Cancer Registry Research Articles

Age-Standardized Cancer Incidence Rate and Ethnicity in Iran

Background: Cancer presents a significant global health challenge, with incidence rates influenced by multifaceted factors including ethnicity. Understanding ethnic disparities in cancer occurrence is crucial for identifying vulnerable populations and designing targeted interventions. In Iran, a diverse ethnic landscape offers a unique opportunity to explore these dynamics. Objectives: This study investigates the association between cancer incidence rates and ethnicity, alongside the impact of the Human Development Index (HDI) on cancer incidence. Methods: Utilizing data from Iran's national cancer registration program, we analyzed the age-standardized incidence rates of 12 common cancers across provinces. Poisson regression and spatial analysis techniques were employed to assess the influence of ethnicities and HDI on cancer incidence. Results: The average standardized incidence rate for all cancers was 154.98 per 100,000 people, with notable variations across provinces and ethnic groups. The Fars ethnicity exhibited the highest incidence rate, while the Baloch ethnicity showed the lowest. Significant spatial autocorrelation was observed for colorectal and stomach cancers, with hotspot analysis revealing regional disparities. Interestingly, only HDI emerged as a significant predictor of cancer incidence in regression models. Conclusions: Contrary to individual-level studies, our province-level analysis found no significant association between cancer incidence and ethnicity in Iran. The prominence of HDI underscores the role of socioeconomic factors in cancer occurrence. While some limitations in this study pose challenges, it highlights the need for targeted research at the individual level to better elucidate cancer-ethnicity associations. Such insights are critical for informing tailored interventions and reducing cancer disparities across diverse ethnic populations in Iran.

District-Wise Distribution and Trends of Cancer Incidence in Nepal: A Five-Year Analysis (2016–2020)

Introduction: Cancer is an increasing public health concern in Nepal. This study utilizes secondary data collected from B.P. Koirala Memorial Cancer Hospital (BPKMCH), the national authority responsible for data collection under the National Cancer Registry. Method: A total of 62,492 new cancer cases were diagnosed and treated between January 1, 2016, and December 31, 2020. Key variables such as name, age, sex, address, and cancer site were recorded to determine the distribution of cancer across the country. These variables were verified and categorized based on topographical regions where patients reported. Cancer cases were coded according to the International Classification of Diseases for Oncology (ICD-O-10, Third Edition), and data analysis was performed using SPSS version 29.0. Results: The highest number of cancer cases was reported from Kathmandu (12.33%), followed by Jhapa (4.10%) and Morang (3.74%) districts. Conclusion: The incidence of cancer showed an increasing trend over the study period. To reduce the cancer burden in Nepal, well-organized awareness campaigns, HPV vaccination, and regular screening programs are strongly recommended.

Trends in Surgical Overtreatment of Prostate Cancer

Overtreatment of prostate cancer is a public health concern that undermines prostate cancer screening efforts. To assess trends in pathologic grade on prostatectomy during the past 2 decades as a surrogate for overtreatment. This retrospective cohort study examined the grade of prostate cancer on final pathology reports among patients undergoing prostatectomy between January 1, 2010, and September 1, 2024, in 2 parallel cohorts: Surveillance, Epidemiology, and End Results (SEER), a nationwide cancer registry, and Michigan Urological Surgery Improvement Collaborative (MUSIC), a statewide clinical registry. The presence of higher-risk features among patients who underwent grade group 1 prostatectomy during this period was also assessed. The primary exposure of interest was year of radical prostatectomy. The primary outcome was the proportion of all prostatectomies that were pathologic grade group 1 (pGG1) on final pathology reports. The secondary outcome was the proportion of pGG1 prostatectomies with a higher-risk preoperative feature, assessed as a binary variable and including at least 1 of the following: more than 50% of biopsy cores positive, prostate-specific antigen of 10 ng/mL or higher, or grade group 2 on biopsy. A total of 162 558 male patients in SEER (median [IQR] age, 63 [57-67] years) and 23 370 in MUSIC (median [IQR] age, 64 [59-69] years) underwent prostatectomy. The proportion of radical prostatectomies resulting in pGG1 on final pathology reports decreased from 32.4% (5852 of 18 071) to 7.8% (978 of 12 500) between 2010 and 2020 in SEER and from 20.7% (83 of 401) to 2.7% (32 of 1192) between 2012 and 2024 in MUSIC. A more recent prostatectomy was associated with a lower likelihood of a pGG1 prostatectomy while controlling for age and race within SEER (odds ratio [OR] per 5 years, 0.41; 95% CI, 0.40-0.42; P < .001) and MUSIC (OR per 5 years, 0.39; 95% CI, 0.36-0.43; P < .001). Within a subset analysis of those prostatectomies that were final pGG1, a more recent prostatectomy was associated with the presence of a higher-risk preoperative feature, including more than 50% of biopsy cores positive, prostate-specific antigen of 10 ng/mL or higher, and grade group 2 on prior biopsy within SEER (OR per 5 years, 1.60; 95% CI, 1.54-1.67; P < .001) and MUSIC (OR per 5 years, 1.60; 95% CI, 1.34-1.90; P < .001). This cohort study found that since 2010, the frequency of pGG1 prostatectomies markedly decreased, and those few that were performed were more likely to have a higher-risk feature. This reduction in the proportion of prostatectomies that are pGG1 likely reflects improved diagnostic pathways, adherence to active surveillance protocols for low-risk cases, and ongoing efforts at both the state and national levels to minimize unnecessary surgical interventions in patients diagnosed with clinically insignificant prostate cancer.

