Multivariate Survival Models Research Articles

Sudden cardiac death after myocardial infarction: individual participant data from pooled cohorts.

Risk stratification of sudden cardiac death after myocardial infarction and prevention by defibrillator rely on left ventricular ejection fraction (LVEF). Improved risk stratification across the whole LVEF range is required for decision-making on defibrillator implantation. The analysis pooled 20 data sets with 140 204 post-myocardial infarction patients containing information on demographics, medical history, clinical characteristics, biomarkers, electrocardiography, echocardiography, and cardiac magnetic resonance imaging. Separate analyses were performed in patients (i) carrying a primary prevention cardioverter-defibrillator with LVEF ≤ 35% [implantable cardioverter-defibrillator (ICD) patients], (ii) without cardioverter-defibrillator with LVEF ≤ 35% (non-ICD patients ≤ 35%), and (iii) without cardioverter-defibrillator with LVEF > 35% (non-ICD patients >35%). Primary outcome was sudden cardiac death or, in defibrillator carriers, appropriate defibrillator therapy. Using a competing risk framework and systematic internal-external cross-validation, a model using LVEF only, a multivariable flexible parametric survival model, and a multivariable random forest survival model were developed and externally validated. Predictive performance was assessed by random effect meta-analysis. There were 1326 primary outcomes in 7543 ICD patients, 1193 in 25 058 non-ICD patients ≤35%, and 1567 in 107 603 non-ICD patients >35% during mean follow-up of 30.0, 46.5, and 57.6 months, respectively. In these three subgroups, LVEF poorly predicted sudden cardiac death (c-statistics between 0.50 and 0.56). Considering additional parameters did not improve calibration and discrimination, and model generalizability was poor. More accurate risk stratification for sudden cardiac death and identification of low-risk individuals with severely reduced LVEF or of high-risk individuals with preserved LVEF was not feasible, neither using LVEF nor using other predictors.

8455 Denosumab Is Associated With Decreased Mortality Compared To Zoledronic Acid In Diabetic Osteoporotic Patients: A Population-Based Cohort Study

Abstract Disclosure: V. Rouach: Consulting Fee; Self; Amgen Inc. H. Gortler: None. Y. Greenman: None. C. Gabriel: None. G. inbalL: None. Background: Anti-resorptive therapies are the mainstay of osteoporosis management, but evidence of their efficacy in the diabetic population is limited and comparison between treatments is lacking. A retrospective analysis, presented at the American Society for Bone and Mineral Research (ASBMR) 2023 Annual Meeting, suggests that denosumab leads to greater reduction in fracture risk than zoledronic acid among treatment-naive postmenopausal women with osteoporosis. However, a comparative study recently published suggests higher mortality with denosumab versus oral bisphosphonates. Aim: To assess the association between zoledronic acid or denosumab and the risk of major osteoporotic fracture and mortality in osteoporotic patients with diabetes type 2. Methods: The study population was identified by electronic records of a diabetes registry cross-linked with an osteoporosis registry of a large provider healthcare organization in Israel. Index date was at osteoporosis registry entry. Demographics, Comorbidity Index (CCI), diabetes complications, bone mineral density (BMD) T-scores, hemoglobin A1c levels, eGFR, purchase of statins, and anti-resorptive agents were collected. Propensity score matching was performed. Kaplan-Meier curves were generated to assess the time to outcomes. Multivariable Cox's proportional hazards survival model was performed. Results: A total of 27503 diabetic osteoporotic patients were identified, 13343 (48%) patients initiated treatment; 12214 (91.5%) started an oral bisphosphonate, 627 (4.7%) zoledronic acid and 502 (3.7%) denosumab. The median follow-up was 8.9 years. The denosumab treated patients were older (75.7 vs 71.9, p<0.01), had longer diabetes duration (8.4 vs 7.2, p<0.01), were more frequently treated with insulin (29.7 vs 23.9, p=0.02) and had a lower eGFR (59.4 vs 75.3, p<0.01). Male/Female ratio, BMI, CCI, smoking status, alcohol consumption, Hip BMD, HbA1c levels, microvascular complications, hypoglycemic events and statins prescriptions were similar. After propensity weighting, we observed a significant reduced risk of death among the denosumab treated patients (RR=0.72 [0.58-0.91]) without a significant difference in the risk of fracture (RR=0.9 [0.82-1.12]). Conclusions: In this cohort of diabetic osteoporotic patients, denosumab was associated with a significantly reduced mortality compared to zoledronic acid, without a significant difference in the risk of fractures. The effect of denosumab on mortality is yet to be explored. Presentation: 6/1/2024

9236 Denosumab Is Associated With Decreased Mortality Compared To Zoledronic Acid In Diabetic Osteoporotic Patients: A Population-Based Cohort Study

