Utilization of External Control Arm to Contextualize Clinical Efficacy of Tisagenlecleucel Treated Among Patients with Relapsed/Refractory Follicular Lymphoma from the Single-Arm Elara Trial

Introduction: While randomized clinical trials (RCT) remain the gold standard for evaluating the safety and efficacy of treatments, interest in the use of external control arm (ECA) has grown when randomization is either infeasible or unethical (Friends of Cancer Research 2019). ECAs can provide contextual evidence to support regulatory approval in such situations. Single arm trials in patients with relapsed/refractory follicular lymphoma (FL) is one example due to the rarity of the disease and lack of standardized treatment in this patient population. The target trial framework (Hernán and Robins 2016) can be used to emulate a hypothetical RCT that ideally would have been performed to estimate the relative efficacy of treatment of interest when non-interventional studies were used to supplement single arm trials. We present an ECA in order to contextualize the efficacy of tisagenlecleucel in the single arm Phase 2 ELARA trial with evidence on standard of care (SOC) derived from the Flatiron Health Research Database (FHRD). Methods: Key components of the target trial include objective, population, treatment assignment, endpoints, start of follow-up period, performed comparison and causal effect of interest. These components were specified collaboratively with clinical and statistical input. The ECA was constructed by applying key eligibility criteria from the ELARA trial to FHRD. Key endpoints including real-world progression and real-world response of the target trial were identified and curated for ECA for comparison with ELARA by reviewing clinically relevant aspects for the population/disease of interest, understanding routine documentation practices of oncology healthcare professionals at both the system level (e.g., across various hematologic and oncologic diseases) as well as at the disease specific level (e.g., NHL), and extracting from the electronic health record (EHR) specific measurements of interest. A performance assessment was conducted to confirm the comparability of real-world and trial endpoints. Methods utilizing propensity scores were used to select a single line of therapy for patients from FHRD, ameliorate the impact of measured confounding and facilitate accurate estimation of the primary causal estimand. Feedback on the proposed indirect comparison was sought a priori from the European Medicines Agency (EMA). Upon feedback from EMA, the ECA was constructed and the causal estimand of interest was assessed. Results: The target trial is defined in Table 1. Feasible eligibility criteria from ELARA were applied to FHRD. Briefly, the ECA included patients diagnosed with FL who had at least two prior lines of therapy, including an anti-CD20 and alkylating agent and had not received a clinical study drug, anti-CD19, gene therapy or allogeneic hematopoietic stem cell transplant. Patients were aged at least 18 years at time of treatment and had at least 3 months of follow up. Some criteria from the trial could not be implemented. Endpoints included real-world response (rwR) and real-world progression (rwP). rwR was defined as the treating clinician's assessment of radiological response based on the assessment of the burden of FL disease over the course of treatment with a given regimen. rwP was defined as an evaluation by the treating clinician of progression in FL disease after systemic treatment. The performance assessment confirmed that the rwR/rwP data can be reliably captured and that the scan frequency for rwR is similar to ELARA. EMA acknowledged the limitations associated with the use of this ECA for indirect comparison including limited sample size, inability to match on certain inclusion/exclusion criteria, impact of missing data, potential misalignment on endpoint definition and unmeasured confounders. However, EMA recognized the usefulness of the ECA in contextualizing the efficacy of tisagenlecleucel in the current therapeutic landscape provided statistical analyses were used to address the limitations as far as possible. Detail results of this indirect comparison are included in a separate ASH abstract. Conclusion: Noting the limitations, an ECA constructed using EHR was considered useful to supplement and contextualize the clinical efficacy of tisagenlecleucel vs SOC from a single arm clinical trial using propensity score based methodologies to account for confounders. Figure 1 Figure 1. Disclosures Hao: Novartis: Current Employment. Hsu: Novartis: Consultancy. Parzynski: Novartis: Consultancy. Lobetti Bodoni: Novartis: Current Employment, Current equity holder in publicly-traded company; Gilead: Other: Travel sponsorship in June 2019; Spouse: F. Hoffmann-La Roche: Current Employment, Current equity holder in publicly-traded company; Spouse: NHS: Ended employment in the past 24 months; Spouse: Sanofi: Consultancy, Membership on an entity's Board of Directors or advisory committees; Spouse: Takeda: Consultancy, Honoraria, Speakers Bureau; Spouse: Harlcok Healthcare: Current holder of individual stocks in a privately-held company; Spouse: Celgene: Honoraria; Spouse: Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Spouse: Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Hampson: Novartis: Current Employment. Masood: Novartis: Current Employment, Current holder of stock options in a privately-held company. Rhodes: Novartis: Consultancy. Wu: Novartis: Consultancy. OffLabel Disclosure: Tisagenlecleucel (Kymriah) an autologous CD19- directed CAR-T-cell therapy, has been approved for children and young adults with relapsed/refractory (r/r) acute lymphoblastic leukemia and, adults with r/r diffuse large B-cell lymphoma.

