated a 4-year follow-up period to evaluate long-term outcomes, including survival and recurrence. Mortality data were thoroughly verified using national registries to minimize missing data and improve data reliability.
Sex-based analysis
Sex was considered a biological variable by tailoring screening protocols to address sex-specific cancer risks. Subgroup analyses were performed to compare cancer detection rates and clinical outcomes between men and women. Breast, gynecological, and prostate cancers were analyzed separately, while colorectal and lung cancers were assessed across both sexes.
Statistical analysis
Descriptive analysis was performed for numerical variables, with mean and standard deviation values used to summarize age distribution. Categorical variables were presented as absolute and relative frequencies (percentages).
Inferential statistical tests were used to analyze the relationships between variables. Student’s t value was used to compare means between age groups with the cancer diagnosis variable to check the assumptions of normality and homogeneity of variances. For the categorical variables, the Chi-squared independence test was applied to evaluate the independence of the variables with cancer diagnosis and to verify the existence of significant associations, preview evaluation supposes (expected frequencies equal or greater than 5).
To improve the internal validity of our findings, a secondary analysis was conducted, restricted to cancers considered primary targets of the institutional screening program. These included breast, cervical, colorectal, prostate, and gastric cancers—all of which had validated screening methods included in the standard test package. In this subanalysis, sensitivity, specificity, accuracy, and likelihood ratios were recalculated. Additionally, a stratified analysis by sex was performed to explore differential test performance.
All statistical analyses were performed in STATA® v. 16.
Ethical considerations
This study was approved by the Ethics Committee of Clinica Aliada, Peru in January 2024. Patient confidentiality and data protection were ensured through an anonymized coding system. Data collection and analysis adhered to the ethical principles outlined in the Declaration of Helsinki, safeguarding patient privacy and integrity. All participants voluntarily attended screenings and provided informed consent for their data use in research. No identifiable information was recorded, and strict measures were implemented to preserve anonymity.
Results
Population characteristics
A total of 31,057 individuals participated in the screening program, with a predominance of women (69.2%) and a mean age of 49.2 years (standard deviation (SD) = 15.4). The distribution of screenings was consistent throughout the 3 years of the study: 33.8% in 2017, 33.1% in 2018, and 33.2% in 2019.
Regarding family history, 19.2% reported a history of breast cancer, followed by gastric cancer (11.6%), prostate cancer (8.1%), and colon cancer (6.8%). Other reported histories included uterine cancer (7.6%), lung cancer (8.3%), and melanoma (2.1%). Among female participants, 67% were premenopausal, while 33% were postmenopausal. Additionally, 64.9% had no history of abortions, whereas 35.1% had experienced at least one.
In terms of lifestyle factors, 12.5% reported alcohol consumption, and 2.6% were current smokers. Among comorbidities, 9.5% had hypertension, and 3% had diabetes mellitus. Further demographic and clinical details are summarized in Table 1.
Table 1.
Characteristics of the patients attending the comprehensive cancer screening program in Lima, Peru, from 2017 to 2019.
| Variables | Cancer diagnosed | |||
|---|---|---|---|---|
| No | Yes | Total | p value α | |
| N | 30,836 (99.3%) | 221 (0.7%) | 31,057 (100.0%) | |
| Age* | 49.14 (15.3) | 59.87 (11.8) | 49.21 (15.3) | <0.001 ± |
| 18–40 | 9099 (99.9%) | 13 (0.1%) | 9112 (29.3%) | <0.001 |
| 41–60 | 13,654 (99.3%) | 95 (0.7%) | 13,749 (44.3%) | |
| 61–95 | 8083 (98.6%) | 113 (1.4%) | 8196 (26.4%) | |
| Year | ||||
| 2017 | 10,417 (33.8%) | 72 (32.6%) | 10,489 (33.8%) | 0.078 |