Abstract Disclosure: V. Rouach: Consulting Fee; Self; Amgen Inc. H. Gortler: None. Y. Greenman: None. C. Gabriel: None. G. inbalL: None. Background: Anti-resorptive therapies are the mainstay of osteoporosis management, but evidence of their efficacy in the diabetic population is limited and comparison between treatments is lacking. A retrospective analysis, presented at the American Society for Bone and Mineral Research (ASBMR) 2023 Annual Meeting, suggests that denosumab leads to greater reduction in fracture risk than zoledronic acid among treatment-naive postmenopausal women with osteoporosis. However, a comparative study recently published suggests higher mortality with denosumab versus oral bisphosphonates. Aim: To assess the association between zoledronic acid or denosumab and the risk of major osteoporotic fracture and mortality in osteoporotic patients with diabetes type 2. Methods: The study population was identified by electronic records of a diabetes registry cross-linked with an osteoporosis registry of a large provider healthcare organization in Israel. Index date was at osteoporosis registry entry. Demographics, Comorbidity Index (CCI), diabetes complications, bone mineral density (BMD) T-scores, hemoglobin A1c levels, eGFR, purchase of statins, and anti-resorptive agents were collected. Propensity score matching was performed. Kaplan-Meier curves were generated to assess the time to outcomes. Multivariable Cox's proportional hazards survival model was performed. Results: A total of 27503 diabetic osteoporotic patients were identified, 13343 (48%) patients initiated treatment; 12214 (91.5%) started an oral bisphosphonate, 627 (4.7%) zoledronic acid and 502 (3.7%) denosumab. The median follow-up was 8.9 years. The denosumab treated patients were older (75.7 vs 71.9, p<0.01), had longer diabetes duration (8.4 vs 7.2, p<0.01), were more frequently treated with insulin (29.7 vs 23.9, p=0.02) and had a lower eGFR (59.4 vs 75.3, p<0.01). Male/Female ratio, BMI, CCI, smoking status, alcohol consumption, Hip BMD, HbA1c levels, microvascular complications, hypoglycemic events and statins prescriptions were similar. After propensity weighting, we observed a significant reduced risk of death among the denosumab treated patients (RR=0.72 [0.58-0.91]) without a significant difference in the risk of fracture (RR=0.9 [0.82-1.12]). Conclusions: In this cohort of diabetic osteoporotic patients, denosumab was associated with a significantly reduced mortality compared to zoledronic acid, without a significant difference in the risk of fractures. The effect of denosumab on mortality is yet to be explored. Presentation: 6/1/2024

Health Insurance Continuity and Mortality in Children and Adolescents/Young Adults with Blood Cancer.

Many uninsured patients do not receive Medicaid coverage until a cancer diagnosis, potentially delaying access to care for early cancer detection and treatment. We examine the association of Medicaid enrollment timing and patterns with survival among children and adolescents/young adults (AYAs) diagnosed with blood cancers, where disease onset can be acute and early detection is critical. We identified 28,750 children and AYAs (0-39 years) newly diagnosed with blood cancers from the 2006-2013 SEER-Medicaid data. Enrollment patterns included continuous Medicaid (preceding through diagnosis), newly gained Medicaid (at/shortly after diagnosis), other noncontinuous Medicaid enrollment, and private/other insurance. We assessed cumulative incidence of death from diagnosis, censoring at last follow-up, five years post-diagnosis, or December 2018, whichever occurred first. Multivariable survival models estimated the association of insurance enrollment patterns with risk of death. One-fourth (26.1%) of the cohort were insured by Medicaid; of these, 41.1% had continuous Medicaid, 34.9% had newly gained Medicaid, and 24.0% had other noncontinuous enrollment. The cumulative incidence of all-cause death five-year post-diagnosis was highest in patients with newly gained Medicaid (30.2%, 95%CI = 28.4-31.9%), followed by other noncontinuous enrollment (23.2%, 95%CI = 21.3-25.2%), continuous Medicaid (20.5%, 95%CI = 19.1-21.9%), and private/other insurance (11.2%; 95%CI = 10.7-11.7%). In multivariable models, newly gained Medicaid was associated with a higher risk of all-cause (hazard ratio = 1.39, 95%CI = 1.27-1.53) and cancer-specific death (hazard ratio = 1.50, 95%CI = 1.35-1.68), compared to continuous Medicaid. Continuous Medicaid coverage is associated with survival benefits among pediatric and AYA patients diagnosed with blood cancers; however, less than half of Medicaid-insured patients have continuous coverage before diagnosis.

Immune Biomarkers at Birth Predict Lower Respiratory Tract Infection Risk in a Large Birth Cohort.

