BACKGROUND:Hepatic artery infusion pump chemotherapy (HAIC) improves colorectal liver metastasis (CRLM) outcomes, but treatment-associated biliary sclerosis (BS) is a major complication. This study evaluates the incidence of and risk factors for BS after HAIC for CRLM. METHODS:This was a single-center retrospective analysis including all patients with CRLM treated with floxuridine between 2000 and 2022. The primary outcome was BS, intractable hyperbilirubinemia, and/or biliary stricture not caused by cancer progression requiring percutaneous or endoscopic stenting or drainage. BS was analyzed using competing risk methods where death was the competing event. RESULTS:Between 2000 and 2022, a total of 2239 patients received HAIC for CRLM, 48% (n=1067) received adjuvant HAIC and 52% (n=1172) were unresectable. There were 128 (6% at 60 months) BS events, and rates did not differ between adjuvant (n=66) and unresectable (n=62) groups (60-month estimate: 6% [95% confidence interval (CI) 5-8] vs 6% [95% CI 4-7]; p=0.362). Operative variables, variant hepatic arterial anatomy, non-gastroduodenal artery cannulation, and extrahepatic perfusion were not associated with BS. In adjuvant cases, BS risk increased with each cycle (hazard ratio [HR] 1.123, p=0.003) and every 100 mg of cumulative floxuridine (HR 1.087, p<0.001) controlling for concurrent colorectal resection and variant vessel ligation. Among unresectable patients, each cycle (HR 1.087, p<0.001) and every 100 mg of cumulative floxuridine (HR 1.054, p<0.001) increased BS risk, controlling for race and concurrent colorectal resection. CONCLUSIONS:The incidence of BS after HAIC for CRLM is 6% after 60 months and does not differ in the adjuvant and unresectable setting due to meaningful differences in survival. Cumulative floxuridine dose and number of HAIC cycles were independently associated with increased BS risk.

The prognostic value of the closest margin and margin of the largest tumor may depend on pathologic response to preoperative chemotherapy in colorectal liver metastases (CRLM). This study investigated the effect of margin status on overall survival (OS) stratified by pathologic response to preoperative chemotherapy. Patients with resected CRLM who received preoperative therapy were included. Responders were stratified by ≥ 75

Survival outcomes for intrahepatic cholangiocarcinoma (IHC) remain poor due to late-stage presentation and limited effective treatment options. Previous studies have suggested age-related differences in outcome, but the results are mixed, and there is substantial variation in the definition of “young.” A retrospective review analyzed 938 consecutive patients with a diagnosis of IHC between 2000 and 2018 at a single institution. Risk factors for IHC, clinical characteristics at diagnosis, and genomics were assessed for potential differences at the extremes of age (≤50 vs ≥70 years). The final cohort (n = 430) included 142 patients 50 years or younger and 288 patients older than 70. The most common genetic alterations were IDH1mut (22

Background:The Institut Mutualiste Montsouris (IMM) 3-level complexity classification has been validated for laparoscopic liver resection (LLR) in several studies with small sample size. However, it has not been well-validated in large studies down to the individual procedure type. Hence, in order to address current limitations in the studies validating the IMM complexity classification, we performed an international multicenter study to validate the IMM complexity classification across its three complexity levels and the categorization of the 11 distinct procedure types. Methods:A retrospective cohort study of 22,252 patients undergoing LLR across 64 centers worldwide between 2005 and 2021 was performed. Baseline characteristics and perioperative outcomes were analyzed across the three difficulty levels and 11 procedure types of the IMM complexity classification. Results:A total of 14,765 patients were included in our final analysis. The main indications for LLR in our study was hepatocellular carcinoma or intrahepatic cholangiocarcinoma (n=7,781, 52.7%) followed by liver metastasectomy (n=3,911, 26.5%). In terms of underlying liver pathology, 5,127 (34.7%) cases had cirrhosis, and 1,214 (8.3%) had portal hypertension. Perioperative outcomes including operative time, open conversion rate, intraoperative blood loss, need for intraoperative blood transfusion, need for Pringle's application, length of stay, postoperative morbidity, major postoperative morbidity and 90-day mortality all demonstrated a significant increasing trend with increasing IMM complexity grades (P<0.001). These trends remained significant following adjustment for baseline characteristics (P<0.001). Notably, when examining the 11 LLR procedure types, all procedures within each IMM complexity grade were individually higher than all procedures in the preceding complexity grade for operative time, blood loss, length of stay, postoperative morbidity and major postoperative morbidity. Conclusions:The three IMM complexity grades were well associated with LLR complexity as determined by key surrogate perioperative measures. Our findings also supported the categorization of the 11 distinct LLR procedures into the three complexity levels.