Machine learning models for predicting missing gender in cancer data

Abstract Lung cancer remains a major global health challenge, with varying survival outcomes observed between male and female patients. This study investigates machine learning-based approaches for imputing missing patient sex in SEER (Surveillance, Epidemiology, and End Results) cancer registry data. We employed multiple machine learning models—including Logistic Regression (LR), Naïve Bayes (NB), Random Forest (RF), Multilayer Perceptron (MLP), and Bagging Classifier (BC)—to predict missing sex labels based on clinical features: survival years, surgery status, age, and race. The dataset was split into training (70%) and testing (30%) subsets. Feature importance analysis identified race as the most influential predictor (importance score: 0.978). Among the models evaluated, RF achieved the highest performance, with an accuracy of 97%, AUC of 98%, recall of 95%, precision of 87%, and an F1-score of 91%. This represents a 15.47% increase in accuracy compared to LR (97% vs. 84%), a 99% increase in precision (87% vs. 0%), a 95% improvement in recall (95% vs. 0%), and a 91% increase in F1-score (91% vs. 0%). Compared to NB, RF improved precision by 3.57% (87% vs. 84%) and F1-score by 2.25% (91% vs. 89%). MLP and BC performed almost similarly, with an accuracy of 97%, AUC of 96% and 97%, recall of 94% and 95%, precision of 90% and 87%, and an F1-score of 92% and 91%, showing a 15.47% improvement in accuracy over LR and a 1.03% gain over NB (97% vs. 96%). The findings highlight the clinical relevance of imputing missing sex data, as sex-based disparities can influence cancer treatment outcomes and survival predictions. By ensuring complete and accurate patient records, the proposed approach can enhance personalized treatment planning and epidemiological studies. In addition to accuracy, computational efficiency is an important consideration: MLP’s high accuracy comes at the cost of increased computational complexity, making it less suitable for large-scale real-time applications as compared with LR or NB. These findings underscore the critical role of “Race” in sex prediction and emphasize the potential of the RF model for accurate sex prediction in lung cancer patients. The findings emphasize the potential for enhanced healthcare practices through precise sex prediction in lung cancer patients, addressing a critical gap in medical research.

The role of MLH1, MSH2 and MSH6 in the development of colorectal cancer in Uganda