| 2018 | 10,211 (33.1%) | 61 (27.6%) | 10,272 (33.1%) | |
| 2019 | 10,208 (33.1%) | 88 (39.8%) | 10,296 (33.2%) | |
| Sex | ||||
| Male | 9496 (99.4%) | 59 (0.6%) | 9555 (30.8%) | 0.188 |
| Female | 21,340 (99.3%) | 162 (0.7%) | 21,502 (69.2%) | |
| Family history | ||||
| Gastric cancer | ||||
| No | 27,269 (99.3%) | 191 (0.7%) | 27,460 (88.4%) | 0.353 |
| Yes | 3567 (99.2%) | 30 (0.8%) | 3597 (11.6%) | |
| Colon cancer | ||||
| No | 28,728 (99.3%) | 205 (0.7%) | 28,933 (93.2%) | 0.813 |
| Yes | 2108 (99.2%) | 16 (0.8%) | 2124 (6.8%) | |
| Prostate cancer | ||||
| No | 28,329 (99.3%) | 201 (0.7%) | 28,530 (91.9%) | 0.618 |
| Yes | 2507 (99.2%) | 20 (0.8%) | 2527 (8.1%) | |
| Ovarian cancer | ||||
| No | 30,014 (99.3%) | 214 (0.7%) | 30,228 (97.3%) | 0.645 |
| Yes | 822 (99.2%) | 7 (0.8%) | 829 (2.7%) | |
| Breast cancer | ||||
| No | 24,920 (99.3%) | 164 (0.7%) | 25,084 (80.8%) | 0.013 |
| Yes | 5916 (99.0%) | 57 (1.0%) | 5973 (19.2%) | |
| Cervical cancer (cervix or endometrium) | ||||
| No | 28,499 (99.3%) | 201 (0.7%) | 28,700 (92.4%) | 0.411 |
| Yes | 2337 (99.2%) | 20 (0.8%) | 2357 (7.6%) | |
| Lung cancer | ||||
| No | 28,282 (99.3%) | 205 (0.7%) | 28,487 (91.7%) | 0.575 |
| Yes | 2554 (99.2%) | 16 (0.8%) | 2570 (8.3%) | |
| Lymphoma | ||||
| No | 29,605 (99.3%) | 211 (0.7%) | 29,816 (96.0%) | 0.687 |
| Yes | 1231 (99.2%) | 10 (0.8%) | 1241 (4.0%) | |
| Melanoma | ||||
| No | 30,185 (99.3%) | 211 (0.7%) | 30,396 (97.9%) | 0.013 |
| Yes | 651 (98.5%) | 10 (1.5%) | 661 (2.1%) | |
| Multiple myeloma | ||||
| No | 30,606 (100.0%) | 221 (0.0%) | 30,827 (99.3%) | 0.198 |
| Yes | 230 (100.0%) | 0 (0.0%) | 230 (0.7%) | |
| Liver cancer | ||||
| No | 30,355 (99.3%) | 215 (0.7%) | 30,570 (98.4%) | 0.168 |
| Yes | 481 (99.2%) | 6 (0.8%) | 487 (1.6%) | |
| Medical history in women | ||||
| First menstrual period (years) | 12.470 (1.617) | 12.398 (1.828) | 12.470 (1.619) | 0.571 |
| Menopause | ||||
| No | 14,346 (99.4%) | 83 (0.6%) | 14,429 (67.0%) | <0.001 |
| Yes | 7012 (98.9%) | 79 (1.1%) | 7091 (33.0%) | |
| Prior abortions | ||||
| No | 13,873 (99.3%) | 93 (0.7%) | 13,966 (64.9%) | 0.045 |
| Yes | 7485 (99.1%) | 69 (0.9%) | 7554 (35.1%) | |
| Breastfeeding | ||||
| No | 4161 (99.2%) | 34 (0.8%) | 4195 (19.5%) | 0.002 |
| Not applicable | 6999 (99.2%) | 32 (0.8%) | 7031 (32.7%) | |
| Yes | 10,198 (99.1%) | 96 (0.9%) | 10,294 (47.8%) | |
| Hormone replacement therapy | ||||
| No | 17,759 (99.4%) | 129 (0.6%) | 17,888 (83.1%) | 0.233 |
| Yes | 3599 (99.2%) | 33 (0.8%) | 3632 (16.9%) | |
| Drinks alcohol | ||||
| No | 26,993 (99.3%) | 196 (0.7%) | 27,189 (87.5%) | 0.606 |
| Yes | 3843 (99.4%) | 25 (0.6%) | 3868 (12.5%) | |
| Smokes | ||||
| No | 30,031 (99.3%) | 214 (0.7%) | 30,245 (97.4%) | 0.605 |
| Yes | 805 (99.1%) | 7 (0.9%) | 812 (2.6%) | |
| High blood pressure | ||||
| No | 27,939 (99.4%) | 165 (0.6%) | 28,104 (90.5%) | <0.001 |
| Yes | 2897 (98.1%) | 56 (1.9%) | 2953 (9.5%) | |
| Diabetes mellitus | ||||
| No | 29,913 (99.3%) | 211 (0.7%) | 30,124 (97.0%) | 0.184 |
| Yes | 923 (98.9%) | 10 (1.1%) | 933 (3.0%) | |
| Body mass index | ||||
| Underweight | 2025 (99.9%) | 3 (0.1%) | 2028 (6.5%) | 0.007 |
| Normal weight | 11,215 (99.3%) | 82 (0.7%) | 11,297 (36.4%) | |
| Overweight | 12,109 (99.3%) | 86 (0.7%) | 12,195 (39.3%) | |
| Obesity | 5442 (99.1%) | 50 (0.9%) | 5492 (17.7%) | |
| Final screening condition | ||||
| Negative result | 27,883 (99.8%) | 43 (0.2%) | 27,926 (89.9%) | <0.001 |
| Positive result | 2953 (94.3%) | 178 (5.7%) | 3131 (10.1%) | |
| Died (4-year follow-up) | ||||
| No | 30,681 (99.4%) | 195 (0.6%) | 30,876 (99.4%) | <0.001 |
| Yes | 155 (85.6%) | 26 (14.4%) | 181 (0.6%) | |
Values expressed as mean and standard deviation.
p value of categorical tests using Chi-squared test.
p value obtained by Student’s t test.
Bold text indicates p-values <0.005.
Patient comparison by cancer diagnosis
Out of 31,057 participants, 221 (0.7%) were diagnosed with cancer without a prior diagnosis (Figure 1). The diagnosed patients had a significantly higher mean age (59.9 years, SD = 11.9) compared to those without a diagnosis (49.1 years, SD = 15.4) (p < 0.001).