Use of Granulocyte Transfusions in Two National Cohorts and Association between Transfusion Dose and Patient Outcomes: The Best Collaborative Study

Granulocytes for transfusion (GTX) continue to be administered, mostly to treat refractory infection 1. Single center series continue to be reported, but evidence of GTX effectiveness remains uncertain. Recent attempts at randomized trials of GTX have not been completed to target although the largest trial, RING 2, identified potentially promising results in a sub-group analysis by dose. GTX can be produced by apheresis (from 'stimulated' donors), aGTX, or as a component pooled from whole blood (WB) donations. Logistic difficulties arise with aGTX, including recruitment of suitable donors willing to undergo cell mobilization and collection. In England GTX are now only supplied as pooled (10 WB donations/pool, mean volume single pooled unit 207 ml, mean granulocyte yield/unit 0.9 x10 10, typical adult dose 2 pools, available 6 days per week). In France, production moved from aGTX to pooled in 2020 (20 WB donations/pool, mean volume single pooled component 428 ml, mean granulocyte yield/unit 1.8 x10 10, typical adult dose 1 pool, available 6 days per week). There have been no recent published national data on the clinical use of pooled granulocytes, or comparisons between large datasets. The aim of this study was to 1) describe the use of GTX, including pooled granulocytes, in two national cohorts of recipients; 2) estimate the effect of dose on patient mortality using statistical methods which could be applied to a larger dataset in the future. A pre-piloted data collection form (DCF) was used to collect prospective audit data on all GTX given to patients in participating hospitals in England from March 2017 -Sept 2020 in England. The PROspective Granulocyte usage and outcomEs Survey (ProGrES) was deemed a national registry/audit with no individual patient consent required for anonymized data. Information on patient characteristics and outcomes was collected at the time of the request for GTX, following completion of GTX and at 28 days and 6 months follow up. DCFs were adapted for international use, and French data were collected from Jan 2018 to Dec 2020. Descriptive analyses were performed to summarize GTX dose, patient characteristics, and outcomes. Using the English data, we investigated the association between GTX dose and mortality within 28 days of final GTX using logistic regression models with and without adjustment for possible confounders (age, infection source, use of renal therapy) and using the target trial framework. Ethical approval was obtained from London School of Hygiene and Tropical Medicine for analysis of English data. Results are presented on 224 (England, pooled) and 139 (France) patients who received GTX. In France, 95 and 44 patients received aGTX and pooled GTX, respectively. Table 1 shows patient features and outcomes. The most frequent underlying diagnosis was acute myeloid leukemia in all groups, and the most common indication was treatment of refractory infection. Median number of transfusions was similar in all groups (5.0, 5.3, 6.0). The mean dose/kg/transfusion was highest in the aGTX group (0.55 x10 9/kg), followed by French pooled (0.40) and English pooled (0.2). Death within 28 days of last GTX varied from 20.2% (French apheresis) to 32.5% (French pooled). Additional analyses were performed on the English data; for death within 28 days, a 0.1 x10 9 increase in GTX dose was associated with a 22% reduction (95% CI -45% to +0%) in odds of death within 28 days of referral; odds ratio 0.78 (95% CI 0.57 to 1.00). After adjustment for potential confounders, the odds ratio was attenuated to 0.92 (95% CI 0.67, 1.23). In summary, GTX continue to be requested by clinicians in patients with poor outcomes, despite an uncertain evidence-base. Regression analysis found dose was associated with lower odds of 28-day mortality. This was not significant after confounder adjustment, but this proof-of-principle analysis was limited by the sample size. A potential dose effect was also seen in the RING study. Our analysis using a target trial approach provides an approach for estimating GTX effectiveness in a larger cohort, which is required given the well-recognized challenges of undertaking randomized trials. The analysis provides a range of sample size calculations for testing an association between dose and outcomes for a defined effect size in a larger international cohort, supported by the BEST Collaborative 3. References: 1. Transfusion 2019;59(1)160 2. Blood 2015;126(18)2153 3. Transfus Med Hemother 2018(45)318 Figure 1 Figure 1. Disclosures Francis-Radice: GSK: Current equity holder in publicly-traded company; Smith & Nephew: Current equity holder in publicly-traded company; Bunzl: Current equity holder in publicly-traded company; Relx: Current equity holder in publicly-traded company; Cancer Research UK: Research Funding. McCullough: Fresenius Kabi: Honoraria; Terumo BCT: Honoraria.

