Restricted Mean Survival Time Research Articles

Optimizing adjuvant treatment options for patients with glioblastoma.

This study focused on minimizing the costs and toxic effects associated with unnecessary chemotherapy. We sought to optimize the adjuvant therapy strategy, choosing between radiotherapy (RT) and chemoradiotherapy (CRT), for patients based on their specific characteristics. This selection process utilized an innovative deep learning method. We trained six machine learning (ML) models to advise on the most suitable treatment for glioblastoma (GBM) patients. To assess the protective efficacy of these ML models, we employed various metrics: hazards ratio (HR), inverse probability treatment weighting (IPTW)-adjusted HR (HRa), the difference in restricted mean survival time (dRMST), and the number needed to treat (NNT). The Balanced Individual Treatment Effect for Survival data (BITES) model emerged as the most effective, demonstrating significant protective benefits (HR: 0.53, 95% CI, 0.48-0.60; IPTW-adjusted HR: 0.65, 95% CI, 0.55-0.78; dRMST: 7.92, 95% CI, 7.81-8.15; NNT: 1.67, 95% CI, 1.24-2.41). Patients whose treatment aligned with BITES recommendations exhibited notably better survival rates compared to those who received different treatments, both before and after IPTW adjustment. In the CRT-recommended group, a significant survival advantage was observed when choosing CRT over RT (p < 0.001). However, this was not the case in the RT-recommended group (p = 0.06). Males, older patients, and those whose tumor invasion is confined to the ventricular system were more frequently advised to undergo RT. Our study suggests that BITES can effectively identify GBM patients likely to benefit from CRT. These ML models show promise in transforming the complex heterogeneity of real-world clinical practice into precise, personalized treatment recommendations.

COVID-19 diagnostic testing and vaccinations among First Nations in Manitoba: A nations-based retrospective cohort study using linked administrative data, 2020-2021.

Differential access to healthcare has contributed to a higher burden of illness and mortality among First Nations compared to other people in Canada. Throughout the Coronavirus Disease 2019 (COVID-19) pandemic, First Nations organizations in Manitoba partnered with public health and Manitoba government officials to ensure First Nations had early, equitable and culturally safe access to COVID-19 diagnostic testing and vaccination. In this study, we examined whether prioritizing First Nations for vaccination was associated with faster uptake of COVID-19 vaccines among First Nations versus All Other Manitobans (AOM). In this retrospective cohort study, we used linked, whole-population administrative data from the Manitoba healthcare system (February 2020 to December 2021) to determine rates of COVID-19 diagnostic testing, infection, and vaccination, and used adjusted restricted mean survival time (RMST) models to test whether First Nations received their first and second vaccine doses more quickly than other Manitobans. The cohort comprised 114,816 First Nations (50.6% female) and 1,262,760 AOM (50.1% female). First Nations were younger (72.3% were age 0 to 39 years) compared to AOM (51% were age 0 to 39 years) and were overrepresented in the lowest 2 income quintiles (81.6% versus 35.6% for AOM). The 2 groups had a similar burden of comorbidities (65.8% of First Nations had none and 6.3% had 3 or more; 65.9% of AOM had none and 6.0% had 3 or more) and existing mental disorders (36.9% of First Nations were diagnosed with a mood/anxiety disorder, psychosis, personality disorder, or substance use disorder versus 35.2% of AOM). First Nations had crude infection rates of up to 17.20 (95% CI 17.15 to 17.24) COVID-19 infections/1,000 person-months compared with up to 6.24 (95% CI 6.16 to 6.32) infections/1,000 person-months among AOM. First Nations had crude diagnostic testing rates of up to 103.19 (95% CI 103.06 to 103.32) diagnostic COVID-19 tests/1,000 person-months compared with up to 61.52 (95% CI 61.47 to 61.57) tests/1,000 person-months among AOM. Prioritizing First Nations to receive vaccines was associated with faster vaccine uptake among First Nations versus other Manitobans. After adjusting for age, sex, income, region of residence, mental health conditions, and comorbidities, we found that First Nations residents received their first vaccine dose an average of 15.5 (95% CI 14.9 to 16.0) days sooner and their second dose 13.9 (95% CI 13.3 to 14.5) days sooner than other Manitobans in the same age group. The study was limited by the discontinuation of population-based COVID-19 testing and data collection in December 2021. As well, it would have been valuable to have contextual data on potential barriers to COVID-19 testing or vaccination, including, for example, information on social and structural barriers faced by Indigenous and other racialized people, or the distrust Indigenous people may have in governments due to historical harms. In this study, we observed that the partnered COVID-19 response between First Nations and the Manitoba government, which oversaw creation and enactment of policies prioritizing First Nations for vaccines, was associated with vaccine acceptance and quick uptake among First Nations. This approach may serve as a useful framework for future public health efforts in Manitoba and other jurisdictions across Canada.

Improved Disease-Free Survival With Adjuvant Chemotherapy After Nephroureterectomy for Upper Tract Urothelial Cancer: Final Results of the POUT Trial.