Figure 1.
Flowchart of the selection of patients undergoing comprehensive cancer screening in Lima, Peru, from 2017 to 2019.
A family history of breast cancer and melanoma was significantly associated with cancer diagnosis. Individuals with a history of breast cancer had a higher prevalence (1.0% vs 0.7%, p = 0.013), with an even higher prevalence among those with melanoma (1.5% vs 0.07%, p = 0.013).
Among female participants, a history of menopause was significantly associated with cancer diagnosis (1.1% vs 0.6%, p < 0.001). Similarly, women with a history of abortion had a slightly higher cancer prevalence (0.9% vs 0.7%, p = 0.045). Regarding body mass index (BMI), individuals with obesity had a higher cancer prevalence (0.9%) compared to those with normal weight (0.7%) or overweight (0.7%; p = 0.007).
In total, 5.7% of patients with a positive screening result were diagnosed with cancer, compared to 0.2% of those that had a negative result (p < 0.001). Moreover, the 4-year mortality rate was significantly higher among cancer patients (14.4%) than in those without a diagnosis (0.6%) (p < 0.001).
The screening program successfully detected 80.5% of diagnosed cancer cases, demonstrating a sensitivity of 80.54% and specificity of 90.42%. The program achieved a correct classification rate of 90.35%, with a positive likelihood ratio of 8.41 and a negative likelihood ratio of 0.22.
To better reflect the screening program’s actual effectiveness, a secondary analysis was conducted including only target cancers—those for which validated screening methods were systematically implemented: breast, cervical, colorectal, prostate, and gastric cancers. In this subgroup, sensitivity was 82.2%, specificity 90.3%, and overall accuracy 90.3%. When stratified by sex, sensitivity and specificity were 82.4% and 88.6% in women, and 81.5% and 94.3% in men, respectively.
Cancer patients by screening classification
Among the 221 diagnosed cases, 80.5% were classified as having a positive screening result, while 19.5% had a negative screening result. Patients in the positive screening group were significantly younger (59.1 years, SD = 11.7) compared to those in the negative screening group (63.3 years, SD = 11.9; p = 0.036). The patients with a positive screening result were predominantly female (77.5%), whereas females constituted only 55.8% of the negative group (p = 0.004).
The time to diagnosis differed significantly between groups: 61.7% of positive-screening patients were diagnosed within the first 3 months, compared to only 7.0% of negative-screening patients (p < 0.001). Imaging abnormalities were the most common reason for detection (42.3%), followed by physical examination findings (22.1%) and a combination of both (16.3%).
Most cases were breast cancer, with a significantly higher proportion among patients with positive screening results (50.0%) compared to those with negative screening results (2.3%). Prostate cancer was the second most common malignancy, with a similar distribution (12.6% in the positive screening group vs 2.3% in the negative screening group). The most frequent cancers among patients with negative screening results were melanoma (18.6%), thyroid cancer (13.95%), and kidney cancer. Diagnosed cases were predominantly detected at early clinical stages (stages 0 and 1; 58.4%), with no significant difference between groups (p = 0.570). Breast cancer remained the most frequently diagnosed malignancy overall, accounting for 40.7% of cases, followed by thyroid (9.3%) and ovarian cancer (9.3%; Figure 2).
Figure 2.
Type of cancer diagnosed according to the results of the comprehensive cancer screening program in Lima, Peru, from 2017 to 2019.
Out of 142 cases with available clinical staging data, 83 (58.4%) were diagnosed at early stages (stage 0 or 1), predominantly involving breast, cervical, and prostate cancers — all of which were included in the core screening package. In contrast, 21 patients (14.8%) were diagnosed at advanced clinical stages (stage 3 or 4), corresponding to diverse cancer types including breast, ovarian, prostate, colorectal, biliary tract, thyroid, kidney, melanoma, multiple myeloma, and brain tumors. These advanced-stage diagnoses often involved tumor types that are less amenable to early detection or not routinely screened in the program. The median survival time for this subgroup was 72.3 months (IQR: 49.2).
Although there was no statistically significant difference in overall mortality between the groups, the survival rate was higher for patients with positive screening results (89.3%) than for those with negative screening results (83.7%; p = 0.306). The distribution of cancer types also revealed sex-specific differences. Among males, prostate cancer was the most prevalent (39.0%), followed by melanoma (15.3%) and thyroid cancer (3.4%). Conversely, among females, breast cancer was the most common diagnosis (54.9%), followed by cervical cancer (6.2%), thyroid cancer (9.9%), and melanoma (6.2%). Certain malignancies, such as ovarian, endometrial, and uterine cancers, were exclusive to females, whereas testicular cancer was diagnosed only in males (Figure 3; Table 2).
Figure 3.
Type of cancer diagnosed by sex at cancer screening program in Lima, Peru, from 2017 to 2019.
Table 2.