Lower respiratory tract infections (LRTIs) remain the leading cause of infant morbidity and mortality worldwide and affect long-term respiratory health. Identifying immunological determinants of LRTI susceptibility may help stratify disease risk and identify therapies. This study aimed to identify neonatal immunological factors predicting LRTI risk in infancy. Cord blood plasma from 191 neonates from the Boston Birth Cohort was analyzed for 28 soluble immune factors. LRTI was defined as bronchiolitis, bronchitis, or pneumonia during the first year of life. Welch's t-test demonstrated significantly higher log10 transformed concentrations of IL-17 and IFNγ in the LRTI group compared to neonates without LRTI in the first year of life (p < 0.05). Risk associations were determined using multivariate survival models. There were 29 infants with LRTIs. High cord blood levels of IFNγ (aHR = 2.35, 95% CI 1.07-5.17), TNF-β (aHR = 2.86, 95% CI 1.27-6.47), MIP-1α (aHR = 2.82, 95% CI 1.22-6.51), and MIP-1β (aHR = 2.34, 95% CI 1.05-5.20) were associated with a higher risk of LRTIs. RANTES was associated with a lower risk (aHR = 0.43, 95% CI 0.19-0.97). Soluble immune factors linked to antiviral immunity (IFNγ) and cytokines mediating inflammatory responses (TNF-β), and cell homing (MIP-1α/b), at birth were associated with an increased risk of LRTIs during infancy.

The Impact of Leg Length and Offset Change on Dislocation Risk Following Primary Total Hip Arthroplasty

BackgroundSoft tissue management in total hip arthroplasty includes appropriate restoration and/or alteration of leg length (LL) and offset to re-establish natural hip biomechanics. The purpose of this study was to evaluate the effect of LL and offset-derived variables in a multivariable survival model for dislocation. MethodsClinical, surgical, and radiographic data was retrospectively acquired for 12,582 patients undergoing primary total hip arthroplasty at a single institution from 1998 to 2018. There were twelve variables derived from preoperative and postoperative radiographs related to LL and offset that were measured using a validated automated algorithm. These measurements, as well as other modifiable and nonmodifiable surgical, clinical, and demographic factors, were used to determine hazard ratios for dislocation risk. ResultsNone of the LL or offset variables conferred significant risk or protective benefit for dislocation risk. By contrast, all other variables included in the multivariable model demonstrated a statistically significant effect on dislocation risk with a minimum effect size of 28% (range 0.72 to 1.54) (sex, surgical approach, acetabular liner type, femoral head size, neurologic disease, spine disease, and prior spine surgery). ConclusionsContrary to traditional teaching and our hypothesis, operative changes in LL and offset did not demonstrate any clinically or statistically significant effect in this large and well-characterized cohort. This does not imply that these variables are not important in individual cases, but rather suggests the overall impact of LL and offset changes is relatively minor for dislocation risk compared to other variables that were found to be highly clinically and statistically significant in this population. These results may also suggest that surgeons do a good job of restoring native LL and offset for patients, which may mitigate their analyzed impact.

Machine learning evaluation of immune infiltrate through digital tumour score allows prediction of survival outcome in a pooled analysis of three international stage III colon cancer cohorts

T-cell immune infiltrates are robust prognostic variables in localised colon cancer. Evaluation of prognosis using artificial intelligence is an emerging field. We evaluated whether machine learning analysis improved prediction of patient outcome in comparison with analysis of T cell infiltrate only or in association with clinical variables. We used data from two phase III clinical trials (Prodige-13 and PETACC08) and one retrospective Italian cohort (HARMONY). Cohorts were split into training (N=692), internal validation (N=297) and external validation (N=672) sets. Tumour slides were stained with CD3mAb. CD3 Machine Learning (CD3ML) score was computed using graphical parameters within the tumour tiles obtained from CD3 slides. CD3 infiltrates in tumour core and invasive margin were automatically detected. Associations of CD3 infiltrates and CD3ML with 5-year Disease-Free Survival (DFS) were examined using univariate and multivariable survival models by Cox regression. CD3 density both in the invasive margin and the tumour core were significantly associated with DFS in the different sets. Similarly, CD3ML score was significantly associated with DFS in all sets. CD3 assessment did not provide added value on top of CD3ML assessment (Likelihood Ratio Test (LRT), p=0.13). In contrast, CD3ML improved prediction of DFS when combined with a clinical risk stage (LRT, p=0.001). Stratified by clinical risk score (High or Low), patients with low CD3ML score had better DFS. In all tested sets, machine learning analysis of tumour cells improved prediction of prognosis compared to clinical parameters. Adding tumour-infiltrating lymphocytes assessment did not improve prognostic determination. This research received no external funding.