Clinical trials frequently include multiple end points that mature at different times. The initial report, typically based on the primary end point, may be published when key planned co-primary or secondary analyses are not yet available. Clinical Trial Updates provide an opportunity to disseminate additional results from studies, published in JCO or elsewhere, for which the primary end point has already been reported.POUT was a phase III, randomized, open-label trial, including 261 patients with muscle-invasive or lymph node-positive, nonmetastatic upper tract urothelial cancer (UTUC) randomly assigned after radical nephroureterectomy to platinum-based chemotherapy (132) or surveillance (129). Primary outcome analysis demonstrated that chemotherapy improved disease-free survival (DFS). At that time, the planned secondary outcome analysis of overall survival (OS) was immature. By February 2022, 50 and 67 DFS events had occurred in the chemotherapy and surveillance groups, respectively, at a median follow-up of 65 months. The 5-year DFS was 62% versus 45%, univariable hazard ratio (HR), 0.55 (95% CI, 0.38 to 0.80, P = .001). The restricted mean survival time (RMST) was 18 months longer (95% CI, 6 to 30) in the chemotherapy arm. There were 46 and 60 deaths in the chemotherapy and control arms, respectively. The 5-year OS was 66% versus 57%, with univariable HR, 0.68 (95% CI, 0.46 to 1.00, P = .049) and RMST difference 11 months (95% CI, 1 to 21). Treatment effects were consistent across chemotherapy regimens (carboplatin or cisplatin) and disease stage. Toxicities were similar to those previously reported, and there were no clinically relevant differences in quality of life between arms. In summary, although OS was not the primary outcome measure, the updated results add further support for the use of adjuvant chemotherapy in patients with UTUC, suggesting long-term benefits.

Abatacept in individuals at high risk of rheumatoid arthritis (APIPPRA): a randomised, double-blind, multicentre, parallel, placebo-controlled, phase 2b clinical trial

Individuals with serum antibodies to citrullinated protein antigens (ACPA), rheumatoid factor, and symptoms, such as inflammatory joint pain, are at high risk of developing rheumatoid arthritis. In the arthritis prevention in the pre-clinical phase of rheumatoid arthritis with abatacept (APIPPRA) trial, we aimed to evaluate the feasibility, efficacy, and acceptability of treating high risk individuals with the T-cell co-stimulation modulator abatacept. The APIPPRA study was a randomised, double-blind, multicentre, parallel, placebo-controlled, phase 2b clinical trial done in 28 hospital-based early arthritis clinics in the UK and three in the Netherlands. Participants (aged ≥18 years) at risk of rheumatoid arthritis positive for ACPA and rheumatoid factor with inflammatory joint pain were recruited. Exclusion criteria included previous episodes of clinical synovitis and previous use of corticosteroids or disease-modifying antirheumatic drugs. Participants were randomly assigned (1:1) using a computer-generated permuted block randomisation (block sizes of 2 and 4) stratified by sex, smoking, and country, to 125 mg abatacept subcutaneous injections weekly or placebo for 12 months, and then followed up for 12 months. Masking was achieved by providing four kits (identical in appearance and packaging) with pre-filled syringes with coded labels of abatacept or placebo every 3 months. The primary endpoint was the time to development of clinical synovitis in three or more joints or rheumatoid arthritis according to American College of Rheumatology and European Alliance of Associations for Rheumatology 2010 criteria, whichever was met first. Synovitis was confirmed by ultrasonography. Follow-up was completed on Jan 13, 2021. All participants meeting the intention-to-treat principle were included in the analysis. This trial was registered with EudraCT (2013-003413-18). Between Dec 22, 2014, and Jan 14, 2019, 280 individuals were evaluated for eligibility and, of 213 participants, 110 were randomly assigned to abatacept and 103 to placebo. During the treatment period, seven (6%) of 110 participants in the abatacept group and 30 (29%) of 103 participants in the placebo group met the primary endpoint. At 24 months, 27 (25%) of 110 participants in the abatacept group had progressed to rheumatoid arthritis, compared with 38 (37%) of 103 in the placebo group. The estimated proportion of participants remaining arthritis-free at 12 months was 92·8% (SE 2·6) in the abatacept group and 69·2% (4·7) in the placebo group. Kaplan-Meier arthritis-free survival plots over 24 months favoured abatacept (log-rank test p=0·044). The difference in restricted mean survival time between groups was 53 days (95% CI 28-78; p<0·0001) at 12 months and 99 days (95% CI 38-161; p=0·0016) at 24 months in favour of abatacept. During treatment, abatacept was associated with improvements in pain scores, functional wellbeing, and quality-of-life measurements, as well as low scores of subclinical synovitis by ultrasonography, compared with placebo. However, the effects were not sustained at 24 months. Seven serious adverse events occurred in the abatacept group and 11 in the placebo group, including one death in each group deemed unrelated to treatment. Therapeutic intervention during the at-risk phase of rheumatoid arthritis is feasible, with acceptable safety profiles. T-cell co-stimulation modulation with abatacept for 12 months reduces progression to rheumatoid arthritis, with evidence of sustained efficacy beyond the treatment period, and with no new safety signals. Bristol Myers Squibb.