IntroductionIn Uganda, colorectal cancer (CRC) is steadily increasing according to the Kampala Cancer Registry. In the West, microsatellite instability is detected in 90% of hereditary nonpolyposis colon cancers (HNPCC) which account for 1–2% of all CRC, and 15% of sporadic CRC. Germline mutations in MLH1 and MSH2 account for 90% of HNPCC in the West, whilst the remainder of cases are due to mutations in MSH6 and PMS2. The aim of this study was to determine the microsatellite instability (MSI) status and determine the proportions of MLH1, MSH2, and MSH6 pathological mutations in Ugandan CRC patients.MethodologyThis was a cross-sectional study carried out between 1st January 2008 to 15th September 2021. Patients were recruited prospectively from 16th September 2019 to 16th September 2021, from Masaka Regional Referral Hospital, Mulago National Referral Hospital, Uganda Martyrs’ Hospital Lubaga and Mengo Hospital. From 1st January 2008 to 15th September 2019, CRC FFPE tissue blocks were obtained from the archives of the Department of Pathology, Makerere University. Data was abstracted from the medical case files for demographics, topography and stage. The histopathological subtype and grade of CRC were obtained by two consultant pathologists from the H&E slides. DNA was extracted from CRC formalin-fixed paraffin-embedded (FFPE) tissue blocks. Library preparation was completed using the Qiagen custom design panel. The custom panel represented 56 genes. The MLH-1, MSH2, MSH6, BRAF and KRAS genes were sequenced using the above library preparation and NGS sequencing. The MSI status was obtained if one of the MSI genes, MLH1, MSH2 or MSH6 was pathologically mutated. If none of the genes was pathologically mutated it was considered MSI negative, microsatellite stable (MSS). Immunohistochemistry was carried out to determine whether MLH1 and PMS2 was MMR proficient or deficient. Categorical data was summarized using frequencies and proportions corresponding to each of the three histopathological subtypes and MSI status subtypes. Continuous and categorical variables were analyzed using the chi-square and Fischer’s exact tests. A p -value ≤ 0.05 was considered statistically significant for all the analyses.ResultsOut of 127 CRC patients, the mean(SD) age of MSI cases was 55.6(16.9) years and of MSS cases was 55.4(15.5) years. The majority were MSS, 75(59.06%) followed by MSI, 52(40.9%). There were 14(11.02%) MLH-1 mutations, 30(23.62%) MSH2 mutations, and 26(20.47%) MSH6 mutations. BRAF mutational analysis showed only 5(3.9%) having pathologic missense BRAF V600 mutations. KRAS mutations consisted of only 8(6.3%) having pathologic missense KRAS mutations.ConclusionsThe high rate of MSI in Ugandan colorectal tumours was mainly associated with a lack of BRAF mutations and a high frequency of MSH2 and MSH6 MMR gene mutations. In CRC patients, identification of the causative mutation is recommended, however in a resource-limited setting, MSI testing and immunohistochemistry is more cost effective. In Ugandan CRC patients who meet at least one of the Bethesda criteria, MSI testing and immunohistochemistry may therefore be offered to obtain the MSI status of the tumour.

Trends over time and inter-hospital variation in the primary treatment approach for T1 colon carcinomas in the Netherlands.

BACKGROUND AND STUDY AIMS This study evaluates the use of local resections (LR) as initial treatment versus primary surgery (PS) for T1 colon carcinoma (CC) in the Netherlands over time, hospital variations, and whether changes in treatment approaches impact 5-year relative survival (RS) and overall survival (OS). PATIENTS AND METHODS This nationwide cohort study included all patients diagnosed with pT1 adenocarcinoma of the colon between 2015-2022, identified from the Netherlands Cancer Registry (NCR). Multilevel, multivariable logistic regression models estimated the probability of undergoing LR per hospital, adjusted for case-mix variables. Hospitals were categorized into low, average, or high attitude towards LR. RS and OS were calculated using multivariable regression analysis. RESULTS A total of 9,650 patients from 73 hospitals were included, with 3,999 (41.4%) receiving PS and 5,651 (58.6%) undergoing LR first. From 2015 to 2022, the national proportion of PS decreased from 53.2% to 29.7%. The RRadj for LR varied across hospitals (RRadj 0.46-1.29). No significant differences in RS or OS were found between high- vs. low attitude centers (5-year RS 99.0% vs. 97.7%, RER 0.97, 95%CI 0.51-1.84 and OS 87.9% vs. 86.4%, HRadj 0.95, 95%CI 0.81-1.11), nor between patients treated after vs. before 2018 (5-year RS 98.7% vs. 98.7%, RER 0.82, 95%CI 0.46-1.46 and OS 86.7% vs. 88.0%, HRadj 0.98, 95%CI 0.85-1.13). CONCLUSIONS While inter-hospital variation exists, LR of T1 CC is increasingly preferred in the Netherlands, leading to a reduction in the number of surgeries without a change in RS or OS.

The effect on colorectal cancer incidence and staging with population-based FOBT-screening in Sweden