Characteristics of cancer patients by result of the comprehensive cancer screening program in Lima, Peru, from 2017 to 2019.
| Variables | Screening result | |||
|---|---|---|---|---|
| Negative (n = 43) | Positive (n = 178) | Total | Test α | |
| Age* | 63.27 (11.9) | 59.05 (11.7) | 59.88 (11.8) | 0.036 ± |
| 18–40 | 2 (4.7%) | 11 (6.2%) | 13 (5.8%) | 0.234 |
| 41–60 | 14 (32.6%) | 81 (45.5%) | 95 (43.0%) | |
| 61–95 | 27 (62.8%) | 86 (48.2%) | 113 (51.2%) | |
| Sex | ||||
| Male | 19 (44.2%) | 40 (22.5%) | 59 (26.7%) | 0.004 |
| Female | 24 (55.8%) | 138 (77.5%) | 162 (73.3%) | |
| Drinks alcohol | ||||
| No | 38 (88.4%) | 158 (88.8%) | 196 (88.7%) | 0.942 |
| Yes | 5 (11.6%) | 20 (11.2%) | 25 (11.3%) | |
| Died (4-year follow-up) | ||||
| No | 36 (83.7%) | 159 (89.3%) | 195 (88.2%) | 0.306 |
| Yes | 7 (16.3%) | 19 (10.7%) | 26 (11.8%) | |
| Time to diagnosis | ||||
| 0–3 months | 3 (7.0%) | 108 (61.7%) | 111 (50.9%) | <0.001 |
| 3–6 months | 17 (39.5%) | 29 (16.6%) | 46 (21.1%) | |
| 6–9 months | 13 (30.2%) | 20 (11.4%) | 33 (15.1%) | |
| 9–12 months | 10 (23.3%) | 18 (10.3%) | 28 (12.8%) | |
| Reason for detection | ||||
| Cytology | 5 (4.8%) | 5 (4.8%) | . | |
| Physical examination | 23 (22.1%) | 23 (22.1%) | ||
| Physical examination, imaging | 17 (16.3%) | 17 (16.3%) | ||
| Physical examination, laboratory | 11 (10.6%) | 11 (10.6%) | ||
| Imaging | 44 (42.3%) | 44 (42.3%) | ||
| Imaging, laboratory test | 1 (1.0%) | 1 (1.0%) | ||
| Laboratory test | 3 (2.9%) | 3 (2.9%) | ||
| Clinical stage | ||||
| 0 | 2 (8.0%) | 16 (13.7%) | 18 (12.7%) | 0.570 |
| 1 | 12 (48.0%) | 53 (45.3%) | 65 (45.7%) | |
| 2 | 5 (20.0%) | 33 (28.2%) | 38 (26.8%) | |
| 3 | 4 (16.0%) | 9 (7.7%) | 13 (9.2%) | |
| 4 | 2 (8.0%) | 6 (5.1%) | 8 (5.6%) | |
Values expressed as mean and standard deviation.
p value obtained by Student’s t test.
p value of categorical tests using Chi-squared test.
Discussion
The results of this large-scale screening study underscore the utility of systematic cancer screening in an asymptomatic population. Our study determined that the annual prevalence of cancer diagnosis among apparently healthy patients was 0.7%. This means that for every 10,000 people screened in a year, 70 will be diagnosed with cancer. Of these 70 cases, it is estimated that 56 will be detected through observation in comprehensive screening, while 14 would remain undetected initially but would be diagnosed at a later stage.
We acknowledge that our primary analysis included all cancer diagnoses recorded during follow-up, without distinguishing between target cancers of the screening program and those not subject to standardized screening protocols. The inclusion of non-target cancers may have affected the performance estimates of the screening strategy. To address this limitation, we performed an additional analysis restricted to cancers with validated screening methods included in the institutional protocol (breast, cervical, colorectal, prostate, and gastric). This subgroup analysis yielded improved diagnostic performance metrics and enhanced the internal validity of our findings.
Time to diagnosis and reasons for detection
The time to diagnosis in screening programs is a crucial indicator of their effectiveness in early cancer detection. In our study, 61.7% of patients with positive screening results were diagnosed within the first 3 months, compared to only 7.0% of those with negative screening results. Our findings on the faster diagnosis in cases with positive screening coincide with what was reported before, 23 where high adherence to the follow-up of abnormal outcomes in breast, cervical and colorectal cancer was documented. This type of population metrics underlines the importance of ensuring structured follow-up processes to translate the findings into early and effective diagnoses.
To minimize the impact of lead-time and reverse-time bias, we opted not to perform formal comparisons of time-to-diagnosis between the two groups. Instead, we present descriptive statistics exclusively for the positive screening group, in which most cases were promptly identified and referred. This approach more accurately reflects the program’s operational success in ensuring early linkage to specialized care following an abnormal result.
These findings emphasize the importance of systematic medical evaluations for early case identification, aligning with previous studies on the impact of risk-stratified follow-up programs. 22 Late diagnoses (6–12 months post-screening) were significantly more frequent in the negative screening result group, likely due to the detection of less common and infrequently screened cancers such as kidney cancer, lymphoma, biliary tract cancer, melanoma, and brain tumors. These cancers often present later due to lower incidence rates and the absence of targeted screening programs.
Previous research 24 has demonstrated that the time to diagnosis varies based on cancer type, healthcare access, and socioeconomic factors. For instance, breast and cervical cancers are often diagnosed early due to established screening protocols, whereas melanoma or gastrointestinal cancers may take longer to detect, particularly in asymptomatic cases.