Identifying optimal ALK inhibitors in first- and second-line treatment of patients with advanced ALK-positive non-small-cell lung cancer: a systematic review and network meta-analysis

ObjectivesTo compare the efficacy, safety and effects on quality of life of different ALK-inhibitors for global and Asian patients with advanced ALK-positive non-small-cell lung cancer (NSCLC).MethodsThe included RCTs were identified through a systematic search of PubMed, EMBASE, Cochrane Library, Clinical Trials.gov, and major cancer conferences. The assessment of progression-free survival (PFS), intracranial PFS, overall survival (OS), and patient-reported outcomes (PROs) was carried out using restricted mean survival time (RMST) model, fractional polynomial model and Royston-Parmar model. Time-invariant hazard ratio (HR) models were also used to validate and supplement the primary analysis. Objective response rate (ORR) and adverse events with any grade, grade 3–5 were assessed through a Bayesian network meta-analysis. The primary measures for OS, PFS, and PROs were HR and RMST. The odds ratio was the metric for evaluating safety, ORR, 12-month PFS rate, 24-month OS rate, and the 12-month non-deterioration rate of PROs. Subgroup analyses based on patient characteristics were performed.ResultsA total of fourteen studies (ten for first-line, four for second-line) consisting of nine treatments (chemotherapy, crizotinib, alectinib [600mg BID], low-dose alectinib [300mg BID], brigatinib, ceritinib, ensartinib, envonalkib, and lorlatinib) were included. In the first-line setting, alectinib showed a significant advantage over crizotinib and had the longest OS among all ALK-inhibitors. Compared to crizotinib, lorlatinib had the best efficacy regarding PFS for global patients, followed closely by alectinib and brigatinib. For Asian patients, alectinib significantly improved PFS compared to other treatments. In second-line, alectinib had the highest PFS for patients pretreated with crizotinib, followed by brigatinib, ceritinib and chemotherapy. Alectinib, irrespective of the dose, was the safest first-line option, whereas lorlatinib, brigatinib, and ceritinib showed poorer safety profiles. Alectinib was also the safest ALK-inhibitor for crizotinib-resistant patients. Brigatinib had the best performance in terms of PROs.ConclusionsConsidering both efficacy and safety, alectinib appears to be the preferable treatment in first-line and second-line, particularly for Asian patients.

Impact of pulmonary complications following esophagectomy on long-term survival: multivariate meta-analysis and restricted mean survival time assessment.

Pulmonary complications (PC) are common after esophagectomy and their impact on long-term survival is not defined yet. The present study aimed to assess the effect of postoperative PCs on long-term survival after esophagectomy for cancer. Systematic review of the literature through February 1, 2023, was performed. The included studies evaluated the effect of PC on long-term survival. Primary outcome was long-term overall survival (OS). Cancer-specific survival (CSS) and disease-free survival (DFS) were secondary outcomes. Restricted mean survival time difference (RMSTD), hazard ratio (HR), and 95% confidence intervals (CI) were used as pooled effect size measures. Eleven studies were included (3423 patients). Overall, 674 (19.7%) patients developed PC. The RMSTD analysis shows that at 60-month follow-up, patients not experiencing PC live an average of 8.5 (95% CI 6.2-10.8; p < 0.001) months longer compared with those with PC. Similarly, patients not experiencing postoperative PC seem to have significantly longer CSS (8months; 95% CI 3.7-12.3; p < 0.001) and DFS (5.4months; 95% CI 1.6-9.1; p = 0.005). The time-dependent HRs analysis shows a reduced mortality hazard in patients without PC at 12 (HR 0.6, 95% CI 0.51-0.69), 24 (HR 0.64, 95% CI 0.55-0.73), 36 (HR 0.67, 95% CI 0.55-0.79), and 60months (HR 0.69, 95% CI 0.51-0.89). This study suggests a moderate clinical impact of PC on long-term OS, CSS, and DFS after esophagectomy. Patients not experiencing PC seem to have a significantly reduced mortality hazard up to 5years of follow-up.

Comparing Survival Extrapolation within All-Cause and Relative Survival Frameworks by Standard Parametric Models and Flexible Parametric Spline Models Using the Swedish Cancer Registry.

In health technology assessment, restricted mean survival time and life expectancy are commonly evaluated. Parametric models are typically used for extrapolation. Spline models using a relative survival framework have been shown to estimate life expectancy of cancer patients more reliably; however, more research is needed to assess spline models using an all-cause survival framework and standard parametric models using a relative survival framework. To assess survival extrapolation using standard parametric models and spline models within relative survival and all-cause survival frameworks. From the Swedish Cancer Registry, we identified patients diagnosed with 5 types of cancer (colon, breast, melanoma, prostate, and chronic myeloid leukemia) between 1981 and 1990 with follow-up until 2020. Patients were categorized into 15 cancer cohorts by cancer and age group (18-59, 60-69, and 70-99 y). We right-censored the follow-up at 2, 3, 5, and 10 y and fitted the parametric models within an all-cause and a relative survival framework to extrapolate to 10 y and lifetime in comparison with the observed Kaplan-Meier survival estimates. All cohorts were modeled with 6 standard parametric models (exponential, Weibull, Gompertz, log-logistic, log-normal, and generalized gamma) and 3 spline models (on hazard, odds, and normal scales). For predicting 10-y survival, spline models generally performed better than standard parametric models. However, using an all-cause or a relative survival framework did not show any distinct difference. For lifetime survival, extrapolating from a relative survival framework agreed better with the observed survival, particularly using spline models. For extrapolation to 10 y, we recommend spline models. For extrapolation to lifetime, we suggest extrapolating in a relative survival framework, especially using spline models. For survival extrapolation to 10 y, spline models generally performed better than standard parametric models did. However, using an all-cause or a relative survival framework showed no distinct difference under the same parametric model.Survival extrapolation to lifetime within a relative survival framework agreed well with the observed data, especially using spline models.Extrapolating parametric models within an all-cause survival framework may overestimate survival proportions at lifetime; models for the relative survival approach may underestimate instead.