Supplementary Figure 1: Number of cancer cases and deaths across indications in 2020, Millions

OBJECTIVE:To compare the incidence of pancreatic adenocarcinoma (PDAC) during surveillance for intraductal papillary mucinous neoplasm (IPMN) to a Surveillance, Epidemiology, and End Results (SEER) matched cohort. BACKGROUND:IPMN are pre-malignant lesions and prescribed lifelong surveillance. Many still develop PDAC despite annual imaging, while others never demonstrate cyst progression. The true magnitude of PDAC risk attributable to IPMN remains unclear. METHODS:Retrospective review of patients under surveillance for IPMN at a high-volume center was performed using an institutional cyst database which includes demographics, serial radiographic cyst features, and pathological outcomes. Standardized incidence ratio (SIR) for PDAC was derived from age and year matched SEER group compared to the study cohort. RESULTS:Overall, 1494 patients underwent 11,107 scans with median surveillance of 55.3 months for 7,681 person-years at risk. Median age was 66 years, 64% were female, and median index cyst size was 1.4 cm. Cumulative incidence of malignancy at 60 months was 0.97% (CI:0.53%-1.70%). Compared to SEER population, PDAC risk was 10-fold higher (CI:7.2-15.6) during surveillance with 26 events (16 IPMN-HGD, 10 PDAC) compared to 2.44 expected. Highest risk was noted with MPD dilation ≥5 mm (SIR:26.5, CI:11-64), and BMI ≥30 (SIR:20.6, CI:11.4-37). Whereas IPMN ≤1 cm (SIR:3.28, CI:0.82-13.1), and age<60 (SIR:6.27, CI:1.58-24.9) were closest to the population risk. Cysts <3 cm and without MPD dilation or solid component (guideline-negative), had SIR:8.40 (CI: 5.15-13.7). CONCLUSION:Risk of malignancy with IPMN is significantly higher than the general population but varies by demographics and cyst features. A personalized surveillance approach may improve early detection in high-risk patients and reduce imaging burden in low-risk groups.

Importance:The best local therapy approach for patients with locally advanced pancreatic cancer (LAPC) remains undefined. Objective:To evaluate hypofractionated ablative radiation therapy (A-RT) followed by evaluation for surgery in patients with LAPC whose cancer remained unresectable after induction chemotherapy. Design, Setting, and Participants:This was a phase 2 single-arm nonrandomized clinical trial conducted at the Memorial Sloan Kettering Cancer Center, New York, New York, from June 2018 to April 2024. Participants with histologically confirmed pancreatic adenocarcinoma unresectable by multidisciplinary review after 3 or more months of modified fluorouracil, leucovorin, irinotecan, and oxaliplatin (mFOLFIRINOX) or gemcitabine plus nab-paclitaxel were eligible. Intervention:Hypofractionated A-RT (67.5 Gy in 15 fractions or 75 Gy in 25 fractions) with capecitabine followed by evaluation for resection. Main Outcomes and Measures:Coprimary end points included resectability and 2-year overall survival. Secondary end points included 90-day surgical adverse events (AEs) and 2-year locoregional and distant metastasis rates. Results:Forty-eight participants underwent A-RT (25 [52.1%] male; median [range] age, 67 [50-80] years; median [range] tumor size 4.1 [2.4-8.3] cm; 47 (98%) with arterial and venous involvement; median [range] carbohydrate antigen 19-9, 88 [<1-1601] U/mL). Forty-five patients (94%) received mFOLFIRINOX. Seventeen participants (34%) underwent a laparoscopy, with resection performed in 13 (27%). Resections included 11 pancreaticoduodenectomies (84.6%) and 2 distal pancreatectomies (15.4%). At median follow-up of 3 years, 2-year overall survival from A-RT for the cohort was 38% (95% CI, 26%-54%) overall and 31% (95% CI, 19%-51%) and 54% (95% CI, 33%-89%) in participants with unresected and resected disease, respectively. Two-year rates of local progression were 11% (95% CI, 3.5%-25%) and 15% (95% CI, 2.2%-40%) in unresected and resected groups, respectively. Two-year rate of distant metastasis was 73% (95% CI, 57%-84%). There were no deaths within 90 days of surgery. Eleven surgical AEs (including 4 Clavien-Dindo grade III) occurred in 12 evaluable participants. Acute and late RT-related grade III or higher AEs occurred in 6 (12.5%) and 12 (26.1%) participants, respectively. The most common late AE was ascites in 7 (15.2%; 3 postsurgery). Conclusions:For patients with LAPC with extensive vascular encasement after induction chemotherapy, hypofractionated A-RT prior to possible surgery was associated with promising overall survival for both ultimately resected and unresected groups. A-RT was not associated with an increase in postsurgical morbidity when surgery was performed and was an effective destination therapy when it was not. Further evaluation of this strategy to maximize outcomes regardless of resection is warranted. Trial Registration:ClinicalTrials.gov Identifier: NCT03523312.