The most common reason for cancer detection in this study was imaging abnormalities (42.3%), followed by findings from physical examinations (22.1%) and a combination of both (16.3%). These results are consistent with existing evidence supporting the role of imaging techniques such as mammography and computed tomography in early neoplasm detection. 25 Cytologic (4.8%) and laboratory findings (2.9%) played a smaller role, likely due to their lower specificity in population screening. Imaging techniques were particularly influential in early diagnoses within the positive-screening group, reinforcing the need to optimize access to high-quality imaging tests in screening programs to improve early detection rates and health outcomes.
Types of cancer diagnosed
Breast cancer was the most frequently diagnosed malignancy (40.7%), with a significant predominance in the positive-screening group compared to the negative-screening group (50.0% vs 2.3%). This underscores the efficacy of routine mammographic screening, which has been shown to reduce breast cancer mortality in longitudinal studies.25,26
Although the program did not establish a formal minimum age threshold for breast cancer screening, most diagnosed women were over 40 years of age. This suggests that the program’s practical implementation was aligned with international guidelines, which recommend initiating mammography in women aged ≥ 40 years with average risk.
Other frequently observed cancers included melanoma (8.6%), thyroid cancer (8.1%), and prostate cancer (10.4%). Melanoma was more prevalent in the negative-screening group (18.6%) compared to the positive-screening group (6.2%). This difference may be attributed to the clinical features of melanoma, which often allow visual diagnosis without intensive observation. The higher incidence in the negative group may also be associated with insufficient preventive measures.
Thyroid cancer was diagnosed in both positive-screening (6.7%) and negative-screening (14.0%) groups, suggesting that its early-stage presentation can be subtle, making detection challenging without specialized tools like ultrasonography, which was not included in the comprehensive screening program. 27
Prostate cancer represented 12.4% of the cases in the positive-screening group, aligning with the evidence supporting PSA testing as an effective tool for prostate cancer screening and early diagnosis. However, its use remains controversial, as studies have shown that PSA has not had a significant impact on reducing cancer-specific mortality.28,29
In contrast, our findings showed that melanoma and thyroid cancer ranked among the most frequently diagnosed cancers in this cohort, while gastric cancer was almost absent. These differences can be explained by the nature of our study population—individuals with private health insurance undergoing routine screening while asymptomatic.
Screening programs, particularly in insured and health-literate populations, tend to detect indolent, early-stage tumors with favorable prognoses, such as thyroid cancer and melanoma. These cancers often go unnoticed in general clinical settings until symptoms arise. Furthermore, individuals with access to dermatologic and endocrine care are more likely to undergo evaluations that detect skin lesions and thyroid nodules incidentally. This pattern of detection has been described in several studies and is supported by findings from a population-based cancer registry in Georgia. In that study, over 83% of thyroid cancers in both men and women were diagnosed at stage I, with more than 40% of those cases being papillary microcarcinomas—tumors of low clinical aggressiveness. Notably, thyroid cancer alone accounted for more than 50% of all stage I cancers diagnosed in women. 30
Of note is the observed association between a history of abortion and cancer diagnosis, which—although statistically significant (0.9% vs 0.7%)—was of small magnitude and derived from a bivariate analysis unadjusted for potential confounders such as age or hormonal history. International evidence does not support a causal relationship between abortion and cancer. 31 Accordingly, we recommend further multivariate research to assess any potential association with greater methodological rigor.
These observations highlight how screening programs may lead to disproportionate detection of low-risk cancers, raising the issue of potential overdiagnosis, particularly for thyroid cancer. On the other hand, cancers such as gastric carcinoma—which are commonly associated with late-stage presentation and nonspecific symptoms—often require symptom-driven diagnostic approaches and are less likely to be identified in asymptomatic screening cohorts.
Our results align with international evidence showing that cancer types identified through screening programs differ significantly from those seen in symptomatic, hospital-based populations. This divergence is influenced by healthcare access, diagnostic infrastructure, and socioeconomic status, which are known to affect both the likelihood of screening uptake and the types of cancers diagnosed. 32
Several cancers diagnosed in this study, such as lymphomas, kidney, biliary tract, and brain tumors, currently lack effective screening methods for early detection. Its diagnosis is usually incidental or symptomatic, which explains its presentation in more advanced stages and the lower capacity of population screening programs to impact its prognosis. This finding underlines the need to differentiate between cancers with consolidated detection strategies (breast, cervix, prostate, colon) and those that, for now, are not susceptible to early diagnosis by screening.
Clinical stages of cancer type
Most diagnosed cases were at early clinical stages (0 and 1, 58.4%). However, it is important to clarify that this figure includes all cancer types diagnosed during the 4-year follow-up, not only those considered as target cancers of the screening program. When focusing on cancers with consolidated screening strategies—such as breast, cervical, and prostate cancer—the proportion of early-stage diagnoses was even higher, reflecting the effectiveness of structured early detection protocols.
Among patients with positive screening results, 13.7% were diagnosed at stage 0 and 45.3% at stage 1. In contrast, negative-screening patients had a higher proportion of advanced-stage diagnoses, including 16.0% at stage 3. However, the comparison of clinical stages at diagnosis between screening patients did not yield statistically significant differences (p = 0.57). Therefore, no conclusive difference in stage distribution can be affirmed between groups.