AimTo investigate colorectal cancer (CRC) incidence and stage of disease in the population invited vs. not invited to the guaiac-based Fecal Occult Blood (gFOBT) and Fecal Immunochemical Test (FIT) colorectal cancer screening program in Stockholm-Gotland, Sweden, 2008–2021, and to estimate the incidence rate by sex and localization in the colorectum.MethodsThe study cohort consisted of all 60-69-years-old residents of the Stockholm-Gotland region 2008–2012 according to the population register. Screening with biennial gFOBT was successively implemented in the region by randomly inviting birth cohorts of the target group to different year of start of screening from 2008 and replaced by FIT with cut-off level 40 µg/g in women and 80 µg/g in men for a positive test in 2015. Record linkage was made to the National Cancer Register and to the Swedish Colorectal Cancer Register (SCRCR). The age-standardized CRC incidence ratio was compared in invited and non-invited during screening and in 70-75-year-olds and assessed overall and by sex, CRC stage and localization.ResultsIn total, 320,989 and 151,533 individuals were invited to a first screening and FIT round, and 5,972 CRCs were diagnosed. During screening, the overall age-adjusted incidence ratio for the gFOBT- and FIT-invited compared to the non-invited was 0.99 (95% CI 0.91–1.07) and 1.03 (95% CI 0.93–1.15), respectively. Post screening, 70–75 years of age, the overall incidence rate was 12% lower among the invited than the non-invited (RR 0.88, 95% CI 0.81–0.97). During FIT screening, the incidence for stage I and proximal CRC was 38 and 23% higher than in the non-invited (RR 1.38, 95% CI 1.09–1.76 and RR 1.23, 95% CI 1.02–1.48 respectively). The incidence post screening was 22% lower regarding stage I CRC, 13% lower in women, and 17% lower for distal CRCs as compared to the non-invited (RR 95% CI 0.78 0.63–0.95, 0.87 0.76-1.00 and 0.83 0.74–0.94 respectively).ConclusionIn the Stockholm-Gotland program, the FIT screening significantly increased the incidence rate in early staged and proximal CRCs as compared to the uninvited, and the significant decrease in the overall CRC incidence post screening was mainly seen in distal, early staged CRCs in women.

Ovarian cancer survival by residual disease following cytoreductive surgery: a nationwide study in Norway.

Residual disease (RD) following cytoreductive surgery is prognostic for epithelial ovarian cancer (EOC) patients. Few studies have evaluated RD and survival by tumor histotype and across continuous RD diameter. 2608 individuals with stages III-IV invasive EOC diagnosed between 2013 and 2022 were identified using the Cancer Registry of Norway. In flexible parametric models, we compared excess mortality comparing RD versus no macroscopic residual disease (NMRD); systemic anti-cancer therapy was evaluated in a sub-cohort from 2019. Excess mortality was assessed across continuous RD size using restricted cubic splines. Among 1849 patients with cytoreductive surgery, survival was worse for individuals with RD (vs. NMRD), excess hazard ratio (EHR) = 2.62 (95% confidence interval = (2.27-3.01)); no heterogeneity was observed by histotype (p = 0.21). Patients with 0.1-0.4 cm RD had 2-fold higher risk of death (EHR = 2.09 (1.63-2.68)) relative to women with NMRD; ~3-fold higher risk was observed for all other categories (e.g., 0.5-0.9 cm, EHR = 2.97 (2.26-3.89); 3.0-20 cm, 2.75 (2.05-3.70)). No significant difference in three-year survival was observed across continuous RD diameter (p ≥ 0.17). NMRD was associated with better survival regardless of neoadjuvant chemotherapy. Achieving NMRD resulted in the best survival outcomes. Among patients with RD, we observed no significant difference in survival by RD diameter.

Trends in Head and Neck Cancer in Saudi Arabia from 2016–2020

Objectives: Head and neck cancer (HNC) is considered the seventh most common cancer worldwide. To evaluate the incidence and geographical distributions of head and neck cancer among the Saudi population in a specific period. Methods: A retrospective and a descriptive study investigating HNC in Saudi population from January 2016 through December 2020 was conducted based on data obtained from the Saudi Cancer Registry SCR. Male and female data were included on lip, tongue, mouth salivary glands, oropharynx, nasopharynx hypopharynx, pharynx, nose and sinuses, and larynx. Age-standardized rate (ASR) and age-specific incidence rate (AIR) were calculated with a focus on age, gender and regional differences. Results: The total number of HNC cases identified by the SCR was 3,232 cases in which males were 2082 (64.4%) and females were 1,150 (35.5%). The mean ASR per 100,000 of HNC was higher in males at 4.94 compared to females at 2.9 over the study period. The mean ASR per 100,000 population ranged from 2.22 in Hail to 5.3 in Jazan, with a national average of 3.68 per 100,000. Positive correlation between HNC incidence and age was noted. Nasopharyngeal cancer showed the highest number of cases for both genders through the same period. Conclusion: Between 2016 and 2020, HNC incidence in Saudi Arabia remained stable, with consistently higher ASR in males. These findings highlight the importance of targeted, region-specific health initiatives, greater public education, and age-targeted screening to reduce the impact of HNC in Saudi Arabia.

