FDA Approval Year Research Articles

Sex distribution in clinical trials of radiologic contrast agents: A 27-year review

PurposeClinical trials play a pivotal role in assessing the safety and efficacy of medical therapies. Addressing sex distribution among enrollees in clinical trials of radiologic contrast agents is essential for ensuring the generalizability of trial outcomes. Previous research has highlighted the influence of demographic factors, particularly sex, on treatment responses, emphasizing the need for equitable representation in clinical trials. Our study aim was to determine the sex distribution of enrollees in clinical trials of radiologic contrast agents. MethodsOur retrospective study included a total of 65 clinical trials conducted between 1990 and 2017 identified on clinicaltrials.gov after a comprehensive review including searching individually for all FDA approved contrast agents. Data collected included the year of FDA approval, the number of participants, sex distribution, trial location, trial phase, and study type. Inter-rater validation ensured data accuracy. ResultsOur analysis revealed fluctuations in sex distribution of trial enrollees. Enrollment of males exceeded females in most years, with a shift towards a more equitable representation in recent trials. Trials conducted in the United States had a higher rate of enrollment by females. Phase I trials had the most balanced representation, whereas Phase IV trials had the highest sex disparity. ConclusionAcross all trials, females made up 47.3 % of enrollees [3316 out of 7016 total enrollees]. Enrollment of males exceeded females in 44 of the 65 trials studied, females outnumbered males in 19 trials, and enrollment was equal between the sexes in 2 trials. While the sex distribution observed across all trials represents an equitable representation of enrollees, the wide variance of sex distribution at the level of individual trials has the potential to limit the generalizability of results.

A generative adversarial network model alternative to animal studies for clinical pathology assessment

Animal studies are unavoidable in evaluating chemical and drug safety. Generative Adversarial Networks (GANs) can generate synthetic animal data by learning from the legacy animal study results, thus may serve as an alternative approach to assess untested chemicals. AnimalGAN, a GAN method to simulate 38 rat clinical pathology measures, was developed with significant robustness even for the drugs that vary significantly from these used during training, both in terms of chemical structure, drug class, and the year of FDA approval. AnimalGAN showed comparable results in hepatotoxicity assessment as using the real animal data and outperformed 12 conventional quantitative structure-activity relationship approaches. Using AnimalGAN, a virtual experiment of 100,000 rats ranked hepatotoxicity of three structurally similar drugs in a similar trend that has been observed in human population. AnimalGAN represented a significant step with artificial intelligence towards the global effort in replacement, reduction, and refinement (3Rs) of animal use.

Oncologic outcomes of secondary cytoreductive surgery in the era of poly (ADP-ribose) polymerase inhibitors maintenance (549)

Oncologic outcomes of secondary cytoreductive surgery in the era of poly (ADP-ribose) polymerase inhibitors maintenance (549)

Biosimilar uptake of filgrastim and impact on spending in Medicare Part D from 2015 to 2019.

67 Background: The introduction of a filgrastim biosimilar in 2014 was associated with substantial cost savings in Medicare Part B and Medicaid programs. However, Medicare Part D is unique since it is unable to directly negotiate prices with drug manufacturers. We sought to investigate the uptake of filgrastim biosimilars and impact on spending among Part D beneficiaries. Methods: We evaluated utilization trends for filgrastim (Neupogen), filgrastim-sndz (Zarxio), and tbo-filgrastim (Granix) using the 2015-2019 Medicare Part D Prescription Drug Event data. We conducted a retrospective cross-sectional review of annual spending, number of beneficiaries, number of claims, spending per beneficiary, and spending per dosage unit. We excluded filgrastim-aafi (Nivestym) due to recent approval and adjusted for inflation using 2019 dollars. Results: In 2019, total aggregate Part D spending on filgrastim products was $78 million. From 2015 to 2019, the biosimilar share of total aggregate spending increased from $1.8 million (2%) to $44 million (56%), with combined biosimilar spending (Zarxio and Granix) eclipsing originator Neupogen in 2018 (within 4 years of FDA approval of Zarxio). Total spending on Neupogen reduced 58% from 2015 to 2019. While biosimilar uptake progressively increased every year, total aggregate spending on all filgrastim forms reduced only 7% from 2015 to 2019 ($84 million to $78 million). For all 3 forms, from 2015 to 2019, trends in spending were: average spending per claim ($3193 to $2549, -20%), average spending per beneficiary ($5880 to $6722, +15%), and average spending per dosage unit of filgrastim ($583 vs $571, -2%). Detailed results in Table. Conclusions: We demonstrate that the significant uptake of biosimilar filgrastim products in Medicare Part D from 2015 to 2019 was associated with a small decrease in aggregate spending, essentially unchanged per unit spending, and increased spending per beneficiary on filgrastim products. Our findings contrast with experiences across Medicare Part B and Medicaid, that demonstrated significant cost savings with biosimilar filgrastim uptake. This may be due to inability of Medicare Part D to directly negotiate prices with manufacturers (in contrast to Medicare Part B and Medicaid), supporting ongoing Congressional policy being debated in the United States Senate (H.R. 3).[Table: see text]

