e19519 Background: The association of chemotherapy with potentially fatal adverse events such as SJS/TEN has not been systematically examined. Case reports in association with agents such as imatinib and docetaxel have been published in the literature. We applied data mining algorithms to assess the risk of SJS/TEN to chemotherapy. Methods: FDA Safety Information and Adverse Event Reporting Program MedWatch database was searched for cases of SJS and TEN reported in association with FDA-approved chemotherapy agents through December 2009. Proportional reporting ratios (PRR>2) and empirical Bayes geometric mean (EBGM>2) with a case count threshold ≥3 were considered signals of disproportionate reporting. Cases obtained by searching this database included those in which chemotherapeutic agents were used to treat cancer and non-cancer-related diseases. Results: Using our criteria for selection of cases, we have identified 252 TEN cases associated with 11 FDA-approved drugs. Similarly, 71 SJS cases in association with 4 FDA-approved chemotherapy agents were identified (Table). Conclusions: By identifying disproportionally reported agents we demonstrated that multiple chemotherapy agents may be associated with SJS/TEN. Although data mining may not be sufficient to establish causality, this method may serve as a supplemental tool to identify chemotherapy drugs that may warrant further investigation. Drugs Cases Signal n PRR 95% CI EBGM 95% CI SJS Bendamustine 8 6.67 3.34-13.28 5.54 2.33-10.85 Chlorambucil 10 2.66 1.43-4.93 2.21 1.16-3.92 Ibritumomab tiuxetan 17 4.64 2.48-7.65 3.52 1.88-6.03 Temozolomide 36 2.16 1.56-2.99 2.07 1.47-2.82 Total 71 TEN Bendamustine 5 9.53 3.97-23.32 7.21 2.39-16.38 Busulfan 17 3.85 2.39-6.18 3.29 2.00-5.245 Cytarabine 55 2.67 2.05-3.48 2.56 1.94-3.30 Chlorambucil 12 6.19 3.52-10.88 5.67 3.04-9.69 Dactinomycin 8 3.81 1.90-7.60 2.82 1.36-5.57 Fludarabine 40 3.62 2.65-4.93 3.37 2.44-4.56 Ibritumomab tiuxetan 4 2.96 1.11-7.86 1.93 0.69-4.52 Lomustine 14 17.17 10.23-28.79 14.81 7.49-24.27 Procarbazine 24 5.51 3.69-8.2 5.18 3.3-7.62 Thiotepa 8 4.84 2.42-9.66 3.38 1.59-6.81 Vincristine 65 2.12 1.65-2.7 2.06 1.6-2.6 Total 252
7229 Background: A debate exists over the extent to which gemcitabine is associated with potentially fatal pulmonary toxicities, namely interstitial pneumonitis. Methods: Herein, The Research on Adverse Drug reactions And Reports (RADAR) program evaluated 1) the quality and clinical characteristics of serious gemcitabine-associated pulmonary ADRs reported to the FDA (1997-present) and 2) reviewed medical literature cases (2001-present). Results: Completeness depending on whether reports were from the clinical trial (n=55) or routine care setting (n=95). Significantly more clinical trial reports reported the date of ADR onset (87% vs. 56%); days until onset (71% vs. 56%); pulmonary clinical signs such as dyspnea or hypoxia (82% vs. 69%), suspected drug relatedness (84% vs. 71%); and confirmed cause of death (83% vs. 67%) [p<.05 for all]. RADAR attributes better reporting in the clinical trial setting to prospective case reports forms mandated by IRBs and clinical trial sponsors. Gemcitabine was used for NSCLC (33%), pancreatic cancer (17%) and unspecified lung cancer (14%) and concomitant chemotherapy was used often (80% vs. 53%, clinical trial vs. not). These severe toxicities [pneumonitis (85%) & pneumonia (15%)] were diagnosed by CT or ultrasound (47%), more than one radiological exam (10%) or clinical means alone (43%); the median days until onset were 41 versus 55 days (clinical trial vs. not); 61% of patients were male; and 36 % of patients died. The medical literature indicates that concomitant chemotherapy significantly raises the risk of pulmonary toxicity. Conclusions: The lack of completeness in non-clinical trial versus clinical trial reports indicates that health professionals should strive to prospectively monitor and better report gemcitabine- associated pulmonary complications. No significant financial relationships to disclose.
