Large scale adverse event data mining for targeted therapies development.

2017 ◽  
Vol 35 (15_suppl) ◽  
pp. 2538-2538
Author(s):  
Mayur Sarangdhar ◽  
Bruce Aronow ◽  
Anil Goud Jegga ◽  
Brian Turpin ◽  
Erin Haag Breese ◽  
...  

2538 Background: Targeted anti-cancer small molecule drugs & immune therapies have had a dramatic impact in improving outcomes & the approach to clinical trials. Increasingly, regulatory approvals are expedited with small studies designed to identify strong efficacy signals. However, this may limit the extent of safety profiling. The use of large scale/big data meta-analyses can identify novel safety & efficacy signals in "real-world" medical settings. Methods: We used AERSMine, an open-source data mining platform to identify drug toxicity signatures in the FDA’s Adverse Event Reporting System of 8.6 million patients. We identified patients (n = 732,198) who received either traditional and targeted cancer therapy & identified therapy-specific toxicity patterns. Patients were classified based on exposures: anthracyclines (n = 83,179), platinum (117,993), antimetabolites (93,062), alkylators (81,507), antimicrotubule agents (97,726), HER2 inhibitors (40,040), VEGFis (79,144), VEGF-TKis (90,734), multi TKis (34,457), anaplastic lymphoma Kis (7,635), PI3K-AKT-mTOR inhibitors (33,864), Bruton TKis (9,247), MEKis (4,018), immunomodulatory agents (174,810), proteasome inhibitors (44,681), immune checkpoint inhibitors (20,287). Pharmacovigilance metrics [Relative Risks & safety signals] were used to establish statistical correlation & toxicity signatures were differentiated using the Kolmogorov–Smirnov test. Results: To validate the use of the AERSMine to detect AEs, we focused on cardiotoxicity. It identified classic drug associated AEs (e.g. ventricular dysfunction with anthracyclines, HER2is & VEGFis; VEGFi hypertension & vascular toxicity; multi TKIs vascular events). AERSMine also identified recently reported uncommon toxicities of myositis/myocarditis with immune checkpoint inhibitors. It indicated a higher frequency of myositis/myocarditis with combination immune checkpoint therapy, paralleling industry corporate safety databases. These toxicities were reported at higher frequencies in patients > 65 yrs. Conclusions: AERSMine “big data” analyses provide a sensitive tool to detect potential new patterns of AEs simultaneously across multiple clinical trials & in the real-world setting.

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 4057-4057
Author(s):  
Ajeet Gajra ◽  
Andrew J Klink ◽  
Chadi Nabhan ◽  
Bruce Feinberg

Introduction: Despite high cure rates (>75%) achieved in adults with classical Hodgkin Lymphoma (cHL) with modern chemotherapy, radiation, anti-CD30 therapy, and stem cell transplant (SCT), a minority of patients are refractory to primary chemotherapy or at relapse accounting for about 1000 deaths annually in the US. cHL has a bimodal distribution seen in young adults and a secondary peak in patients over age 60. While patients over age 60 comprise less than 25% of new cases, they account for over 60% of deaths from cHL signifying poor prognosis in this group. The approval of immune checkpoint inhibitors (IO) in cHL has offered patients another class of agents to achieve disease response. The two approved agents in this drug class, nivolumab (NIVO) and pembrolizumab (PEMBRO) were approved in R/R cHL in May 2016 and March 2017 respectively. The common adverse effects of immunotherapy (IO) in clinical trials include fatigue, pyrexia, cough, musculoskeletal pain, diarrhea, rash, hypothyroidism and hypertransaminasemia. The objective of this real-world analysis is to assess the utilization of IO (NIVO and PEMBRO) in patients with cHL who have been treated outside of clinical trials and ascertain adverse events (AEs) and time to treatment discontinuation in patients aged ≥60 compared to those <60 years. Methods: Patients with at least 1 claim for cHL and treated with either NIVO or PEMBRO in second line or later (2L+; i.e., R/R disease) from May 2016 to October 2018 were identified from the Symphony Integrated Dataverse, a large US claims database containing linked longitudinal prescription, medical, and hospital claims. The IDV contains claims for 280 million active unique patients representing over 73% of specialty prescriptions, 58% of medical claims, and 30% of hospital claims volume in the US. Patients included in the study cohort had a 6-month pre-index period with no claims with a diagnosis of HL, were at least 18 years old at initiation of NIVO or PEMBRO, and no evidence of participation in a clinical trial, including receiving NIVO or PEMBRO prior to their approvals (May 2016 and March 2017, respectively). Patient characteristics and treatment patterns were summarized using descriptive statistics. Median and 95% confidence interval (CI) for duration of IO agent were calculated using Kaplan-Meier estimates. Results: Of the 338 patients with R/R cHL treated with an immune checkpoint inhibitor, 247 (73%) received NIVO, and 91 (27%) received PEMBRO. Median age at initiation of the first IO agent in the R/R setting was 58 years (range 18-80), 48% were over the age of 60 and 59% were male. Nearly two-thirds (62%) had commercial insurance, and patients resided across all 4 US regions. Half (50%) of patients received their IO agent in second line (i.e., first therapy in RR setting), 31% in third line, and 19% in fourth line or later. With a median follow-up of 13.4 months from R/R disease, median time to discontinuation of IO agent was 5.7 months (95% CI 4.8-7.5). Most common AEs among at least 10% of patients during treatment with an IO agent were fatigue (22%), anemia (16%), dyspnea (14%), and cough (12%); anemia was more common among patients ≥60 years old (22%) compared to those <60 years old (11%) (P<0.01). The times to treatment discontinuation used as a surrogate for duration of response were not significantly different by agent used or by age (Table). Conclusions: IO therapy with both NIVO and PEMBRO is widely used in R/R cHL irrespective of age. Despite a lower incidence of cHL in patients ≥60 almost half (48%) of patients in this database treated with IO were older. Possible explanations include greater propensity of refractory disease in older adults and possible preference for IO agents given that the AE profile is distinct from that of cytotoxic chemotherapy. Almost half of IO is being used early in the course of disease in the 2nd line. The AE profile is similar to that reported in clinical trials and does not differ significantly by age with the exception of anemia, which was more common in patients aged ≥60 years. Time to treatment discontinuation did not differ by age or by the agent used. Despite its inherent limitations, this analysis adds real-world evidence to the safety and efficacy data of IOs in cHL outside clinical trials. Disclosures Gajra: Cardinal Health: Employment. Klink:Cardinal Health: Employment. Nabhan:Aptitude Health: Employment. Feinberg:Cardinal Health: Employment.