Objective Colorectal cancer (CRC) is the second leading cause of cancer-related deaths worldwide, with incidence and mortality rising in low- and middle-income (LMICs) countries. Expanding the number of healthcare professionals trained in colonoscopy is crucial to addressing the increasing cancer burden. We developed a colonoscopy simulation-based training course and assessed whether training on locally developed low-fidelity (LF) simulation models resulted in skill acquisition and confidence comparable to training on high-fidelity (HF) models. Design A randomized crossover trial was conducted as part of a colonoscopy training program, which consisted of a 9-week online lecture series and a 4-day in-person colonoscopy training workshop, using HF and LF simulators. Performance was assessed using the Mayo Colonoscopy Skills Assessment Tool (MCSAT) and the Global Assessment of Gastrointestinal Endoscopic Skills (GAGES) tool. The primary outcome was the difference in median exam scores. Setting The study was conducted at the Obafemi Awolowo University Teaching Hospital in Ile-Ife, Nigeria, from January 4th, 2024, to May 3rd, 2024. Participants Participants were surgeons, surgical residents, and gastroenterologists of varying levels of experience from Nigeria. Results A total of twenty-one trainees enrolled in the online Zoom lectures, and 18 participants were randomized in the simulation course. Median MCSAT and GAGES scores improved over time in both HF and LF groups. For GAGES scores, only Exam 2 (after lectures and training on a simulation model) showed statistically significant differences for scope navigation (p = 0.046), use of strategies (p = 0.019), and the overall GAGES score (p = 0.041). No significant differences were observed at Exam 4 (the final exam of the course), suggesting that skills had converged by the end of training. Both HF and LF groups demonstrated significant increases in confidence. Conclusions This pilot trial suggests training on LF colonoscopy models results in similar colonoscopy skill acquisition and confidence compared to HF colonoscopy. LF colonoscopy training models can be used as cost-effective teaching tools in LMICs with limited resources and financial investments.