Observed survival benefit with limited exposure of durvalumab in unresectable stage III non-small cell lung cancer at a large community-based institution.

e20539 Background: The PACIFIC trial ushered in a paradigm shift in the management of unresectable, non-metastatic non-small cell lung cancer (NSCLC), demonstrating improvement in 12,24,36-month overall survival (OS) and leading to the 2018 FDA approval for durvalumab in unresectable or locally advanced stage III NSCLC. With almost 3 years of FDA approval, we performed a retrospective analysis of patient experiences and outcomes at Levine Cancer Institute analyzing patient data to assess survival and potential points of clinical significance. Methods: Patients over the age of 18, who met criteria similar to the PACIFIC trial (i.e. unresectable or locally advanced stage III NSCLC) from February 2018 through September 2020 were analyzed. Those who were receiving active treatment at the data cutoff were excluded. Patient characteristics, prior treatment, durvalumab administration, immune-related adverse events (irAEs), and efficacy data were summarized and evaluated. OS and progression free survival (PFS) were evaluated with Kaplan Meier methods. Results: A total of 159 patients were evaluated. 40.9% were female and 59.1% were male. The median age was 67 (range 38-83 years). Of note, 86.8% of patients were white, whereas 13.2% were nonwhite. 50.3% patients experienced an irAE. The most common reasons for discontinuation of durvalumab were completion (at least 24 doses), progressive disease, or toxicity (33.3%, 30.8%, 26.4%, respectively). The median number of doses of durvalumab received was 14 (range 1-26 doses). The median PFS was 15.3 months with 12-and 24-month PFS being 54% and 41.1 %, respectively. Median OS was 42 months with 12-and 24-month OS being 78.1% and 67.8%, respectively. Our analysis compared outcomes in those who completed adjuvant durvalumab versus those who did not complete adjuvant therapy (Table). Conclusions: Data shows the best survival in those who completed durvalumab (comparable to historic values) and novel data shows a perceived survival benefit in those completing 12 doses compared to those who did not. Thus, partial treatment may provide a survival advantage. Further multivariate analysis will look for possible correlations to increased immune events and inability to complete therapy. Further investigation will delve into this cohort’s small proportion of non-white patients, evaluating for possible barriers to care that may lead to more patients being diagnosed with stage IV NSCLC.[Table: see text]