Rotterdam Oncology Documentation (RONCDOC) – a high-quality data warehouse and tissue collection for head and neck cancer

BackgroundEvery year, almost 900.000 people are diagnosed with head and neck cancer (HNC) worldwide. HNC contains many different subsites and a large variability in tumor biology. This often results in small and/or heterogeneous study populations. Developing overarching databases is an efficient solution to collect and analyze data of these smaller subsets of patients and to facilitate data sharing among research groups. The few existing large databases often include only basic characteristics. In addition, hospital-based cohorts that include more variables are often not collected consecutively, resulting in selection bias. Therefore, we established a hospital-based cancer registry system “Rotterdam Oncology Documentation” (RONCDOC), a complete and consecutive data warehouse and tissue collection for HNC, directly registered at the source. The primary aim of this paper is to report on our data collection protocol in order to make the RONCDOC data accessible and reusable for other researchers, and to offer a blue print to other consortia planning to establish their own data warehouse.MethodsData collected in the Netherlands Cancer Registry (NCR) of patients with HNC were obtained from the Netherlands comprehensive cancer organization (IKNL) and merged with corresponding data from the electronic patient file (EPF). The data were manually verified using the EPF, and enriched with additional variables from the EPF according to an extensive data entry protocol. Furthermore, a comprehensive validation protocol was developed to guarantee the quality of the data. Tissue microarrays (TMAs) were constructed from resection specimens of patients with primary oral squamous cell carcinoma.ConclusionWith RONCDOC, we have established a consecutive and high-quality data warehouse for HNC. This paper outlines the essential steps for establishing such a data warehouse, offering a blueprint for other consortia.Trial registrationThis study was approved by the ethics committee of the Erasmus Medical Center (MEC-2016–751).

Every Third Baby with Lymphoma Under 3 Years of Age Has Inborn Error of Immunity

Background and Aims In children under 3 years of age, lymphoma is extremely rare and may be a manifestation of inborn errors of immunity (IEIs). The aim of this study was to determine TREC/KREC copy numbers and search for Slavic founder mutations (RAG1 p.Lys86ValfsTer33, IL7R p.Ser44Arg, NBN1 p.Lys219AsnfsTer16, ATM p.Glu1978Ter, UNC13D p.Arg782SerfsTer12) in patients with lymphoma up to 3 years of age from the Belarusian Cancer Registry (No.: 0170100025) over a 26-year period. Methods From 1998 to 2024, 39 patients younger than 3 years (29 males and 10 females) were diagnosed with lymphoma. The median age was 2.1 years (from 50 days to 2.8 years). 4 patients had Hodgkin's lymphoma (HL), 5—diffuse large B cell lymphoma (DLBCL), 13—lymphoblastic lymphoma (LL), 10—Burkitt lymphoma (BL), 2—peripheral T cell lymphoma (PTCL), 4—anaplastic large cell lymphoma (ALCL), 1—non-Hodgkin's lymphoma (NHL) of unspecified type. DNA was isolated from archival samples from 36 patients: bone marrow smears (n = 12), frozen bone marrow cells (n = 11), and peripheral blood cells (n = 13). Results Lymphoma in children aged 0-3 years accounted 3% (39/1237) of all pediatric lymphoma cases up to 18 years of age, including 7% (35/481) of NHL and 0.5% (4/756) of HL patients. 15/39 (38%) patients died and 3 were lost to follow-up. TREC/KREC copy numbers was reduced in 12/36 (33%) patients, 7 of whom were dead. Low TRECs/KRECs were present in all patients with DLBCL, 2 with ALCL, 2 with LL, and single cases with HL, BL, and PTCL. Homozygosity for the underlying Slavic variant of RAG1 [p.Lys86ValfsTer33] was found in a patient with DLBCL at the age of 14 months. One 12-month-old patient with PTCL was homozygous for the founder Slavic UNC13D variant [p.Arg782SerfsTer12]. Conclusions Low TREC/KREC was detected in one-third of children aged 0-3 years with lymphoma and may be used as a step 1 method to suspect the IEIs.

Tattoos and cutaneous squamous cell carcinoma: a population-based case-control study.