Automated digital oncologic data review for breast cancer research: AI-enabled mCODE and field-of-interest extraction framework.

e12647 Background: Enhancing outcomes in cancer care and research can be optimally achieved by accelerating data extraction from unstructured electronic health resources. Understanding the association between treatment response markers and long-term outcomes is crucial for refining cancer therapeutic approaches and improving patient prognosis. In this study, we explore the association of oncologic outcomes with pathologic complete response (pCR) in high-risk early-stage breast cancer treated with neoadjuvant chemotherapy (NACT) using automated oncologic data extraction framework. Methods: Patients with primary stage I-III breast cancer who received NACT followed by mastectomy, with or without immediate breast reconstruction from January 2010 to August 2023 were retrospectively analyzed. We extracted pCR results and data on clinical and pathological stage, tumor subtype, grade, and treatment regimens and factors affecting postoperative complications from electronic medical records based on the AI enabled-mCODE (Minimal Common Oncology Data Elements) and fields of interest extraction framework. Employing mixed-effects Cox models, we explored the association between pCR and event-free survival (EFS), distant recurrence free Survival (DRFS) and overall survival (OS). Survival curves of patients were estimated using Kaplan-Meier method. Multivariate analysis was performed to evaluate the predictive factors for pCR. Results: Overall, 2725 patients were analyzed. The median age was 50.6 years. The median follow-up was 67 months. pCR was achieved in 31.0%. Hazard ratio for pCR vs. non-pCR were 0.34 for EFS (95% CI, 0.25-0.45), 0.32 for DRFS (95% CI, 0.23-0.43), 0.21 for OS (95% CI, 0.13-0.34), varying from 0.13 to 0.56 for EFS, 0.12 to 0.48 for DRFS and 0.07 to 0.40 for OS. The presence of pCR remained prognostic for event-free survival in multivariate models adjusted for age, tumor size, and nodal status at baseline. Conclusions: This study confirms that patients achieving pCR after NACT have better long-term survival outcomes that patients who do not. Oncologic data extraction from unstructured electronic medical records using AI-enabled extraction framework can be expected to enhance robust cancer research and generate valuable insights from real-world practice settings for tailored therapeutic decision-making. [Table: see text]

Genomic analysis of recurrent early stage and de novo metastatic colorectal cancer among veterans.

e15593 Background: The Veterans Health Administration (VHA) diagnoses approximately 4,000 cases of colon cancer each year and conducts Next Generation Sequencing (NGS) on recurrent and metastatic cases. The molecular profile of cancers resected for cure that recur is poorly characterized and holds promise to clarify both the risk and mechanisms of recurrence. We sought to characterize these genetic differences. Methods: A national database from the VHA consisting of colon cancer patients was analyzed for those with recurrent disease who have received NGS testing from 2015-2021. Data was analyzed by stage and presence of DNA mismatch repair (MMR) (MLH1, MSH2, PMS2, MSH6) and DNA damage repair (DDR) abnormalities (BRCA1, BRCA2, ATM, BRIP1, BARD1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D) in addition to receipt of chemotherapy and overall survival with univariate and multivariate cox proportional hazards analyses and multivariate logistic regression analyses. Chi squared analyses were also performed comparing presence of DNA abnormalities between early recurrent and stage IV cancers. Results: From309,318 cases with colon cancer codes, we identified 2,366 colon cancer patients who had NGS and had recurrent or de novo metastatic disease. 1,072 patients had initial staging annotated (I n = 89, II n = 194, III n = 320 , IV n = 453). The average age was 69.41 years; 70.5% white, 21.5% Black and 7.8% other. Among all patients, DNA MMR abnormalities (n = 117) were associated with better overall survival, after controlling for age and race (HR 0.6, 95% CI 0.42-0.85, p &lt; 0.005). Among recurrent early stage (I-III) and stage IV patients, MMR (n = 55 and n = 21, respectively) is associated with improved survival in multivariable models (HR 0.39, 95% CI 0.33-0.46, p &lt; 0.001) . Among the 4 DNA MMR genes, MSH2 mutations had the strongest association with survival (univariate HR 0.31, 0.15-0.62, p &lt; 0.001). Among patients with MMR abnormalities, those who received either only capecitabine or only fluorouracil (without irinotecan or oxaliplatin) had worse overall survival (HR 2.6, 1.34 -5.04, p &lt; 0.005). This effect was more pronounced in early stage recurrent patients versus patients with de novo metastatic disease (HR 5.1, 1.87 – 13.9, p &lt; 0.001). Finally, we also found that early stage patients with recurrent disease are more likely to have MMR abnormalities compared to de novo metastatic (8.8% vs 4.6%, p &lt; 0.01) and confirmed this finding after controlling for race and age. (OR 2.13, p &lt; 0.005). Conclusions: Among Veteranswith colon cancer, resected recurrent colon cancer is more likely to have DNA MMR abnormalities than de novo metastatic disease. Veterans with recurrent and or metastatic disease with DNA MMR have better outcomes regardless of stage at diagnosis. These findings suggest different pathogenesis for those with recurrent disease after curative resection, and for those with de novo metastatic disease.