Outcomes of Cardiac Diffuse Large B-Cell Lymphoma in the Rituximab Era

Introduction: Cardiac lymphomas are very rare with diffuse large B-cell lymphoma (DLBCL) considered to be the most common histology. These lymphomas can be either local "primary cardiac" disease, or part of dissemination by systemic lymphoma. There is limited data regarding outcomes of patients with this disease. We sought to evaluate the outcomes of cardiac DLBCL in both pre- and rituximab eras using a large retrospective database. Methods: The public Surveillance, Epidemiology and End Results (SEER) database was used to identify all patients diagnosed with DLBCL and heart as the primary disease site. Data cutoff in the database was in 2016. Patients with missing date of diagnosis or survival data were excluded. Patients were divided into two groups based on diagnosis year, with rituximab cohort included all DLBCLs diagnosed in 2006 and later (2006 was FDA approval year of rituximab in untreated DLBCL). Treatment effect (surgery, radiation, chemotherapy) was analyzed when available. survival was estimated using the Kaplan-Meier method, and compared using Log-Rank test. Cox proportional hazards models were used for adjusted survival analyses. Results: Total of 106 patients were included in the final analysis, baseline characteristics are summarized in table 1. Median age at diagnosis was 69.5 years with only about 10% of patients being 50 years or younger. Most patients were white 71 (67%), had local stage I/II disease 68 (64.2%), and belonged to the rituximab era group 71 (67%). Most patients had chemotherapy 82 (77.4%), while only 25 (23.6%) had surgery, and 16 (15.1%) had radiotherapy. Median overall survival (OS) for the entire cohort was 22 months (0-292). Median OS was 16 months (95% CI, 0.55 -31) for the pre-rituximab group, and was 26 months (95% CI, 7.5 - 45) for the rituximab group which were not statistically different (p-value =0.340). Median lymphoma-specific survival (LCS) was 30 months (95% CI, 8.0 -52) for the pre-rituximab group, and was 36 months (95% CI, 16 - 158) for the rituximab group which were not statistically different (p-value =0.295). OS and LCS were also not different between the two era groups when stratified by chemotherapy (figure 1). On univariate analysis, Chemotherapy was associated with better OS [HR = 0.472, 95% CI (0.277-0.806); p-value = 0.006] but not LCS [HR = 0.690, 95% CI (0.341-1.396); p-value = 0.302]. Using a multivariate analysis model, both OS and LCS were associated with lymphoma stage, insurance status and age but not with diagnosis era or chemotherapy (table 2). Conclusion: Cardiac DLBCLs are rare and affecting mostly the elderly. No significant improvement in outcomes were noted in the current rituximab era. Age, disease stage, and having health insurance were associated with better outcomes. The role of chemotherapy needs further evaluation in larger studies. Disclosures No relevant conflicts of interest to declare.

Evaluation of Serious Postmarket Safety Signals Within 2 Years of FDA Approval for New Cancer Drugs.

We examined how often new serious safety signals were identified by the U.S. Food and Drug Administration within the first 2 years after approval for new molecular entities (NMEs) for treatment of cancer that required specific regulatory actions described here. We identified, for all NMEs approved for treatment of cancer or malignant hematology indications between 2010 and 2016, substantial safety-related changes within the first 2 years after approval, which included a new Boxed Warning or Warning and Precaution; requirement for (or modification of existing) Risk Evaluation and Mitigation Strategies (REMS); and withdrawal from the market because of safety concerns. Fifty-five NMEs were approved between 2010 and 2016: 32 (58%) under regular approval (RA) and 23 (42%) under accelerated approval (AA). Of these 55 NMEs, 9 (16%) had substantial safety-related changes after approval. Across all 55 NMEs, one was temporarily withdrawn from the market for safety reasons (1.8%); one (1.8%) required a new REMS; nine required labeling revisions-new Boxed Warnings were required for two NMEs (3.6%), and new Warnings and Precautions subsections were required for eight (14.6%). One drug (ponatinib) was responsible for several of the substantial safety-related changes (withdrawal, REMS, Boxed Warnings). One of 32 NMEs approved under RA required a new Warning and Precaution, whereas 7 of 23 NMEs approved under AA had substantial safety-related changes in the first 2 years after approval. Based on our analysis we conclude that although there was a greater incidence of substantial safety-related changes to AA drugs versus RA drugs, the majority of these were changes to the Warnings and Precautions and did not substantially alter the benefit-risk profile of the drug. The majority of new cancer drugs (84%) approved in the U.S. do not have new substantial safety information being added to the label within the first 2 years of approval. Unprecedented efficacy seen in contemporary cancer drug development has led to early availability of effective cancer therapies based on large effects in smaller populations. More limited premarket safety data require diligent postmarketing safety surveillance as we continue to learn and update drug labeling throughout the product lifecycle.