Despite its high incidence, melanoma was predominantly diagnosed at early stages within the negative-screening group, highlighting the importance of visual assessments and prompt biopsies. However, less frequent cancers such as lymphomas or kidney cancer showed a more uniform distribution across clinical stages, suggesting potential limitations in the sensitivity of screening for these malignancies. 33
The distribution of cancer types by consultation reason suggests that breast cancer predominated in the positive-screening group due to well-established mammographic surveillance programs. In contrast, melanoma and thyroid cancer were more common in the negative-screening group, indicating lower utilization or availability of early detection strategies for these cancers.
The overall analysis of clinical stages confirms that most diagnoses occurred at early stages (58.4% at stages 0 and 1), which is associated with better prognoses. However, a significant proportion of advanced-stage cases (14.8% at stages 3 and 4) underscores the need to enhance screening coverage and effectiveness. Similarly, in the registration of lung screening with low-dose tomography, Silvestri et al. 34 reported a substantial change toward stage I diagnoses (53.5% of cases), which confirms that screening programs can significantly modify the distribution of stages at the time of diagnosis.
The data are consistent with the findings of Allemani et al., 35 who reported that early diagnosis is more common in countries with health systems prioritizing access to diagnostic imaging and screening programs. For example, in breast cancer, 70%–80% of cases in countries with regular screening programs are diagnosed at early stages, as observed in this analysis for the group with positive screening results. In cancers such as melanoma, the clinical stages at diagnosis often depend on patients’ initial perception and the speed at which they seek medical attention, highlighting the importance of active surveillance programs to achieve timely diagnosis.
From a public health perspective, this study highlights the necessity of policies promoting universal access to screening programs. High-income countries have achieved substantial advancements in early breast and cervical cancer detection, reducing mortality through increased public health investment, funding for early detection initiatives, well-defined screening guidelines, and national cancer control plans. 36 Estimates suggest that mammographic screening alone contributes to a 15%–30% reduction in breast cancer mortality. 37
However, universal screening for all cancer types across the entire adult population is neither sustainable nor cost-effective. Our findings support the prioritization of structured screening programs targeting cancers with higher detection frequencies and validated diagnostic methods. For example, breast cancer in women aged ≥ 40 years, cervical cancer in women aged 25–65 years, prostate cancer in men aged ≥ 50 years, and colorectal cancer in adults aged 50–75 years. In contrast, we do not recommend the routine use of upper gastrointestinal endoscopy in asymptomatic individuals, nor population-based screening for thyroid cancer or melanoma, due to their low diagnostic yield and the lack of validated screening strategies in our setting. These recommendations aim to optimize the program’s impact based on principles of equity, cost-effectiveness, and long-term sustainability within the framework of a national cancer control policy.
The clinical translation of these findings is reflected in survival rates. In Peru, according to national data for 2018, 66,627 new cases of cancer and 33,098 deaths were reported, which corresponds to an overall survival close to 50.3%. 4 In contrast, in our study, survival was 88.2%. This difference suggests that systematic screening programs not only favor diagnosis in early stages, but are also associated with a tangible improvement in survival.
Implementation of screening programs also reflects the ethical obligation that arises after the detection of a disease. Identifying cancer in an asymptomatic person implies the responsibility of ensuring timely access to confirmatory diagnosis, comprehensive treatment and follow-up, otherwise the potential benefit of screening is lost. Likewise, the costs and benefits of screening programs must be weighed, both at the individual and population levels. Although screening can reduce mortality in certain cancers, it also carries risks of overdiagnosis, false positives and unnecessary procedures. Therefore, the design of these strategies must be based on a rigorous analysis of cost-effectiveness and risk-benefit, ensuring that the resources invested are translated into real and sustainable clinical benefits for the population.
This study presents several limitations that must be considered when interpreting its findings. First, its retrospective observational design precludes establishing causal relationships. Second, there was heterogeneous adherence to the screening package, as not all participants completed the full set of recommended tests. Third, it was not possible to reconstruct the diagnostic process in detail for patients with negative screening results, limiting the interpretation of false negatives. Fourth, the database lacked information on treatment and clinical outcomes, preventing an assessment of therapeutic effectiveness or survival. Fifth, the fixed 4-year follow-up period may have been insufficient to capture all incident cancers, especially slow-growing tumors.
Additional methodological considerations include the absence of individual-level data on test completion, preventing linkage of cancer diagnoses to specific screening components. This limitation restricted test-specific performance evaluation. Furthermore, the fixed follow-up window did not align with the recommended periodicity for different screening modalities (e.g. annual mammography vs decennial colonoscopy), potentially affecting the classification of false negatives. Additionally, the low frequency of gastric cancer may be attributed to the limited diagnostic yield of endoscopy in asymptomatic individuals, the lack of a national screening program in Peru, and the absence of routine Helicobacter pylori testing in preventive visits. Despite these constraints, restricting the performance analysis to target cancers with validated screening methods confirmed the robustness of the program, with sensitivity remaining high (> 80%) and improved specificity among men.