Machine learning identification of hypoxia-related genes and prognostic risk-scoring model for effective survival stratification in hepatocellular carcinoma.

e16261 Background: Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer-related mortality globally, where treatment and prognostic assessment have important implications in clinical practice. Hypoxia, as a common feature within solid tumors, can directly change the tumor microenvironment, which affects the efficacy of cancer treatment and prognosis. In this study, we constructed and validated a hypoxia-based prognostic model using bioinformatics and machine learning. Methods: Two public datasets, GSE14520 and GSE41666, were collected from the Gene Expression Omnibus: (1) HCC tumor tissues compared to adjacent normal tissues (N = 214) and (2) HepG2 cells under normoxic and hypoxic conditions (N = 6). Differential expression analysis was performed to identify HCC characteristic genes and hypoxia-related genes, including their common genes (HCC-Hypoxia Overlap genes, HHOs). Using RNA-seq data of HCC patients (N = 367) from the TCGA Liver Cancer (LIHC) database, univariate Cox regression models were identified, and the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm selected hypoxia-characteristic genes for the multivariate survival model. A hypoxia-related risk score was calculated based on the model of these characteristic genes and dichotomized cases into high-risk (HR) and low-risk (LR) groups. The model was validated using liver cancer cases (N = 232) from the International Cancer Genome Consortium database (ICGC-LIRI-JP). Results: Through differential expression analysis of the two datasets, we identified 52 HHOs. Univariate Cox analysis of these HHOs indicated that 21 genes were significantly associated with HCC patient survival. Through LASSO regression analysis, a total of 9 characteristic genes, including CENPA, KIF20A, DLGAP5, HMMR, UPB1, AFM, CABYR, PHLDA2, and N4BP2L1 were ultimately retained in the survival model. Based on these 9 genes, TCGA-LIHC samples were classified into HR and LR groups, and Kaplan-Meier (KM) analysis revealed significant differences in survival outcomes (p &lt; 0.032). Risk scoring of the ICGC-LIRI-JP validation set classified samples into HR and LR. KM analysis showed that the survival times of patients in the HR group were significantly shorter than those in the LR group (p &lt; 0.0001). Receiver Operating Characteristic Analysis analysis of the survival model showed area under the curve values of 0.815, 0.774, and 0.771 at 1, 2, and 3 years, respectively, demonstrating high performance in risk stratification. Conclusions: This study established a prognostic risk-scoring model based on 9 characteristic genes associated with hypoxia. This model can effectively stratify risks among HCC patients and demonstrate excellent performance in predicting survival. These findings may offer new biomarkers and therapeutic targets for the personalized treatment of HCC.

Donor Blood Tests do Not Predict Pancreas Graft Survival After Simultaneous Pancreas Kidney Transplantation; a National Cohort Study.

Simultaneous pancreas-kidney (SPK) transplantation improves quality of life and limits progression of diabetic complications. There is reluctance to accept pancreata from donors with abnormal blood tests, due to concern of inferior outcomes. We investigated whether donor amylase and liver blood tests (markers of visceral ischaemic injury) predict pancreas graft outcome using the UK Transplant Registry (2016-2021). 857 SPK recipients were included (619 following brainstem death, 238 following circulatory death). Peak donor amylase ranged from 8 to 3300U/L (median = 70), and this had no impact on pancreas graft survival when adjusting for multiple confounders (aHR = 0.944, 95% CI = 0.754-1.81). Peak alanine transaminases also did not influence pancreas graft survival in multivariable models (aHR = 0.967, 95% CI = 0.848-1.102). Restricted cubic splines were used to assess associations between donor blood tests and pancreas graft survival without assuming linear relationships; these confirmed neither amylase, nor transaminases, significantly impact pancreas transplant outcome. This is the largest, most statistically robust study evaluating donor blood tests and transplant outcome. Provided other factors are acceptable, pancreata from donors with mild or moderately raised amylase and transaminases can be accepted with confidence. The use of pancreas grafts from such donors is therefore a safe, immediate, and simple approach to expand the donor pool to reach increasing demands.

Sex-Specific Survival and Treatment Delay in Oropharyngeal Squamous Cell Carcinoma.

As the majority of oropharyngeal squamous cell carcinoma (OPSCC) is diagnosed in males, outcomes among females are not well-characterized. We identify sex-specific factors in OPSCC to refine female prognostication. Retrospective cohort. National Cancer Database (NCDB). OPSCC cases from the 2004 to 2019 NCDB were identified. Sociodemographic, clinical, and treatment characteristics (including timing between diagnosis and treatment administration) were compared between sexes. Multivariable Cox proportional hazard regression models were constructed to characterize survival in overall and female-only cohorts. Similar multivariable binomial logistic regression and survival models were constructed to assess odds of treatment delays and their effects on survival, respectively. A total of 192,973 OPSCC patients were identified; 36,695 (19%) were female. Females had more human papillomavirus (HPV) negative, lower clinical T and N stage, and higher comorbidity disease. Females experienced lower survival in HPV negative (hazard ratio, HR = 1.11, P < .001) but not HPV-positive disease. Females were more likely to have any treatment initiated over the median of 28 days (odds ratio, OR = 1.04, P = .014) or delays in adjuvant radiotherapy initiation over 6 weeks (OR = 1.11, P = .032). Treatment delay over 60 days (HR = 1.17, P = .016) and delay in adjuvant therapy initiation (HR = 1.24, P = .02) were associated with worse survival among females. In one of the largest analyses of OPSCC, females had poorer survival than males, specifically in HPV-negative disease, despite presentation with less advanced disease. Notably, delays in any treatment initiation and adjuvant radiotherapy initiation were more likely in HPV-negative women and associated with worse survival, highlighting potential systemic weaknesses contributing to poor prognosis among females.