PATIENT ACCESS TO PCSK9 INHIBITORS IS SUCCESSFUL UTILIZING A DEDICATED PROTOCOL AND CARDIOVASCULAR TEAM

After 3 years of FDA approval of PCSK9 inhibitors (PCSK9i) and guidelines recommending use, patient access and approval by payers remains low leading to under treatment. Establishing clinical pathways standardize patient care, therapy access, and outcomes. With PCSK9i, a healthcare system protocol

Use of Proprietary Names by Prescribers for Generic Products.

Proprietary names are often used when prescribing drug products in the United States. The purpose of this study is to describe prescribers' use of proprietary names for generic products, branded-generic names, on prescription orders and to identify prescribing practice trends to inform the development and evaluation of new proprietary names. To identify Abbreviated New Drug Application (ANDA) with branded-generic names approved between January 2003 and December 2012, we utilized the database provided by the FDA Office of Communications, Drugs@FDA . A national outpatient retail prescription database, IMS's Vector One: National (VONA) was used to identify prescribing trends by examining data for branded-generic names identified in Drugs@FDA as they were written on prescriptions for years 2003 to 2012, the last year of data collection for VONA. IMS Health, IMS National Sales Perspectives (IMS NSP) was used to retrieve the date that product sales were first reported (launch date). Our search of Drugs@FDA identified 65 distinct branded-generic names approved between January 2003 and December 2012. Data show that most of these products with branded-generic names are written on prescriptions and sold to pharmacies within a year of FDA approval. In some cases, the use of branded-generic names persists for up to 9 years after drug approval. This descriptive study confirmed that branded-generic names are used in prescribing. Thus, evaluation of orthographic and phonetic similarities between proposed proprietary names and branded-generic names is necessary when formulating and evaluating new proprietary names.

Examining the relationship between cost of novel oncology drugs and their clinical benefit over time.

6598 Background: The average launch price of oncology drugs has increased by 10% annually from 1995 to 2013. The purpose of this study was to determine if the clinical benefit of novel oncology drugs has increased proportionally over time and is correlated with launch price. Methods: Novel oncology drugs fromrandomized controlled trials (RCTs) cited for clinical efficacy evidence in drug approvals between January 2006 and August 2015 were identified. For each drug, only the first FDA approved indication was included. To determine clinical benefit, all included RCTs were scored using the ASCO Value Framework and the ESMO Magnitude of Clinical Benefit Scale. The launch price for the FDA approval year of each drug was extracted from RedBook. Each drug’s 28-day cost was determined using the dosage schedule outlined in the respective RCT and was adjusted to 2015 USD using the consumer price index. The relationships between 28-day drug cost and FDA approval year, and between incremental drug cost (difference in total drug cost between experimental and control arms accounting for treatment duration) and FDA approval year were examined using generalized linear regression models (gamma distribution and log link). Ordinary least square models were used to evaluate the relationship between ASCO/ESMO scores and FDA approval year. Spearman’s correlation coefficients between 28-day/incremental drug costs and ASCO/ESMO scores were also calculated. Results: Forty RCTs were included in this analysis. The 28-day drug cost was significantly associated with FDA approval year (p = 0.04), with an average increase of 8.5% per year. Incremental drug cost was also significantly associated with FDA approval year (p < 0.001) with an increase of 28.6% per year. The mean ASCO and ESMO scores were 26 and 3, respectively. Both scores were not statistically associated with FDA approval year (p = 0.73 and p = 0.86, respectively) and were also not correlated with 28-day or incremental drug costs (all rho < = 0.2). Conclusions: Novel oncology drugs are not priced according to their clinical benefit. The rising cost of novel oncology drugs over time is not associated with an increase in their clinical benefit, suggesting a decrease in their value over time.

Rates of PD-L1 expression testing in U.S. community-based oncology practices (USCPs) for patients with metastatic non-small cell lung cancer (mNSCLC) receiving nivolumab (N) or pembrolizumab (P).