OBJECTIVES:Head and neck cancer (HNC) remains a major global cause of cancer morbidity and mortality, with disproportionately poor outcomes in low- and middle-income countries. Substantial cross-national disparities suggest an important role for health system strengthening. We evaluated associations between national health system characteristics and global HNC outcomes. METHODS:We conducted a cross-sectional ecological analysis of 185 countries using sex-stratified, age-standardized incidence and mortality estimates from the International Agency for Research on Cancer (IARC) GLOBOCAN 2022 database. The primary outcome was the composite mortality-to-incidence ratio (MIR) for aggregated HNCs, including cancers of the lip and oral cavity, oropharynx, larynx, nasopharynx, hypopharynx, and salivary gland. Eleven national health system indicators capturing health financing, workforce density, service availability, socioeconomic development, and gender equity were evaluated. Univariable linear regressions identified candidate variables using Bonferroni correction (p < 0.0045), followed by multivariable modeling with assessment for multicollinearity using variance inflation factors. RESULTS:All 11 health system indicators were significantly associated with HNC MIR on univariable analysis (p < 0.001 for all). In multivariable analysis of 123 countries with complete data, higher Universal Health Coverage (UHC) service coverage index and higher gross domestic product (GDP) per capita were independently associated with lower (improved) HNC MIR (model R2 = 0.70). Findings were consistent in sex-stratified analyses. CONCLUSIONS:Across countries, progress toward universal health coverage and greater national economic capacity was independently associated with improved HNC outcomes. These findings may help to inform efforts at the level of health systems to improve HNC outcomes worldwide. LEVEL OF EVIDENCE:N/A.

PURPOSEBreast cancer (BC) is a significant health challenge in Nigeria, exacerbated by early onset, advanced-stage diagnosis, and high prevalence of triple-negative tumors. Access to genetic testing and counseling is scarce, with minimal capacity for hereditary cancer services. Despite these barriers, there is strong interest in expanding care to include genetic testing and improve understanding of familial risk. The purpose of this study was to develop and assess the effectiveness of a BC genetics education program for Nigerian health care providers (HCPs).METHODSA multidisciplinary international team developed a four-module hybrid education program combining asynchronous online learning and an in-person didactic session. Invitations were circulated to HCPs in tertiary hospitals across Nigeria. Knowledge improvement was assessed using standardized pre- and postmodule tests.RESULTSThirty-one physicians and nurses participated. All online modules had significant knowledge improvement, with the largest score increases in BRCA1/2 genetic counseling (mean change, 1.9 [95% CI, 1.3 to 2.5]; P < .001) and BRCA1/2 clinical management (mean change, 1.6 [95% CI, 1.2 to 2.1]; P < .001). The subsequent in-person workshop had additional, albeit smaller, module increases. Aggregated analysis showed a 23.0% increase in knowledge after the online training (P < .001), with a further 10.1% gain after the in-person workshop (P = .007). Overall knowledge improved from 45.0% at baseline to 87.0% post-training, representing a 43.0% absolute gain (P < .001).CONCLUSIONThis hybrid training program significantly improved provider knowledge of hereditary BC genetics in Nigeria and offers a scalable, culturally tailored model for expanding BC genetic services in low-resource settings. While promising, the modest sample size and limited follow-up warrant further evaluation and broader rollout to confirm long-term effectiveness.

PURPOSE:Intrahepatic cholangiocarcinoma (iCCA) has been a contraindication for liver transplantation due to frequent recurrence and poor survival. We sought to determine the true proportion of transplantation-eligible iCCA patients and their outcomes without transplantation. METHODS:Data from patients evaluated for iCCA at two academic medical centers between 2008 and 2018 were analyzed retrospectively. Overall survival was determined for patients categorized as eligible for liver transplantation based on six criteria: unresectability, no extrahepatic disease, stable disease after chemotherapy, functional status (maximum age, 72 years), no major comorbidity, and no involvement of major vasculature. RESULTS:Of 1407 iCCA patients with sufficient data, 327 (23%) had advanced liver-confined disease. Most (n = 180 [55%]) progressed or died before 6 months of therapy. Others were ineligible owing to resection after pretreatment (n = 47), poor performance status (n = 8), advanced age (n = 32), comorbidity (n = 19), or extensive extrahepatic vascular involvement (n = 2). Only 39 patients (12% of 327; 2.8% of 1,407; 95% confidence interval [CI] 2-3.8%) met eligibility, with more than half (n = 22 [56%]) treated with hepatic arterial infusion pump chemotherapy. Their median survival from diagnosis was 39 months (95% CI 28-62 months); 3-year survival from estimated eligibility was 46% (95% CI 32-65%). CONCLUSIONS:Patients who meet stringent eligibility criteria for liver transplantation constitute approximately 3% of patients with iCCA. The observed survival exceeds benchmarks established for systematic therapy alone and is likely attributable to favorable tumor biology and hepatic artery infusion chemotherapy.