The prevalence of tattoos in western countries is about 20%. Tattoo ink may contain carcinogenic compounds. The aim of this study was to investigate if tattoo exposure is associated with an increased risk of cutaneous squamous cell carcinoma in individuals. In this population-based case-control study, 2857 cases aged 20 to 60 years, diagnosed between 2014 and 2017, were identified in the Swedish Cancer Registry. Statistics Sweden identified 3 random age- and sex-matched controls per case from the Swedish Total Population Register using incidence-density sampling. In 2019, participants answered a questionnaire regarding lifestyle factors, including tattoos and sun habits. We used logistic regression to investigate if tattoo exposure was associated with the relative risk of cutaneous squamous cell carcinoma. 61% of the cases and 53% of the controls replied to the questionnaire. Among the 1600 cases and the 4551 controls that participated, 15.1% and 17.6% had at least one tattoo before the index date. We found no increased risk of cutaneous squamous cell carcinoma in tattooed compared with non-tattooed individuals (incidence rate ratio, 0.95; 95% confidence interval, 0.78-1.15). Tattoo exposure was not associated with the risk of cutaneous squamous cell carcinoma in this first study of the association. However, more epidemiologic studies are needed before consensus regarding a lack of association can be reached.

Population-based analysis of breast cancer incidence and mortality: overall and age-specific temporal trends over 40-year period in Girona, Spain.

Breast cancer (BC) incidence and mortality in women have changed over time. This study aims to analyze population-level incidence and mortality trends over 40years of observation. Population-based study of BC conducted by Girona Cancer Registry covering the period 1980-2019. Age-standardized incidence and mortality rates were calculated. Poisson change-point regression models were used to analyze trends, calculating the annual percentage change (APC). A total of 12,283 diagnoses of invasive BC between 1980 and 2019. The overall age-standardized incidence rate was 109.9 (95% confidence intervals (CI) 104.4; 115.4) cases per 100,000 women-years. Trend analyses showed a statistically significant incidence increase of 4.2% per year from 1980 to 1994 (95%CI 3.3; 5.1), and a stabilization between 1994 and 2019, with an APC of 0.28% (95%CI - 0.04; 0.56). These trends were similar for the age groups 0-49years and 50-69years. In women over 69years of age, an increase in incidence of 4.4% (95%CI 2.8; 6.0) per year was observed between 1980 and 1995 followed by a non-statistically significant decrease of - 0.35% (95%CI - 0.86; 0.15) between 1995 and 2019. The overall age-standardized mortality rate was 30.3 (95%CI 29.3; 31.3) cases per 100,000 women-years. Mortality rate trends showed a statistically significant decrease of - 1.87% (95%CI - 2.38; - 1.37) per year since 1992. There has been a stabilization in the incidence of BC and a gradual decline in BC mortality in women. The introduction of mammography in the mid-1990s, alongside early detection and treatment due to screening programs may play a significant role in the reduction of BC burden in women of all ages.

Age-specific trends in colorectal, appendiceal, and anal tumour incidence by histological subtype in Australia from 1990 to 2020: a population-based time-series analysis.

Early-onset bowel cancer incidence (age <50 years) has increased worldwide and is highest in Australia, but how this varies across histology and anatomical site remains unclear. We aimed to investigate appendiceal, proximal colon, distal colon, rectal, and anal cancer incidence trends by age and histology in Australia. Cancer incidence rate data were obtained from all Australian cancer registries (1990-2020 period). Birth cohort-specific incidence rate ratios (IRRs) and annual percentage change in rates were estimated using age-period-cohort modelling and joinpoint regression. After excluding neuroendocrine neoplasms, early-onset cancer incidence rose 5-9% annually, yielding 5,341 excess cases (2 per 100,000 person-years; 12% appendix, 45% colon, 36% rectum, 7% anus; 20-214% relative increase). Trends varied by site, period, and age: appendiceal cancer rose from 1990-2020 in 30-49-year-olds; colorectal cancers rose from around 1990-2010 in 20-29-year-olds and from 2010-2020 in 30-39-year-olds; anal cancer rose from 1990-2009 in 40-49-year-olds. Across all sites, IRRs increased with successive birth cohorts since 1960. Notably, adenocarcinoma incidence in the 1990s versus 1950s birth cohort was 2-3-fold for colorectum and 7-fold for appendix. The greatest subtype-specific increases occurred for appendiceal mucinous adenocarcinoma, colorectal non-mucinous adenocarcinoma, and anal squamous cell carcinoma. Only later-onset (age ≥50) colorectal and anal adenocarcinoma rates declined. Appendiceal tumours, neuroendocrine neoplasms (all sites), anorectal squamous cell carcinomas, and colon signet ring cell carcinomas rose across early-onset and later-onset strata. Appendiceal, colorectal, and anal cancer incidence is rising in Australia with variation across age and histology, underscoring the need to identify factors driving these trends. ALM is supported by an Australian Government Research Training Program Scholarship, Rowden White Scholarship, and WP Greene Scholarship. DDB is supported by a National Health and Medical Research Council of Australia (NHMRC) Investigator grant (GNT1194896), a University of Melbourne Dame Kate Campbell Fellowship, and by funding awarded to The Colon Cancer Family Registry (CCFR, www.coloncfr.org ) from the National Cancer Institute (NCI), National Institutes of Health (NIH) [award U01 CA167551]. MAJ is supported by an NHMRC Investigator grant (GNT1195099), a University of Melbourne Dame Kate Campbell Fellowship, and by funding awarded to the CCFR from NCI, NIH [award U01 CA167551].