Reoperation-free survival after hip screws or hip arthroplasty for undisplaced femoral neck fractures in the elderly.

Our primary aim was to assess reoperation-free survival at one year after the index injury in patients aged ≥ 75 years treated with internal fixation (IF) or arthroplasty for undisplaced femoral neck fractures (uFNFs). Secondary outcomes were reoperations and mortality analyzed separately. We retrieved data on all patients aged ≥ 75 years with an uFNF registered in the Swedish Fracture Register from 2011 to 2018. The database was linked to the Swedish Arthroplasty Register and the National Patient Register to obtain information on comorbidity, mortality, and reoperations. Our primary outcome, reoperation, or death at one year was analyzed using restricted mean survival time, which gives the mean time to either event for each group separately. Overall, 3,909 patients presenting with uFNFs were included. Of these patients, 3,604 were treated with IF and 305 with primary arthroplasty. There were no relevant differences in age, sex, or comorbidities between groups. In the IF group 58% received cannulated screws and 39% hook pins. In the arthroplasty group 81% were treated with hemiarthroplasty and 19% with total hip arthroplasty. At one year, 32% were dead or had been reoperated in both groups. The reoperation-free survival time over one year of follow-up was 288 days (95% confidence interval (CI) 284 to 292) in the IF group and 279 days (95% CI 264 to 295) in the arthroplasty group, with p = 0.305 for the difference. Mortality was 26% in the IF group and 31% in the arthroplasty group at one year. Reoperation rates were 7.1% in the IF group and 2.3% in the arthroplasty group. In older patients with a uFNF, reoperation-free survival at one year seems similar, regardless of whether IF or arthroplasty is the primary surgery. However, this comparison depends on the choice of follow-up time in that reoperations were more common after IF. In contrast, we found more early deaths after arthroplasty. Our study calls for a randomized trial comparing these two methods.

Fertility-sparing surgery vs standard surgery for early-stage cervical cancer: difference in 5-year life expectancy by tumor size

Cervical cancer incidence among premenopausal women is rising, and fertility-sparing surgery serves as an important option for this young population. There is a lack of evidence on what tumor size cutoff should be used to define candidacy for fertility-sparing surgery. We sought to describe how the association between fertility-sparing surgery (compared with standard surgery) and life expectancy varies by tumor size among patients with cervical cancers measuring ≤4 cm in largest diameter. Our secondary objective was to quantify the probability of undergoing adjuvant radiotherapy among patients who underwent fertility-sparing surgery as a function of tumor size. We identified patients in the National Cancer Database aged ≤45 years, diagnosed with stage I cervical cancer with tumors ≤4 cm between 2006 and 2018, who received no preoperative radiation or chemotherapy, and who underwent either fertility-sparing surgery (cone or trachelectomy, either simple or radical) or standard surgery (simple or radical hysterectomy) as their primary treatment. Propensity-score matching was performed to compare patients who underwent fertility-sparing surgery with those who underwent standard surgery. A flexible parametric model was employed to quantify the difference in life expectancy within 5 years of diagnosis (restricted mean survival time) based on tumor size among patients who underwent fertility-sparing and those who underwent standard surgery. In addition, among those who underwent fertility-sparing surgery, a logistic regression model was used to explore the relationship between tumor size and the probability of receiving adjuvant radiation. A total of 11,946 patients met the inclusion criteria of whom 904 (7.6%) underwent fertility-sparing surgery. After propensity-score matching, 897 patients who underwent fertility-sparing surgery were matched 1:1 with those who underwent standard surgery. Although the 5-year life expectancy was similar among patients who had fertility sparing surgery and those who had standard surgery regardless of tumor sizes, the estimates of life-expectancy differences associated with fertility-sparing surgery were more precise among patients with smaller tumors (1-cm tumor: restricted mean survival time difference,-0.10 months; 95% confidence interval,-0.67 to 0.47) than among those with larger tumors (4-cm tumor: restricted mean survival time difference,-0.11 months; 95% confidence interval,-3.79 to 3.57). The probability of receiving adjuvant radiation increased with tumor size, ranging from 5.6% (95% confidence interval, 3.9-7.9) for a 1-cm tumor to 37% (95% confidence interval, 24.3-51.8) for a 4-cm tumor. Within 5 years of diagnosis, young patients with stage I cancers measuring ≤4 cm had similar survival outcomes after either fertility-sparing surgery or standard surgery. However, because few patients with tumors >2 cm underwent fertility-sparing surgery, a clinically important survival difference could not be excluded in this population.