Background/Objectives: The iBreast Exam (iBE) electronically palpates the breast to identify possible abnormalities. The purpose of this study was to assess iBE feasibility and compare it to Clinical Breast Exam (CBE) for breast lesion detection. Methods: Prospective evaluation of 300 asymptomatic women, ≥18 years old, with CBE, iBE, and mammography was performed. Sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV) of iBE and CBE for detecting suspicious breast lesions were calculated using breast imaging as the reference standard. For women with one year follow up, the sensitivity, specificity, PPV, and NPV for cancer detection were calculated. Results: 300 women (mean age 58.9 years) underwent CBE, iBE, and mammography. In 2/300 (0.7%), CBE was positive; in 1/300 (0.3%), iBE was positive; and in 24/300 (8%), screening mammograms were positive. Nine had suspicious imaging findings with biopsy (three malignant and six benign). Of three cancers, all visualized mammographically, CBE and iBE detected an ipsilateral breast abnormality in one woman and missed two cancers (<2 cm). Sensitivity, specificity, NPV, and PPV of iBE and CBE were similar, with no statistically significant difference in NPV or PPV for detection of suspicious breast findings or breast cancer (p > 0.05). Conclusions: Mammography detected all breast cancers in our cohort and remains the standard of care. iBE is feasible to perform. Our pilot data demonstrates iBE performed similarly to CBE by trained nurse practitioners. Given our small study population, further investigation is warranted into the potential use of iBE where trained healthcare practitioners are not readily available.

Background In Nigeria, cancer patients often pay for care out-of-pocket, leading to catastrophic health expenditure (CHE). However, data on the true costs and economic burden of cancer care are limited. This study prospectively analyzes direct and indirect out-of-pocket costs for colorectal cancer (CRC) care at a tertiary hospital in Southwest Nigeria. Methods Patients newly diagnosed with CRC between August 2019 and April 2024 were enrolled. Trained research assistants administered a context-specific questionnaire at admission and six months later. Patients reported household income and all cancer-related expenditures. CHE was defined using three standard thresholds: healthcare costs exceeding 40 % of capacity-to-pay or 10 % and 25 % of annual income. Results Data were collected from 50 patients with a mean age of 57.9 years (SD 14.3). Twelve percent of patients (6/50) presented with stage I disease. Forty-two percent (21/50) had stage II disease, while 46 % (23/50) had stage III or IV disease at presentation. 24 % (12/50) of patients had right-sided disease, while 38 % (19/50) had rectal cancer, 4 (21 %) of whom received neoadjuvant radiotherapy. Ninety percent (45/50) of patients received systemic chemotherapy. The mean annual capacity-to-pay for the cohort was $3930.71(SD $5108.10), while the mean cost of care was $5286.16(SD $2919.77). Indirect costs, including travel, lodging, and lost income, accounted for $2144.04(SD $2478.68) of the total cost of care. Between 95.2 % and 100 % of our cohort experienced a CHE. Only 20 % (10/50) of our cohort had health insurance. Conclusions More than 95 % of patients seeking care for CRC at a tertiary care facility in Southwest Nigeria experience a CHE because of out-of-pocket costs associated with accessing care. There is the need for more studies on interventions to reduce these cost barriers for patients. Policy Summary A more effective and accessible health insurance scheme is urgently needed in Nigeria to protect CRC patients from CHE.