Finally, the potential selection bias inherent in the scope of this study must be considered, since it was developed in a private clinic and among people with private health insurance. As mentioned before, 38 the effectiveness observed in some designs is not always generalized to clinical environments due to selection bias and suboptimal adherence to follow-up recommendations. This population is characterized by a higher socioeconomic level, better access to specialized medical services and greater health literacy, which may have influenced detection rates and the type of cancers diagnosed (e.g. higher proportion of thyroid cancer and melanoma in early stages). Therefore, the results are not directly extrapolable to the general population of Peru or to the users of the public health system.
In future implementations, we recommend systematic recording of individual test performance and adherence, adoption of personalized follow-up intervals, and exclusion of cancer types lacking validated screening strategies to enhance analytical precision and clinical applicability. The findings of this study emphasize that timely diagnosis and detection methods are critical in early cancer identification. Imaging-based screening was particularly effective for detecting prevalent cancer types such as breast cancer, whereas complementary methods, including cytology and laboratory testing, played a more limited role. The association between early-stage diagnoses and the most frequently detected cancers among positive-screening patients underscores the importance of risk-stratified screening strategies tailored to individual risk profiles.
These results reinforce the need to personalize screening approaches based on cancer type and patient-specific risk factors to optimize early detection and improve clinical outcomes.
Acknowledgments
We would like to thank ALIADA Preventive Services for their adequate attention to all patients.
Footnotes
ORCID iD: Virgilio E. Failoc-Rojas 
https://orcid.org/0000-0003-2992-9342
Author contributions: ALGL: Conceptualization, data curation, supervision, writing-original draft, writing–review, and editing. REAC: Investigation, software, methodology, writing-original draft, writing–review and editing. CC: Conceptualization, data curation, supervision, writing-original draft, writing–review and editing and VEFR: Conceptualization, data curation, formal analysis, supervision, writing-original draft, writing–review and editing.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement: The data will be made available at reasonable request to the corresponding author.
References
- 1. Cao Y. Lack of basic rationale in epithelial-mesenchymal transition and its related concepts. Cell Biosci 2024; 14(1): 104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Cancer Today [Internet], https://gco.iarc.who.int/today/ (2025, accessed 22 August 2025).
- 3. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024; 74(3): 229–263. [DOI] [PubMed] [Google Scholar]
- 4. Ministerio DS. Situación del cáncer en el Perú, 2021. [Internet], https://www.dge.gob.pe/portal/docs/tools/teleconferencia/2021/SE252021/03.pdf (2021, accessed 22 August 2025).
- 5. Guida F, Kidman R, Ferlay J, et al. Global and regional estimates of orphans attributed to maternal cancer mortality in 2020. Nat Med 2022; 28(12): 2563–2572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ministry of Health of Peru. Identifican principales causas de mortalidad en el Perú [Internet]. https://www.gob.pe/institucion/minsa/noticias/32055-identifican-principales-causas-de-mortalidad-en-el-peru (2014, accessed 19 August 2025).
- 7. del Pacífico U. Generación Plateada: Informe sobre los adultos 50+ y el emprendimiento en Perú [Internet]. Emprende UP. 2022 https://emprendeup.pe/generacion-plateada-informe-sobre-los-adultos-50-y-el-emprendimiento-en-peru/ (2022, accessed 19 August 2025). [Google Scholar]
- 8. Instituto Nacional de Estadistica e Informatica. Población de Lima Metropolitana supera los 10 millones 151 mil habitantes [Internet]. https://m.inei.gob.pe/prensa/noticias/poblacion-de-lima-metropolitana-supera-los-10-millones-151-mil-habitantes-14160/ (2021, accessed 19 August 2025).
- 9. De La Cruz-Vargas JA, Ramos W, Chanduví W, et al. Proportion of cancer cases and deaths attributable to potentially modifiable risk factors in Peru. BMC Cancer 2024; 24(1): 477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ministry of Health of Peru. Minsa continúa realizando acciones a nivel nacional para la detección temprana del cáncer de próstata [Internet]. https://www.gob.pe/institucion/minsa/noticias/980265-minsa-continua-realizando-acciones-a-nivel-nacional-para-la-deteccion-temprana-del-cancer-de-prostata (2024, accessed 19 August 2025).
- 11. Ministry of Health of Peru. Programa Presupuestal 0024: Prevención y Control del Cáncer [Internet], https://www.minsa.gob.pe/presupuestales/doc2024/reporte-seguimiento/Reporte%202023-I_PP%200024.pdf (2023, accessed 19 August 2025).
- 12. Schiffman JD, Fisher PG, Gibbs P. Early detection of cancer: past, present, and future. Am Soc Clin Oncol Educ Book Am Soc Clin Oncol Annu Meet 2015; 35(1): 57–65. [DOI] [PubMed] [Google Scholar]
- 13. ¿Qué exámenes se usan para detectar el cáncer? [Internet]. https://www.cancer.gov/espanol/cancer/deteccion/examenes-de-deteccion (2015, 22 August 2025).