Historical redlining and mortality in children, adolescents, and young adults with cancer in California, 2000-2019.

Historical redlining, or the Home Owners Loan Corporation (HOLC) program's racially biased mortgage risk monitoring maps in the 1930s, is implicated in shaping modern neighborhoods and health outcomes. This retrospective cohort study evaluates the association between redlining and mortality in young cancer patients. Using the California Cancer Registry, we identified patients <25 years old diagnosed with malignant cancer between 2000-2019. HOLC maps were spatially joined with patient address at diagnosis to determine redlining status (A "Best", B "Still Desirable", C "Declining", D "Hazardous"). Census tract-level U.S. Census and American Community Survey data were appended to determine modern neighborhood characteristics. The Kaplan-Meier method was used to evaluate overall survival and multivariable Cox proportional hazards models to estimate the associations between HOLC grade and mortality, adjusting for clinical and multilevel social drivers of health. In total 8,108 patients resided in HOLC-graded neighborhoods among 51,084 patients statewide. Overall survival at 5 years was inferior for patients who resided in D graded neighborhoods at diagnosis vs A graded neighborhoods (80.3%, 95% CI: 78.6-81.8 vs 88.5%, 95% CI: 84.3-91.6). Adjusting for clinical characteristics, patients in D graded neighborhoods experienced greater mortality (HR 1.32, 95% CI: 1.12-1.56) compared with those in A and B graded neighborhoods. Additional adjustment for insurance attenuated the effect (HR 1.17, 95%CI: 1.00-1.36) and for neighborhood socioeconomic status marginally attenuated the effect (HR 0.96, 95% CI: 0.81-1.13). Findings suggest enduring legacy effects of historical redlining on young individuals with cancer, potentially mediated social factors including health insurance.

Father's occupation and colorectal cancer in his adult offspring.

Growing evidence suggests transmission of colorectal cancer risk through the maternal line. There is scant information about transmission through the paternal line, despite plausible evidence from mammal experiments. We examined the association between paternal occupation and colorectal cancer in the Child Health and Development Studies, a multi-generational cohort followed for 60 years. Pregnant mothers completed in-person interviews at enrollment (1959-1966) and reported demographic and health-related information for her and her husband, including occupation. Colorectal cancer in adult (age ≥18 years) offspring was ascertained from a population-based cancer registry. Cox proportional hazards models were used to estimate adjusted hazard ratios (aHR), with follow-up accrued from birth through cancer diagnosis, death, or last contact. Paternal occupations included: 37.0% professional, technical, or managerial; 13.7% clerical or sales; 30.5% crafts or operative; and 17.5% service work or labor. Over 716,133.5 person-years of follow-up, 79 offspring were diagnosed with colorectal cancer (median age at diagnosis: 50 years [range: 23-59 years]). Offspring of fathers employed in crafts, operative, service, or labor occupations had higher incidence rates of colorectal cancer (15.66 per 100,000 person-years; 95% CI 11.47, 20.89) compared to professional, technical, or managerial occupations (6.84 per 100,000 person-years; 95% CI 3.73, 11.47). Risk associated with these occupations remained elevated after adjustment for maternal race, paternal education, and offspring year of birth (aHR 1.99; 95% CI 1.03, 3.87). Our findings support possible transmission of cancer risk through the paternal line.

Clustering of childhood acute leukemia in Finland: a nationwide register-based study.