Abstract TMP104: Timing of Recurrence After Ischemic Stroke by Subtype: A Population-Based Study

Introduction: Prior work suggests stroke recurrence timing may vary by subtype, with large vessel disease being associated with earlier recurrence; however, these studies were conducted prior to the current era of secondary stroke prevention strategies. We examined whether the timing of recurrence after ischemic stroke varied by subtype in a large population-based study. Methods: In 2015, we identified all acute ischemic stroke events from hospitals in the Greater Cincinnati and Northern Kentucky area. Events in the last 6 months of 2014 were oversampled for black patients. Trained nurses extracted clinical data, and stroke specialists subtyped and adjudicated cases. We tracked events for 3 years, considering only ischemic strokes after day 14 as a recurrence. To assess recurrence timing variations, we stratified strokes by subtype and analyzed their restricted mean survival times (RMST) at 1, 3, 12, 24, and 36 months. RMST is the area under the survival curve up to a specific follow up point, denoting the average event-free time during that interval. Results: Of 2577 patients with index stroke, 332 recurrent strokes were seen over the ensuing 3 years. Half of the recurrences occurred within 8 months from the index event. At 12 months, large vessel (LV), undetermined, and other identified cause showed a statistically significant smaller RMST compared to small vessel (SV). This difference was nonsignificant at later time points, suggesting later recurrence for SV. There were no statistically significant differences in RMST between cardioembolic, LV, undetermined, and other identified at any time point. Conclusions: Recurrence after initial SV stroke was less likely to occur in the first year when compared to LV, undetermined etiology, or other identified causes of stroke. There was no evidence of difference in recurrence timing between cardioembolic, LV, other identified, and undetermined etiology.

Real world evidence comparison of first-line (1L) immune-oncology(IO)/tyrosine kinase inhibitor (TKI) vs. IO/IO combination therapy in renal cell carcinoma (RCC).

402 Background: IO/IO and IO/TKI are the preferred first-line combinations for treatment of RCC. While these combinations have been superior to sunitinib, no head-to-head clinical trials have been completed, which leaves uncertainty about the preferred combination. Here, we have leveraged ASCO Cancerlinq data to compare overall survival for RCC patients (pts) receiving first line IO/IO vs. IO/TKI combinations. We hypothesized that IO/IO and IO/TKI would have similar survival. Methods: We performed a retrospective cohort study comparing IO/IO vs IO/TKI therapy for pts with metastatic RCC. We used the Atropos health platform, which is a commercial platform and queried ASCO CancerLinq data for this study. We identified pts with kidney cancer who received IO/IO or IO/TKI combinations between 2019 and 2023. For our primary analysis, we included only pts with ICD10 code consistent with clear cell [cc]RCC, but also did a secondary analysis including all pts with kidney cancer. The IO/TKI pts received either avelumab/axitinib, pembrolizumab/axitinib, nivolumab/cabozantinib, pembrolizumab/lenvatinib, or atezolizumab/cabozantinib within two weeks of each other. The IO/IO pts were identified as pts receiving ipilimumab and nivolumab within 2 weeks of each other without interceding TKI. We conducted a survival analysis of time to death using Cox proportional hazards regression to compare the two groups. The analysis was performed under three different confounder adjustment scenarios: without adjustment, with basic matching on sex and age, and with high dimensional propensity score (hdPS) using inverse probability of treatment weighting. Results: A total of 584 pts (286 IO/IO and 296 IO/TKI) with kidney cancer and 146 with ccRCC (75 IO/IO and 71 IO/TKI) were included. Mean age was 63-65 years and 30-35% female sex across groups. Table shows the restricted mean survival time (RMST) and HR for the Cox models. In the ccRCC cohort, 20 pts with IO/IO and 7 with IO/TKI died with RMST of 947 and 1071 days, respectively (hdPS HR 0.35, p=0.03). In the RCC cohort, 76 pts with IO/IO vs 49 with IO/TKI died with RMST 1126 vs 1017 (hdPS HR 0.61, p=0.04). Conclusions: In this exploratory, hypothesis-generating analysis of CancerLinq data, pts treated with IO/TKI had longer OS vs those treated with IO/IO. Limitations include the retrospective nature of the study, low power, lack of randomization, and potential unmeasured confounding factors (e.g., confounding by indication). Further investigation and prospective validation are needed to confirm these findings. [Table: see text]

Assessing the Performance of Alternative Methods for Estimating Long-Term Survival Benefit of Immuno-oncology Therapies

ObjectivesThis study aimed to determine the accuracy and consistency of established methods of extrapolating mean survival for immuno-oncology (IO) therapies, the extent of any systematic biases in estimating long-term clinical benefit, what influences the magnitude of any bias, and the potential implications for health technology assessment. MethodsA targeted literature search was conducted to identify published long-term follow-up from clinical trials of immune-checkpoint inhibitors. Earlier published results were identified and Kaplan-Meier estimates for short- and long-term follow-up were digitized and converted to pseudo–individual patient data using an established algorithm. Six standard parametric, 5 flexible parametric, and 2 mixture-cure models (MCMs) were used to extrapolate long-term survival. Mean and restricted mean survival time (RMST) were estimated and compared between short- and long-term follow-up. ResultsPredicted RMST from extrapolation of early data underestimated observed RMST in long-term follow-up for 184 of 271 extrapolations. All models except the MCMs frequently underestimated observed RMST. Mean survival estimates increased with longer follow-up in 196 of 270 extrapolations. The increase exceeded 20% in 122 extrapolations. Log-logistic and log-normal models showed the smallest change with additional follow-up. MCM performance varied substantially with functional form. ConclusionsStandard and flexible parametric models frequently underestimate mean survival for IO treatments. Log-logistic and log-normal models may be the most pragmatic and parsimonious solutions for estimating IO mean survival from immature data. Flexible parametric models may be preferred when the data used in health technology assessment are more mature. MCMs fitted to immature data produce unreliable results and are not recommended.