HMG-CoA reductase inhibitors and COVID-19 mortality in Stockholm, Sweden: A registry-based cohort study

Background The relationship between statin treatment and Coronavirus Disease 2019 (COVID-19) mortality has been discussed due to the pleiotropic effects of statins on coagulation and immune mechanisms. However, available observational studies are hampered by study design flaws, resulting in substantial heterogeneity and ambiguities. Here, we aim to determine the relationship between statin treatment and COVID-19 mortality. Methods and findings This cohort study included all Stockholm residents aged 45 or older (N = 963,876), followed up from 1 March 2020 until 11 November 2020. The exposure was statin treatment initiated before the COVID-19-pandemic, defined as recorded statin dispensation in the Swedish Prescribed Drug Register between 1 March 2019 and 29 February 2020. COVID-19-specific mortality was ascertained from the Swedish Cause of Death Registry. Hazard ratios (HRs) were calculated using multivariable Cox regression models. We further performed a target trial emulation restricted to initiators of statins. In the cohort (51.6% female), 169,642 individuals (17.6%) were statin users. Statin users were older (71.0 versus 58.0 years), more likely to be male (53.3% versus 46.7%), more often diagnosed with comorbidities (for example, ischemic heart disease 23.3% versus 1.6%), more frequently on anticoagulant and antihypertensive treatments, less likely to have a university-level education (34.5% versus 45.4%), and more likely to have a low disposable income (20.6% versus 25.2%), but less likely to reside in crowded housing (6.1% versus 10.3%). A total of 2,545 individuals died from COVID-19 during follow-up, including 765 (0.5%) of the statin users and 1,780 (0.2%) of the nonusers. Statin treatment was associated with a lowered COVID-19 mortality (adjusted HR, 0.88; 95% CI, 0.79 to 0.97, P = 0.01), and this association did not vary appreciably across age groups, sexes, or COVID-19 risk groups. The confounder adjusted HR for statin treatment initiators was 0.78 (95% CI, 0.59 to 1.05, P = 0.10) in the emulated target trial. Limitations of this study include the observational design, reliance on dispensation data, and the inability to study specific drug regimens. Conclusions Statin treatment had a modest negative association with COVID-19 mortality. While this finding needs confirmation from randomized clinical trials, it supports the continued use of statin treatment for medical prevention according to current recommendations also during the COVID-19 pandemic.

1282Causal inference in multi-cohort studies using the target trial approach

Focus of Presentation Utilising data from multiple cohorts to address causal questions in health research has become increasingly widespread due to a number of advantages. These include improved precision of estimates, in particular to investigate effect heterogeneity as well as rare events and exposures, and the ability to examine the replicability of findings. However, undertaking causal inference in multi-cohort studies also faces several challenges, which makes clear causal thinking even more important than in single-cohort studies. We propose the use of the “target trial” framework for the conduct of causal inference in multi-cohort studies. Findings Using two case studies, the first considering the effect of maternal mental health on emotional reactivity and the second examining the influence of exposure to adversity on inflammatory outcomes in childhood, we describe and demonstrate how the target trial approach enables clear definition of the target estimand and systematic consideration of sources of bias. Considering the target trial as the reference point allows the identification of potential biases within each study, so that analysis can be planned to reduce them. Furthermore, the interpretation of findings is assisted by an understanding of the unavoidable biases that may be compounded when pooling data from multiple cohorts, or that may explain discrepant findings across cohorts. Conclusions/Implications Use of the target trial framework in multi-cohort studies helps strengthen causal inferences through improved analysis design and clarity in the interpretation of findings. Key messages The target trial framework, already well-established for casual inference in single-cohort studies, is recommended for the conduct of causal inference in multi-cohort studies.