IntroductionAlthough palliative care (PC) is recognized as a component of comprehensive care throughout the life course and universal health coverage, more than 85% of the global need for PC is unmet. Nigeria has isolated PC provision, with a narrow scope, limited availability, and inadequate support for PC services, despite having adequate legislation to support the operation in the country. Given the rising incidence of chronic illnesses that are amenable to PC in Nigeria, this study aimed to explore healthcare professionals' (HCPs') knowledge and attitudes towards PC and identify targets for improving PC delivery nationally.MethodsUsing a mixed-methods convergent design, an online survey was distributed to multidisciplinary HCPs in Nigeria. The survey included validated tools that assessed knowledge and attitudes related to PC, pain management, and end-of-life (EOL) care. Five focus group discussions (FGDs) were conducted and qualitatively analyzed using rapid matrix analysis to triangulate themes with survey responses.ResultsFrom February to December 2024, 117 HCPs from 26 hospitals across five of six geopolitical zones of Nigeria completed the quantitative survey. Most respondents were nurses (67%) or physicians (26%), worked in public teaching hospitals (70%), and provided care for patients with cancer (94%). Over half (56%) received some PC training, though many expressed interest in additional training on pain management (65%), EOL communication skills (50%), and EOL ethics and decision-making (49%). FGDs from three geopolitical zones revealed 3 major themes: (1) HCPs have a shared high-level understanding of the goals of PC with varied perceptions around PC delivery; (2) Challenges in the practical delivery of PC include systemic barriers, opioid costs and stigma, and patient receptivity; (3) Delivering PC impacts HCPs' job satisfaction and takes a psychological toll. Participant-identified strategies for overcoming barriers to PC delivery included the development of standardized PC protocols, opioid policy changes, patient education about PC, and HCP training and support.ConclusionHCPs in this study demonstrated broad interest in and knowledge about PC. However, systemic barriers and limited patient and caregiver receptivity limit PC uptake in clinical practice. Future initiatives should involve clinical training related to PC communication, community education programs, and implementation of contextually appropriate protocols and policies for PC delivery.

Clinical practice guidelines (CPGs) were developed to standardize and optimize cancer care delivery in low- and middle-income countries (LMICs). The aim of this scoping review is to identify implementation science (IS) frameworks and strategies to promote CPG adoption and adherence in LMICs. We identified studies that describe, develop, reference, or utilize IS frameworks or strategies to deliver or evaluate adoption of oncology CPGs in LMICs. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR), we searched Medline, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), African Journals Online, Latin American and Caribbean Health Sciences Literature, Scopus, Web of Science, and PsycINFO on 3/21/2022, 12/20/2022, 01/26/2024, and 10/03/2025. Publications in all languages and of all study types were eligible for inclusion. Titles, abstracts, and full text were screened by two reviewers with conflicts resolved by a third reviewer. Excluded studies did not use an IS framework or strategy, did not focus on cancer care, or were conducted in a high-income country. Searches identified 17,806 unique publications for title/abstract screening. 182 studies met criteria for full-text review; 35 were included. Twenty-four studies were country-specific, most commonly India (n = 5) and Nigeria (n = 5). The most frequent CPGs referenced were national/country-specific guidelines (n = 13) and Breast Health Global Initiative (BHGI) guidelines (n = 9). Only 16 full-text publications described original research to promote or evaluate evidence-based CPG interventions in LMICs. Established IS frameworks used in more than one study were the Exploration, Preparation, Implementation, and Sustainment (EPIS) framework (n = 2) and Consolidated Framework for Implementation Research (CFIR) (n = 2). There is limited research utilizing IS frameworks and strategies to promote CPG adoption and adherence in LMICs and substantial heterogeneity in reporting. Greater utilization of IS frameworks, strengthening implementation capacity and improving consistency in reporting, are needed to improve CPG uptake in LMICs. Open Science Framework (https://osf.io/nb75s).