Quantified treatment effect at the individual level is more indicative for personalized radical prostatectomy recommendation: implications for prostate cancer treatment using deep learning

BackgroundThere are potential uncertainties and overtreatment existing in radical prostatectomy (RP) for prostate cancer (PCa) patients, thus identifying optimal candidates is quite important.PurposeThis study aims to establish a novel causal inference deep learning (DL) model to discern whether a patient can benefit more from RP and to identify heterogeneity in treatment responses among PCa patients.MethodsWe introduce the Self-Normalizing Balanced individual treatment effect for survival data (SNB). Six models were trained to make individualized treatment recommendations for PCa patients. Inverse probability treatment weighting (IPTW) was used to avoid treatment selection bias.Results35,236 patients were included. Patients whose actual treatment was consistent with SNB recommendations had better survival outcomes than those who were inconsistent (multivariate hazard ratio (HR): 0.76, 95% confidence interval (CI), 0.64–0.92; IPTW-adjusted HR: 0.77, 95% CI, 0.61–0.95; risk difference (RD): 3.80, 95% CI, 2.48–5.11; IPTW-adjusted RD: 2.17, 95% CI, 0.92–3.35; the difference in restricted mean survival time (dRMST): 3.81, 95% CI, 2.66–4.85; IPTW-adjusted dRMST: 3.23, 95% CI, 2.06–4.45). Keeping other covariates unchanged, patients with 1 ng/mL increase in PSA levels received RP caused 1.77 months increase in the time to 90% mortality, and the similar results could be found in age, Gleason score, tumor size, TNM stages, and metastasis status.ConclusionsOur highly interpretable and reliable DL model (SNB) may identify patients with PCa who could benefit from RP, outperforming other models and clinical guidelines. Additionally, the DL-based treatment guidelines obtained can provide priori evidence for subsequent studies.

Decision To Delivery Time and Its Predictors Among Mothers Who Underwent Emergency Cesarean Delivery At Selected Hospitals of Northwest Ethiopia, 2023: Prospective Cohort Study.

The decision to delivery time is the interval between the decision and the childbirth by emergency caesarean delivery. The Royal College of Obstetricians and Gynecologists and the American College of Obstetricians and Gynecologists recommend that the decision to delivery time interval is less than 30 min. Additionally, the decision to delivery time varies across institutions and countries. The aim of this study was to determine the decision to delivery time and its predictors among women who underwent emergency cesarean delivery at selected hospitals of Northwest Ethiopia, 2023. An institutional-based prospective cohort study was conducted at selected hospitals of Northwest Ethiopia, among women who underwent emergency cesarean delivery from November 1 to January 30, 2023. A total of 285 participants were enrolled, and data collected using structured and pre-tested questionnaires. A systematic sampling technique was used. Data were entered into Epi-Data version 4.6 and then exported to STATA 15 for further analysis. The log rank test was utilized to compare group differences. The time is estimated by using the Kaplan-Meier curve and Cox proportional-hazard regression analysis was carried out to determine the predictors. From 285 participants, 56 (21.8%) women delivered within the recommended 30 min. The overall median survival time was undetermined and the restricted mean survival time was 48.9 min (95% CI: 47.4-50.5). The average decision to delivery time is affected among women who hesitate to accept consent (AHR: 0.17, 95% CI: 0.02-1.25), cord prolapses (AHR: 1.36, 95% CI: 0.46-3.94), rank of surgeon (AHR: 0.42. 95% CI: 0.42-1.08), no free operation room table (AHR: 0.27, 95% CI: 0.28-0.94), regional anesthesia (AHR: 0.56, 95% CI: 0.25-1.28), and use of a bladder flap (AHR: 0.33, 95% CI: 0.16-0.85). Overall decision to delivery times among women who underwent emergency cesarean section at selected hospitals were longer than the recommended time.

Transvaginal Polytetrafluoroethylene Mesh Surgery for Pelvic Organ Prolapse: One-Year Safety and Efficacy Results

ObjectivesTo evaluate the efficacy and safety of the polytetrafluoroethylene (PTFE) mesh by comparing conventionally used polypropylene (PP) mesh in tension-free vaginal mesh (TVM) surgery for pelvic organ prolapse (POP). MethodsWe conducted an observational cohort study of patients who underwent TVM using a PTFE or PP mesh. PTFE was used from June 2019 to May 2021, and PP mesh from January 2018 to May 2019. Outcomes included POP recurrence, perioperative complications, and patient satisfaction. Restricted mean survival time (RMST) was used to analyze POP recurrence, comparing the time to recurrence between the two groups at 1 year after TVM. ResultsOf 171 patients, 104 underwent PP mesh placement (PP group) and 67 underwent PTFE mesh placement (PTFE group). POP recurrence was observed in 10 and nine patients in the PP and PTFE groups, respectively. The mean time until the recurrence in the PTFE group was significantly shorter than that in the PP group (RMST difference: -20.3 days; 95% CI, -40.1 to -0.5; P=0.044). Subgroup analysis revealed the mean time until recurrence was significantly shorter in the PTFE group for postoperative periods 3 months or less, ages >70 yrs, and POP stage ≥3. There were no intervention cases in either group and no significant differences in the perioperative complications. Patient satisfaction was greater in the PTFE group after three months postoperatively. ConclusionTVM surgery with a PTFE mesh is more prone to recurrence than that with a PP mesh, but with higher patient satisfaction. Within three months of surgery, elderly patients and those with advanced stage POP require care to prevent recurrence.