- 14. Bartolomé-Moreno C, Melús-Palazón E, Vela-Vallespín C, et al. Recomendaciones de prevención del cáncer. actualización 2024. Aten Primaria 2024; 56: 103128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Abugattas Saba J, Manrique Hinojosa J, Vidaurre Rojas T. Mamografía como instrumento de tami zaje en cáncer de mama. Rev Peru Ginecol Obstet 2015; 61(3): 311–319. [Google Scholar]
- 16. Molina R. El antígeno prostático específico en el diagnóstico precoz del cáncer de próstata. Med Integral Med Prev Asist En Aten Primaria Salud 2000; 36(6): 199–202. [Google Scholar]
- 17. Camacho-Nájera M, Armienta-Sarabia R, Hernández-Gómez ME, et al. Detección de cáncer colorrectal por colonoscopia: Resultados de 1 año de reportes histopatológicos. Endoscopia 2020; 32: 110–115. [Google Scholar]
- 18. Marmot MG, Altman DG, Cameron DA, et al. The benefits and harms of breast cancer screening: an independent review. Br J Cancer 2013; 108(11): 2205–2240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Puga O, Belmar F, Pertossi E. Prevención y detección precoz del cáncer cervicouterino. Rev Médica Clínica Las Condes 2024; 35(2): 95–105. [Google Scholar]
- 20. Wender R, Wolf AMD. Increasing cancer screening rates in primary care. Med Clin North Am 2020; 104(6): 971–987. [DOI] [PubMed] [Google Scholar]
- 21. Revilla-López J, Anampa-Guzmán A, Marquez LC, et al. Cancer cases detected in the prevention and control service of a private cancer clinic in Peru. Infect Agent Cancer 2019; 14: 44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Smith RA, Andrews KS, Brooks D, et al. Cancer screening in the United States, 2019: a review of current American Cancer Society guidelines and current issues in cancer screening. CA Cancer J Clin 2019; 69(3): 184–210. [DOI] [PubMed] [Google Scholar]
- 23. Barlow WE, Beaber EF, Geller BM, et al. Evaluating screening participation, follow-up, and outcomes for breast, cervical, and colorectal cancer in the PROSPR Consortium. J Natl Cancer Inst 2019; 112(3): 238–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Neal RD, Tharmanathan P, France B, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer 2015; 112 Suppl 1: S92–S107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Gangnon RE, Sprague BL, Stout NK, et al. The contribution of mammography screening to breast cancer incidence trends in the United States: an updated Age–Period–Cohort model. Cancer Epidemiol Biomarkers Prev 2015; 24(6): 905–912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Tabar L, Yen MF, Vitak B, et al. Mammography service screening and mortality in breast cancer patients: 20-year follow-up before and after introduction of screening. Lancet 2003; 361(9367): 1405–1410. [DOI] [PubMed] [Google Scholar]
- 27. Haugen BR, Alexander EK, Bible KC, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association Guidelines Task Force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016; 26(1): 1–133. 2016. Jan. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ilic D, Djulbegovic M, Jung JH, et al. Prostate cancer screening with prostate-specific antigen (PSA) test: a systematic review and meta-analysis. BMJ 2018; 362: k3519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Remmers S, Bangma CH, Godtman RA, et al. Relationship between baseline prostate-specific antigen on cancer detection and prostate cancer death: long-term follow-up from the European randomized study of screening for prostate cancer. Eur Urol 2023; 84(5): 503–509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Batavani T, Kereselidze M, Chikhladze N, et al. Early and late detection of cancer in Georgia: Evidence from a population-based cancer registry, 2018–2019. Cancer Epidemiol 2022; 80: 102216. [DOI] [PubMed] [Google Scholar]
- 31. Tong H, Wu Y, Yan Y, et al. No association between abortion and risk of breast cancer among nulliparous women: Evidence from a meta-analysis. Medicine 2020; 99(19): e20251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Arnold M, Rutherford MJ, Bardot A, et al. Progress in cancer survival, mortality, and incidence in seven high-income countries 1995-2014 (ICBP SURVMARK-2): a population-based study. Lancet Oncol 2019; 20(11): 1493–1505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ferlay J, Colombet M, Soerjomataram I, et al. Cancer statistics for the year 2020: an overview. Int J Cancer 2021; 149(4): 778–789. [DOI] [PubMed] [Google Scholar]
- 34. Silvestri GA, Goldman L, Tanner NT, et al. Outcomes from more than 1 million people screened for lung cancer with low-dose CT imaging. Chest 2023; 164(1): 241–251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Allemani C, Matsuda T, Di Carlo V, et al. Global surveillance of trends in cancer survival 2000-14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet Lond Engl 2018; 391(10125): 1023–1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Duggan C, Trapani D, Ilbawi AM, et al. National health system characteristics, breast cancer stage at diagnosis, and breast cancer mortality: a population-based analysis. Lancet Oncol 2021; 22(11): 1632–1642. [DOI] [PubMed] [Google Scholar]
- 37. Khrouf S, Letaief Ksontini F, Ayadi M, et al. Breast cancer screening: a dividing controversy. Tunis Med 2020; 98(1): 22–34. [PubMed] [Google Scholar]
- 38. Lund JL, Rivera MP, Su IH, et al. Estimating the effects of cancer screening in clinical practice settings: the role of selective uptake and suboptimal adherence along the cancer screening continuum. Cancer Epidemiol Biomarkers Prev 2024; 33(8): 984–988. [DOI] [PMC free article] [PubMed] [Google Scholar]