Importance:Colorectal cancer (CRC) incidence is increasing in sub-Saharan Africa, including Nigeria, where more than half of patients die within 1 year due to late diagnosis and limited treatment. Without ready access to screening, identifying risk factors is critical. However, regional data are scarce, and it is unclear whether risk factors from high-income populations apply to Nigerians, who may have distinct clinicodemographic patterns and tumor characteristics. Objective:To assess whether 13 CRC risk factors identified in high-income populations are associated with CRC in adults in Nigeria. Design, Setting, and Participants:This multicenter case-control study was conducted in 6 hospitals across 3 of Nigeria's 6 geopolitical zones, with recruitment from April 2020 to September 2024. Participants were patients newly diagnosed with CRC and cancer-free controls matched on age, sex, location, education, and recruitment site. Data were analyzed from April 2023 to November 2025. Exposures:Family history of cancer, anthropometric factors (adult height, body mass index [BMI], and somatotype or body size over the life course), physical activity, and dietary and lifestyle factors (processed meat, red meat, fruit, vegetables, fiber, alcohol, smoking, and diabetes). Main Outcomes and Measures:The associations of risk factors with CRC were assessed using odds ratios (ORs) and 95% CIs, estimated using multivariable unconditional logistic regression. Results:Among 2063 participants (1103 [53.5%] male; median [IQR] age, 54 [42-64] years) enrolled, most had secondary education or higher (1756 participants [85.2%]), lived in urban areas (1817 participants [88.1%]), and were Yoruba (1524 participants [73.9%]). Cases were older than controls (median [IQR] age, 56 [45-66] vs 52 [41-63] years) but were otherwise similar. Family history of cancer (OR, 1.55; 95% CI, 1.06-2.26), adult height (OR per 5-cm increase, 1.09; 95% CI, 1.00-1.20), body size or type over the life course (tertile 3 vs tertile 1: OR, 1.88; 95% CI, 1.30-2.72), and processed meat intake (lowest vs highest tertile of intake: OR, 1.43; 95% CI, 1.07-1.91) were positively associated with CRC risk, and an inverse U-shaped association was identified between physical activity and CRC risk (OR vs first quartile of activity, quartile 2: 0.63; 95% CI, 0.44-0.88; quartile 3: 0.52; 95% CI, 0.34-0.79; quartile 4: 0.77; 95% CI, 0.58-1.02), consistent with data from high-income populations. BMI was inversely associated with CRC risk (OR vs BMI 18.5-24.9, BMI <18.5: 2.63; 95% CI, 1.79-3.86; BMI ≥30: 0.42; 95% CI, 0.28-0.63), likely reflecting prediagnostic weight loss. No associations with diabetes, smoking, or intakes of red meat, fruit, vegetables, fiber, or alcohol were observed, likely due to low prevalence or variability. Conclusions and Relevance:This case-control study of CRC risk factors in adults in Nigeria found CRC risk patterns similar to those observed in high-income populations. These insights offer immediate opportunity to tailor education, prevention, and early detection in the region, especially as economic development leads to lifestyles more similar to high-income populations. These data provide a foundation for prospective research and investigations in other African populations.

e15655 Background: Although the role of chemotherapy in the treatment of non-metastatic colorectal cancer (CRC) has been established little is known on its role and treatment outcomes in the Nigerian cohort. Recent data shows a unique biology and a prevalence of 28% of mismatch repair deficient CRC in the Nigerian population. It is unclear if this unique biology informs similar treatment approach in established guidelines. We set out to study the treatment outcomes of surgery, and chemotherapy in stage II and III CRC in Nigeria. Methods: This is a retrospective study on 381 patients with stage 2 and 3 CRC from 2013 to 2023. Patients were staged using the 8 th edition AJCC TNM. Data on these patients were retrieved from the African Research Group for Oncology (ARGO) database. Patients in each stage were subdivided into 5 groups based on the modality of treatment received (surgery alone; neoadjuvant therapy + surgery; surgery + adjuvant therapy; neoadjuvant therapy + surgery + adjuvant therapy; and chemotherapy alone). Primary end points of the study were overall survival (OS) and disease-free survival (DFS); and secondary end point was to identify the modality of treatment most appropriate for each stage. Data was analysed using Kaplan Meier curves and log rank test for survival outcomes. Results: The study showed that 40.6% of stage 2 and 40.4% of stage 3 CRC patients had surgery alone, while 59.4% and 59.6% of stage 2 and 3 respectively had combined chemotherapy and surgery. The median disease-free survival and overall survival was similar for both stages (10 months, p = 0.508; 9 months, p = 0.722 respectively). Patients with stage 3 who had either neoadjuvant chemotherapy or adjuvant chemotherapy had better OS when compared with patients who had surgery alone (p = 0.021; p = 0.001 respectively). Conclusions: This study advocates for the use of (neo)adjuvant chemotherapy for stage 3 CRC in Nigeria. This is in line with current NCCN and ASCO guidelines for stage 3 CRC. Prospective studies would be needed to fully explore the role of combined treatment modality (surgery and chemotherapy) in the non-metastatic setting.