Survival outcomes for rural patients with advanced prostate cancer: A SEER investigation.

22 Background: It is unknown if the impact of rural versus urban settings may impact outcomes in men with metastatic prostate cancer, and in particular survival. This study aimed to examine differences in survival for patients in the United States with de novo metastatic prostate cancer, according to urban-rural status. Methods: This retrospective cohort study utilised data from the publicly available Surveillance, Epidemiology, and End Results database. Data on demographics, urban-rural status, histopathology, and survival were extracted for men aged 18-75 diagnosed with metastatic prostate cancer between 2009-2018. Rurality was re-classified from the rural-urban continuum code which ranged from 1 (most urban) to 9 (most rural). Individuals missing rurality status or survival-related data were excluded. Overall and cancer-specific survival between individuals living in urban and rural settings were analysed using multivariable Cox regression and adjusted restricted mean survival time modelling. Subgroup analyses were performed for histological subtypes of prostate cancer. Sensitivity analyses were performed using varying definitions of rurality. Results: Altogether, 21,291 individuals were included, with 2,510 living in rural settings, defined as rural-urban continuum code ≥ 4. The rural and urban participants differed in age, race, U.S. region, and marital status. There was no association between urban-rural status and overall survival (adjusted hazard ratio [HR] = 1.03, 95% confidence interval [CI]: 0.97-1.09, 11,333 events) or cancer-specific survival (1.03, 0.97-1.10, 9,380 events) using Cox regression models. Adjusted restricted mean survival time modelling demonstrated that urban patients lived 2.29 months longer than rural patients (95% CI: 0.61-3.97). Sub-analyses of neuroendocrine, intraductal, and other histological subtypes did not show an association between urban-rural status and overall survival. A more selective definition of rurality (rural-urban continuum code ≥ 5) resulted in a persisting difference in overall mean survival time (2.12 months, 0.24-4.01) using restricted mean survival time modelling. Similarly, a broader definition of rurality (rural-urban continuum code ≥ 3) also resulted in a difference in overall mean survival time (1.98 months, 0.59-3.36). Conclusions: This retrospective analysis demonstrated that a U.S. individual with metastatic prostate cancer who resided rurally died sooner compared to those from urban areas, suggesting greater disease impact in rural settings.

Deep learning-guided adjuvant chemotherapy selection for elderly patients with breast cancer.

The efficacy of adjuvant chemotherapy in elderly breast cancer patients is currently controversial. This study aims to provide personalized adjuvant chemotherapy recommendations using deep learning (DL). Six models with various causal inference approaches were trained to make individualized chemotherapy recommendations. Patients who received actual treatment recommended by DL models were compared with those who did not. Inverse probability treatment weighting (IPTW) was used to reduce bias. Linear regression, IPTW-adjusted risk difference (RD), and SurvSHAP(t) were used to interpret the best model. A total of 5352 elderly breast cancer patients were included. The median (interquartile range) follow-up time was 52 (30-80) months. Among all models, the balanced individual treatment effect for survival data (BITES) performed best. Treatment according to following BITES recommendations was associated with survival benefit, with a multivariate hazard ratio (HR) of 0.78 (95% confidence interval (CI): 0.64-0.94), IPTW-adjusted HR of 0.74 (95% CI: 0.59-0.93), RD of 12.40% (95% CI: 8.01-16.90%), IPTW-adjusted RD of 11.50% (95% CI: 7.16-15.80%), difference in restricted mean survival time (dRMST) of 12.44 (95% CI: 8.28-16.60) months, IPTW-adjusted dRMST of 7.81 (95% CI: 2.93-11.93) months, and p value of the IPTW-adjusted Log-rank test of 0.033. By interpreting BITES, the debiased impact of patient characteristics on adjuvant chemotherapy was quantified, which mainly included breast cancer subtype, tumor size, number of positive lymph nodes, TNM stages, histological grades, and surgical type. Our results emphasize the potential of DL models in guiding adjuvant chemotherapy decisions for elderly breast cancer patients.

P752 Difficult-to-treat inflammatory bowel disease: effectiveness and safety of 4th and 5th lines

Abstract Background Many patients with inflammatory bowel disease (IBD) have signs or symptoms of active disease despite multiple treatment attempts. This emerging concept is defined difficult-to-treat IBD. The objective of this study was to investigate for the first time treatment persistence, efficacy and safety of biologics or small molecules used in 4th or 5th line therapy. Methods We reviewed all consecutive patients with IBD treated at Nancy University Hospital between July 2022 and April 2023 with the 4th or 5th line treatment for at least three months. The primary outcome was to assess the persistence rate of 4th and 5th line therapy. Results We enrolled 82 patients with IBD (4th line: 44; 5th line: 38). On Kaplan-Meier analysis the duration of risankizumab, ustekinumab or vedolizumab therapy did not differ significantly (p&amp;gt;0.05) as 4th and 5th line treatment. The restricted mean survival time analysis showed that persistence rate of risankizumab was the highest as 4th line therapy (risankizumab vs. vedolizumab: 36.0 and 29.4 weeks respectively, p=0.008; risankizumab vs. ustekinumab: 36.0 and 32.8 weeks respectively, p=0.035). In multivariate regression, Crohn’s disease diagnosis (Odd ratio 4.6; 95% confidence interval 1.7-12.4) was significantly associated with treatment persistence. Conclusion In this first real-world setting, risankizumab could have longer persistence rate as 4th line treatment for IBD than other agents. Persistence of biological agents was greater in Crohn’s disease than in ulcerative colitis. More studies are needed to compare treatment efficacy in patients with difficult-to-treat IBD.