Repeat corneal transplantation in Southern China: Indications, surgical technique, outcomes, and risk factors for repeat keratoplasty failure.

To report the indications, surgical techniques, and outcomes of repeat keratoplasty and evaluate the risk factors for graft failure in the Chinese population. The medical records of 216 patients (243 cases) who underwent at least two keratoplasties at a leading eye hospital in southern China between 2011 and 2020 were retrospectively reviewed. Indications and surgical procedures for repeat corneal transplantation were analyzed. Kaplan-Meier survival analysis was used to determine the graft survival rate after repeat keratoplasty. A multivariable survival model was used to assess the risk factors. Repeated keratoplasties increased continuously from 2011 to 2020 (P = 0.002). The most common primary indication was infectious keratitis (38.7%), and the most common reason for repeat keratoplasty was graft rejection (30.04%). Regraft techniques included penetrating keratoplasty (PK) in 165 cases (67.9%), deep lamellar keratoplasty (DALK) in 52 cases (21.40%), and endothelial keratoplasty (EK) in 26 cases (10.7%). Median survival was 5.3, 6.8, and 6.4 years for PK, DALK, and EK, respectively. The 5-year survival rate was 53.5%, 66.6%, and 69.8% for PK, DALK, and EK, respectively. The median LogMAR visual acuity was 1.4 for PK, 0.75 for DALK, and 1.2 for EK at the end of the follow-up. Multivariate analysis revealed that graft rejection is a risk factor for repeat keratoplasty failure (P = 0.002). DALK and EK may provide better outcomes than PK in treating graft failure. Preventing and treating postoperative graft rejection may be key to improving regraft survival. These findings will aid in the management of failed corneal grafts.

Symptom Burden and Time from Symptom Onset to Cancer Diagnosis in Patients with Early-Onset Colorectal Cancer: A Multicenter Retrospective Analysis.

Background: The incidence of colorectal cancer (CRC) is decreasing in individuals >50 years due to organised screening but has increased for younger individuals. We characterized symptoms and their timing before diagnosis in young individuals. Methods: We identified all patients diagnosed with CRC between 1990-2017 in British Columbia, Canada. Individuals <50 years (n = 2544, EoCRC) and a matched cohort >50 (n = 2570, LoCRC) underwent chart review to identify CRC related symptoms at diagnosis and determine time from symptom onset to diagnosis. Results: Across all stages of CRC, EoCRC presented with significantly more symptoms than LoCRC (Stage 1 mean ± SD: 1.3 ± 0.9 vs. 0.7 ± 0.9, p = 0.0008; Stage 4: 3.3 ± 1.5 vs. 2.3 ± 1.7, p < 0.0001). Greater symptom burden at diagnosis was associated with worse survival in both EoCRC (p < 0.0001) and LoCRC (p < 0.0001). When controlling for cancer stage, both age (HR 0.87, 95% CI 0.8-1.0, p = 0.008) and increasing symptom number were independently associated with worse survival in multivariate models. Conclusions: Patients with EoCRC present with a greater number of symptoms of longer duration than LoCRC; however, time from patient reported symptom onset was not associated with worse outcomes.

Risk of Major Adverse Cardiovascular Events After SARS-CoV-2 Infection in British Columbia: A Population-Based Study

BackgroundCOVID-19 is associated with increased risk of post-acute cardiovascular outcomes. Population-based evidence for long periods of observation is still limited. MethodsThis population-based cohort study was conducted using data (2020-2021) from the British Columbia COVID-19 Cohort. The exposure of interest was severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, identified through reverse transcription–polymerase chain reaction (RT-PCR) assay. Individuals who tested positive (exposed) on RT-PCR were matched to negative controls (unexposed) on sex, age, and RT-PCR collection date in a 1:4 ratio. Outcomes of interest were incident major adverse cardiovascular events and acute myocardial infarction, identified more than 30 days after RT-PCR collection date. The association between SARS-CoV-2 infection and cardiovascular risk was assessed through multivariable survival models. Population attributable fractions were computed from Cox models. ResultsWe included 649,320 individuals: 129,864 exposed and 519,456 unexposed. The median duration of follow-up was 260 days; 1,786 events (0.34%) took place among the unexposed, and 702 (0.54%) in the exposed. The risk of major adverse cardiovascular events was higher in the exposed (adjusted hazard ratio [aHR] 1.34; 95% confidence interval [CI], 1.22-1.46), with greater risk observed in those who were hospitalized (aHR 3.81; 95% CI, 3.12-4.65) or required intensive care unit admission (aHR 6.25; 95% CI, 4.59-8.52) compared with the unexposed group. The fraction of cardiovascular events attributable to SARS-CoV-2 was 7.04% (95% CI, 4.67-9.41%). Comparable results were observed for acute myocardial infarction. ConclusionsSARS-CoV-2 infection was associated with higher cardiovascular risk, with graded increase across the acute COVID-19 severity, contributing to 7% of incident major adverse cardiovascular events. These findings suggest that long-term monitoring of cardiovascular risk is required in COVID-19 survivors.