2014 ◽  
Vol 32 (15_suppl) ◽  
pp. 3057-3057
Author(s):  
Wei-Wu Chen ◽  
Albiruni R. A. Razak ◽  
Philippe L. Bedard ◽  
Lillian L. Siu ◽  
Aaron Richard Hansen

2021 ◽  
Author(s):  
Vanita Noronha ◽  
George Abraham ◽  
Vijay Patil ◽  
Amit Joshi ◽  
Nandini Menon ◽  
...  

2020 ◽  
Vol 8 (Suppl 3) ◽  
pp. A310-A310
Author(s):  
Krishna Gunturu ◽  
Muhammad Awidi ◽  
Rojer Ranjit ◽  
Brendan Connell ◽  
Rachel Carrasquillo ◽  
...  

BackgroundICI revolutionized modern Oncology landscape and being utilized in metastatic to adjuvant and neo-adjuvant settings. As Oncologists, we are treating cancer patients with ICI every day, yet there is still a lot that is unknown about these drugs. We don’t have clear understanding of the efficacy and toxicity when sequencing one ICI for another. We conducted a retrospective review of real world data at Lahey Hospital and Medical Center to understand further and to pave path for prospective studies to understand this issue further to improve patient care.MethodsWe retrospectively reviewed Oncology patient charts who received ICI between January1, 2014 to December 18, 2018. Total 483 patients received ICI during this time frame and 22 of these patients received a second ICI either as monotherapy or in combination with other ICI or chemotherapy.ResultsA total of 22 patients received subsequent ICI after the initial ICI as showed in table 1. 15 of the 22 (68%) patients were transitioned from one ICI to another monotherapy. 11 of these patients were transitioned secondary to disease progression (73%), three had immune related adverse events and one was switched per standard of care. One patient had ICI re-challenge. Three patients had a transition from ICI monotherapy to combination ICI therapy. One patient went onto chemo-immunotherapy and 2 patients transitioned from combination ICI to chemo-immunotherapy.Abstract 284 Table 1Real world data of sequencing immune checkpoint inhibitors (ICI) after initial ICIConclusionsICI therapy is evolving and patients are being treated with multiple lines of ICI. In current practices, ICI is frequently being transitioned from cytotoxic T-lymphocyte antigen 4 (CTLA-4) and programmed cell death 1 (PD-1) or its ligand, programmed cell death ligand 1 (PD-L1) classes or combined with chemotherapy or targeted therapy. It would be prudent to explore the effects of sequencing these medications either as a monotherapy or in combination with other therapies to better serve our patients and to prevent financial toxicity.


Sign in / Sign up

Export Citation Format

Share Document