Add-on effects of oral tocopherol supplementation to surgical varicocelectomy on the outcome of assisted reproductive technology: a single-center pilot study report.

Varicocelectomy is well known to improve the pregnancy outcome of patients with clinical varicoceles in assisted reproductive technologies as well as spontaneous conception. Therefore, this study aimed to evaluate the additional effects of oral antioxidant therapy after varicocelectomy on the pregnancy outcome in the assisted reproductive technology setting. This study was a retrospective cohort study. The subjects were couples among whom the male partner had undergone varicocelectomy and was scheduled for subsequent assisted reproductive technology. Pregnancy outcomes were followed retrospectively in 62 couples with male partners who received tocopherol (antioxidant group) and 37 couples who did not (control group). The tocopherol and control groups were assigned dependent on the decision of the physician in charge and the patient's request. The clinical pregnancy rates per couple and embryo transfer, time to pregnancy, and the number of cycles during transfer to pregnancy were evaluated. No significant differences were observed in the pregnancy rate per couple (antioxidant group 70.9% vs. control group 64.9%, P = 0.55) and per embryo transfer (50.4% vs. 39.6%, P = 0.22). Regarding the time to event analyzed by adjusted restricted mean survival time, the mean time to pregnancy was significantly shorter in the antioxidant (tocopherol) group (14.2 vs. 17.4 months, P = 0.025). No significant difference was observed in the embryo transfer cycle to pregnancy (mean embryo transfer cycles: 2.6 vs. 3.0, P = 0.238). Additional oral tocopherol nicotinate as antioxidant therapy after varicocelectomy was shown to shorten the time to pregnancy. It is recommended that add-on effects be tested in more well-designed randomized controlled trials to examine whether it improves assisted reproductive outcomes.

Efficacy of immune checkpoint inhibitors in patients with HIV-associated unresectable HCC (uHCC): Propensity-score matched analyses from two international consortia.

442 Background: HIV-associated HCC is an incompletely characterized disease with poor prognosis. People living with HIV (PWH) are commonly excluded from clinical trials, leading to a paucity of high-level evidence for optimal management. Whether ICIs are tolerated and effective in HIV-associated uHCC remains unclear. Methods: Using data from the CATCH-IT consortium and from a global dataset recruiting patients with HCC from 14 centers in 3 continents, we selected patients with HIV-associated uHCC treated with ICIs and compared their outcomes with a matched cohort of HIV negative (HIV-) patients. Primary endpoints were overall (OS) and progression-free survival (PFS). Propensity score matching (PSM) between the two groups was performed for the following variables: age, sex, HCC aetiology, Child-Pugh class (CP-C), ECOG-PS, BCLC stage, alpha-fetoprotein (AFP) levels, treatment line, portal vein tumor thrombosis (PVTT), and extrahepatic spread (EHS). Restricted mean survival time (RMST) difference analyses stratified by HIV status were performed for OS and PFS at 3, 6 and 9 months. Results: We accrued 46 PWH and 400 HIV-, treated mostly in the first-line setting (58.7% vs. 43.7%) with either atezolizumab plus bevacizumab (28.3% vs 17.5%) or anti-PD-1 monotherapy. Median age at ICI start was 62 (38-76) and 64 (18-87); viral hepatitis (HCV: 58.7%; HBV: 34.8%) were the prevalent aetiologies in PWH, non-viral (46%) was the most common in HIV-; most patients had BCLC-C HCC (84.8% and 82.3%). In PWH, median CD4 count was 345 cells/mm3 (IQR: 200-465); viral load was undetectable in 45 patients, all were established on combined anti-retroviral therapy. Following PSM, 44 PWH and 117 HIV- were included. Median OS was 7.5 months (95%CI: 5.2-NR) in PWH and 10.8 months (95%CI: 7.3-19.5) in HIV- (HR: 0.89; 95%CI: 0.56-1.42; p=0.58). Presence of PVTT, AFP&gt;400 ng/mL, and ECOG-PS=1 but not HIV status were associated with worse OS in uni- and multivariable models. Median PFS was 2.8 months (95%CI: 2.6-4.8) for PWH vs 3.0 months (95%CI: 2.4-5.3) for HIV- (HR: 0.83-95%CI:0.56-1.24 ;p=0.38); HIV status was not associated with worse PFS. RMST confirmed no OS or PFS differences based on HIV status. In matched cohorts, neither ORR (21% vs. 19.1%, p=0.97) nor DCR (44.7% vs. 55.4%, p=0.19) were associated with HIV status. All grade immune-related adverse events (irAEs) were more frequent in HIV (44.0% vs 21.7%, p=0.003), with similar incidence of grade ≥3 irAEs (22.2% vs 13.0%). PWH had lower incidence of dermatological toxicities (4.3% vs 22%, p=0.003) and a trend towards lower incidence of hepatotoxicity (6.5% vs 18%, p=0.059). Conclusions: This study provides practice-informing evidence to support the use of ICIs in PWH and uHCC. Our findings encourage the inclusion of patients with well-controlled HIV in prospective clinical trials.