Cardiac MRI-Enriched Phenomapping Classification and Differential Treatment Outcomes in Patients With Ischemic Cardiomyopathy.

Significant controversy continues to confound patient selection and referral for revascularization and mitral valve intervention in patients with ischemic cardiomyopathy (ICM). Cardiac magnetic resonance (CMR) enables comprehensive phenotyping with gold-standard tissue characterization and volumetric/functional measures. Therefore, we sought to determine the impact of CMR-enriched phenomapping patients with ICM to identify differential outcomes following surgical revascularization and surgical mitral valve intervention (sMVi). Consecutive patients with ICM referred for CMR between 2002 and 2017 were evaluated. Latent class analysis was performed to identify phenotypes enriched by comprehensive CMR assessment. The primary end point was death, heart transplant, or left ventricular assist device implantation. A multivariable Cox survival model was developed to determine the association of phenogroups with overall survival. Subgroup analysis was performed to assess the presence of differential response to post-magnetic resonance imaging procedural interventions. A total of 787 patients were evaluated (63.0±11.2 years, 24.8% women), with 464 primary events. Subsequent surgical revascularization and sMVi occurred in 380 (48.3%) and 157 (19.9%) patients, respectively. Latent class analysis identified 3 distinct clusters of patients, which demonstrated significant differences in overall outcome (P<0.001). Latent class analysis identified differential survival benefit of revascularization in patients as well as patients who underwent revascularization with sMVi, based on phenogroup classification, with phenogroup 3 deriving the most survival benefit from revascularization and revascularization with sMVi (hazard ratio, 0.61 [0.43-0.88]; P=0.0081). CMR-enriched unsupervised phenomapping identified distinct phenogroups, which were associated with significant differential survival benefit following surgical revascularization and sMVi in patients with ICM. Phenomapping provides a novel approach for patient selection, which may enable personalized therapeutic decision-making for patients with ICM.

O172: Donor blood tests and pancreas outcome in simultaneous pancreas kidney transplantation, a national cohort study

Abstract Introduction Simultaneous pancreas-kidney (SPK) transplantation improves quality of life in people with diabetes and end-stage renal disease, and limits the progression of diabetes related complications. Many surgeons are reluctant to accept the pancreas of a donor with raised amylase, due to concern of potentially inferior outcomes. We aim to ascertain whether donor amylase and liver blood tests (a marker of visceral ischaemic injury) predict pancreas transplant outcome. Methods This retrospective cohort study used the NHS registry on adult SPK transplantation (2016-2021). Adjusted regressions models assessed the effect of donor amylase and liver blood tests on pancreas transplant outcome. Results 857 SPK recipients were included (619 following brainstem death, 238 following circulatory death). Peak donor amylase ranged from 8-3300U/L (median=70). Donor peak amylase had no significant impact on pancreas graft survival when adjusting for multiple confounders (aHR=0.944, 95% CI=0.754-1.81). Median peak alanine transaminase and aspartate transaminase was 67U/L and 72U/L (range 8-5930 and 0-7910U/L). Neither of these influenced pancreas graft survival in multivariable models (aHR=0.967, 95% CI=0.848-1.102 and aHR=0.908, 95% CI=0.771-1.070, respectively). Restricted cubic splines were used to assess the relationship between donor blood tests and pancreas graft survival without assuming linear relationships; these confirmed neither amylase, nor transaminases, significantly impact on pancreas transplant outcome. Conclusions Donor amylase and transaminases do not predict transplant outcomes. Therefore, raised donor amylase or transaminases should not be considered a barrier to organ utilisation. The use of pancreas grafts from donors with hyperamylasaemia is a safe, immediate, and simple approach to expand the donor pool.