Acute leukemia is the most common childhood malignancy, with suspected contributions from environmental factors and immune responses to common pathogens. A recent meta-analysis indicated possible spatiotemporal clustering, though the findings were hindered by data quality limitations. We investigated spatial and spatiotemporal clustering of childhood leukemia using advanced methods and complete residential histories. We included patients aged 0-17years diagnosed in 1990-2019, using data from the Finnish Cancer Registry. A 1:3 age- and sex-matched case-control design was employed and residential history data with exact coordinates was collected. Clustering was evaluated using the Cuzick-Edwards test, Knox test, Kulldorff's scan statistic, and Jacquez's Q statistic. The dataset included 1,626 childhood leukemia cases (median age 5.0years, 54% male). The Knox test revealed no evidence of spatiotemporal clustering. However, the Cuzick-Edwards test revealed spatial clustering at diagnosis addresses for children under 1year (OR 1.35, 95% CI 1.14-1.57). Further analysis with Jacquez's Q test using complete residential histories identified significant spatiotemporal clustering in young children (ages 1.5-5.99years) with acute lymphoblastic leukemia (ALL, p = 0.037). We also tested for co-incidence between leukemia and type 1 diabetes but found no clustering. Overall, we found limited evidence for clustering. In the subgroup analyses, significant spatiotemporal clustering in acute lymphoblastic leukemia cases among children aged 1.5-5.99years was observed, coinciding with the peak incidence in early childhood. Previous research has shown that this age group has distinct genetic characteristics and may possess a unique etiology.

Enhancing Malignancy Detection and Tumor Classification in Pathology Reports: A Comparative Evaluation of Large Language Models.

Cancer registries require accurate and efficient documentation of malignancies, yet current manual methods are time-consuming and error-prone. This study evaluates the effectiveness of large language models (LLMs) in classifying malignancies and detecting tumor types from pathology reports. Using a synthetic dataset of 227 reports, the performance of four LLMs and a score-based algorithm was compared against expert-labeled gold standards. The LLMs, particularly GPT-4o and Llama3.3, demonstrated high sensitivity and specificity in both malignancy detection and tumor classification, significantly outperforming traditional algorithms. LLMs enhance the accuracy and efficiency of cancer data classification and hold promise for improving public health monitoring and clinical decision-making.

Thyroid Cancer Incidence and Trends in United States and Canadian Pediatric, Adolescent, and Young Adults.

Thyroid cancer incidence has risen in both the United States and Canada, despite differing healthcare systems. While overdiagnosis likely partly explains this trend in adults, its impact on younger populations is unclear. We used the North American Association of Central Cancer Registries, which included 133,808 thyroid cancer cases from the United States and Canada, to assess incidence trends among pediatric, adolescent, and young adult (PAYA) populations. Age-adjusted incidence rates (AAIR) per 100,000 person-years (PY) were compared using rate ratios (RR), stratified by sex, age, race/ethnicity (United States only), and histology. Joinpoint regression estimated annual percentage changes (APC) and average APCs (AAPC) in AAIRs. From 1995 to 2014, thyroid cancer incidence increased by 137%. Significant increases occurred across all age groups (0-14, 15-24, 25-34, 35-39 years). The rate increase was highest for papillary thyroid cancer (AAPC = 5.50, 95% CI 5.06, 5.94), and among individuals aged 35-39 years (AAPC = 5.99, 95% CI 4.84, 7.15). Of racial/ethnic groups in the United States, non-Hispanic White individuals had the highest AAIR (6.22 per 100,000 PY). Mortality has changed minimally. Over the past two decades, thyroid cancer incidence has increased in individuals under 40. While evidence suggests that overdiagnosis primarily accounts for this trend, other contributing factors cannot be ruled out. Further research and surveillance of the drivers of increased incidence are critical.

Development of a Synthetic Oncology Pathology Dataset for Large Language Model Evaluation in Medical Text Classification.

Large Language Models (LLMs) offer promising applications in oncology pathology report classification, improving efficiency, accuracy, and automation. However, the use of real patient data is restricted due to legal and ethical concerns, necessitating privacy-compliant alternatives. This study aimed to develop a synthetic oncology pathology dataset to serve as a benchmark for LLM evaluation, enabling reproducible and privacy-preserving AI research. A total of 227 synthetic pathology reports were generated using Microsoft Copilot, ChatGPT Plus, and Perplexity Pro to ensure structural and linguistic diversity. The dataset included cases of prostate (n=75), lung (n=78), and breast (n=74) cancer, evenly distributed between malignant (n=113) and benign (n=114) findings. Reports were reviewed and classified by three independent cancer registrars using a consensus-based validation process. The dataset provides a structured, clinically relevant benchmark for evaluating LLM performance in pathology text classification. It enables AI model assessment without compromising data privacy, paving the way for scalable and ethical AI-driven oncology documentation.