The prognostic effect of tumor-associated macrophages in stage I-III colorectal cancer depends on T cell infiltration

BackgroundTumor-associated macrophages (TAMs) are associated with unfavorable patient prognosis in many cancer types. However, TAMs are a heterogeneous cell population and subsets have been shown to activate tumor-infiltrating T cells and confer a good patient prognosis. Data on the prognostic value of TAMs in colorectal cancer are conflicting. We investigated the prognostic effect of TAMs in relation to tumor-infiltrating T cells in colorectal cancers.MethodsThe TAM markers CD68 and CD163 were analyzed by multiplex fluorescence immunohistochemistry and digital image analysis on tissue microarrays of 1720 primary colorectal cancers. TAM density in the tumor stroma was scored in relation to T cell density (stromal CD3+ and epithelial CD8+ cells) and analyzed in Cox proportional hazards models of 5-year relapse-free survival. Multivariable survival models included clinicopathological factors, MSI status and BRAFV600E mutation status.ResultsHigh TAM density was associated with a favorable 5-year relapse-free survival in a multivariable model of patients with stage I–III tumors (p = 0.004, hazard ratio 0.94, 95% confidence interval 0.90–0.98). However, the prognostic effect was dependent on tumoral T-cell density. High TAM density was associated with a good prognosis in patients who also had high T-cell levels in their tumors, while high TAM density was associated with poorer prognosis in patients with low T-cell levels (pinteraction = 0.0006). This prognostic heterogeneity was found for microsatellite stable tumors separately.ConclusionsThis study supported a phenotypic heterogeneity of TAMs in colorectal cancer, and showed that combined tumor immunophenotyping of multiple immune cell types improved the prediction of patient prognosis.

A Longitudinal Analysis of Flavored Cigar Use and Cigar Smoking Cessation Among US Adults.

Flavorings in cigars increase their appeal, mask the harsh taste of tobacco, and may hinder successful cigar smoking cessation; however, limited evidence has examined whether flavors are associated with short- or long-term cigar smoking cessation. Using restricted data from the Population Assessment of Tobacco and Health Study Waves 1-5, we examined whether flavored cigar use was associated with 30-day-plus and 1-year-plus cigar smoking cessation among US adults. Multivariable discrete-time survival models were fit to a nationally representative sample of US adult (18+) respondents who had a current, established cigar use, smoked five or more days in the past 30 days, and did not exclusively smoke traditional premium cigars at baseline. Models adjusted for age, sex, race and ethnicity, income, cigar and cigarette smoking intensity, and blunt use. At baseline, 44.6% of respondents (n = 674) were 18-34 years old, 75.0% were male, 56.7% were non-Hispanic White, 78.9% had household incomes of <$50,000, and 56.2% smoked flavored cigars. In fully adjusted models, flavored cigar use was associated with a lower risk of 30-day-plus (HR = 0.76, 95% CI = 0.60, 0.97) but not 1-year-plus cigar smoking cessation (HR = 0.81, 95% = 0.62, 1.05). We found that flavored cigar use was associated with a lower risk of short-term but not long-term cigar smoking cessation. More work is needed to understand the dynamics of cigar smoking transitions, including initiation, cessation, and relapse, particularly in larger cohorts and among those who exclusively use cigars or dual-use cigars and cigarettes. As local and some state jurisdictions continue to adopt partial or complete bans of flavored cigar products and the United States Food and Drug Administration considers a national ban of all characterizing flavors in cigars, there is a need for more longitudinal work examining the associations between flavorings in cigars and short and long-term cigar-smoking behaviors, including but not limited to initiation, cessation, intensity of use, and relapse, particularly in diverse cohorts.

CONTRIBUTION OF CHRONIC CONDITIONS TO MORTALITY: DIFFERENCES BY RACE AND ETHNICITY

Abstract Health disparities exist in the timing, order, and accumulation of chronic somatic and neurodegenerative conditions, which are associated with increased health care utilization, expenditure, and mortality. We assessed the contribution of coexisting conditions to mortality among Hispanic, non-Hispanic (NH) White, and NH Black older adults in the United States to measure their burden. We used nationally-representative data from the Health and Retirement Study (HRS) from 1998-2016 (n=11,532). This analysis estimates the absolute additive contributions of seven somatic and two mental/neurodegenerative conditions using a longitudinal average attributable fraction derived from a survey-weighted multivariable discrete survival model adjusted for sociodemographic characteristics (sex, education, net worth, body-mass index, coupled, insured, and any of six ADLs lasting three or more months “because of physical, mental, emotional, or memory problems”). Conditions accounted for over 60% of mortality in each racial/ethnic group (64.0%, 62.2%, and 70.5% in Hispanic, NH White, and NH Black adults, respectively). Although we found that cardiovascular diseases (heart disease and hypertension) were the greatest contributor to morality in each racial/ethnic group, the impact of other conditions varied by race/ethnicity. Leading contributors to mortality were diabetes (12.0%) and dementia (7.2%) for Hispanic participants; cancer (13.0%) and lung disease (10.7%) for NH White participants; and diabetes (11.5%), lung disease (7.4%), and dementia (7.3%) in NH Black participants. This study provides insights into racial and ethnic differences in the contribution of chronic somatic and neurodegenerative conditions to mortality and may help target efforts for specific preventative care services to populations most impacted by them.