11596 Background: FDA approved N for previously treated squamous mNSCLC in 3/2015 and non-squamous mNSCLC in 10/2015 regardless of PD-L1 expression. P was approved for previously treated mNSCLC in 10/2015 for patients whose tumors express PD-L1 as determined by an FDA-approved companion diagnostic (CoDx). Post-FDA approval treatment patterns of N and P, including PD-L1 testing patterns, remain unclear. Methods: We conducted a retrospective analysis using de-identified patient-level electronic health record (EHR) data aggregated from USCPs by Flatiron Health in accordance to an institutional review board approved protocol. Data was abstracted using both structured and unstructured EHR content with 10% duplicate abstraction to confirm data quality. All patients in the database with clinically-confirmed mNSCLC diagnosed from 1/1/11 to 3/31/16 and documented order or administration of N or P were included in the analyses. PD-L1 expression testing results were abstracted from unstructured data in EHR documents. Results: Within the first year of FDA approval, 1,362 mNSCLC patients received N or P (96.2% N; 3.3% P, 0.5% N+P). The median duration of treatment was 113 days. Overall, 11.3% of patients were tested for PD-L1 expression, 42.2% of whom tested positive. PD-L1 expression testing increased quarterly (Q) from 3.0% in Q2 2015 to 12.8% in Q1 2016. An FDA-approved CoDx was used 30.5% of the time. Among patients who received N, 90.5% were not tested for PD-L1 expression, 3.2% tested positive, and 4.7% tested negative for PD-L1. Among patients who received P, 32.7% were not tested for PD-L1 expression, 50.0% tested positive, and 1.9% tested negative. Conclusions: We observed low rates of PD-L1 expression testing in USCPs for patients with mNSCLC receiving N or P. Testing rates increased quarterly following FDA-approval. The majority of patients were not tested using FDA-approved CoDx. Most patients who were not tested for PD-L1 expression or tested negative received N. Clinician education will be important as data on optimal PD-L1 testing strategies guiding immunotherapy treatment decisions accumulates.

Hepatotoxicity with Vismodegib: An MD Anderson Cancer Center and Research on Adverse Drug Events and Reports Project.

BackgroundOn 30 January 2012, the US FDA approved vismodegib (Erivedge®, Genentech, CA, USA) for the management of both metastatic and locally advanced basal cell carcinoma.ObjectiveOur objective was to identify evidence of hepatotoxicity with vismodegib in the FDA Adverse Event Reporting System (FAERS) in treated patients in two National Cancer Institute Comprehensive Cancer Centers.MethodsFAERS was searched for reports dated 1 January 2009 through 31 December 2015 using terms including hedgehog pathway and vismodegib and hepatic-related terms such as liver, jaundice, and hepatitis, among others. Disproportionality analyses with estimates of proportional reporting ratio and empirical Bayesian geometric mean were conducted. A comprehensive literature review was conducted, and the clinical databases at the University of Texas MD Anderson Cancer Center and Robert H. Lurie Comprehensive Cancer Center of Northwestern University were searched.ResultsTwo cases of severe liver dysfunction were published (Common Terminology Criteria for Adverse Events [CTCAE] class III), and 94 reports of adverse events (AEs) were detected in FAERS, 35 of which were serious AEs. Safety notifications related to hepatotoxicity have not been issued by the manufacturer or the FDA, although vismodegib is listed in LiverTox and the European Medicines Agency website.ConclusionWe identified a detectable safety signal for hepatotoxicity for vismodegib within 4 years of FDA approval. Vismodegib should be used in patients with severe liver disease only after careful consideration, and concomitant hepatotoxic medications should be avoided. Rapid dissemination of such safety concerns is expected to result in fewer serious hepatotoxic AEs and more optimal outcomes for patients with cancer receiving vismodegib.Electronic supplementary materialThe online version of this article (doi:10.1007/s40268-016-0168-2) contains supplementary material, which is available to authorized users.

Impact of Reimbursement Cuts on the Sustainability and Accessibility of Dopamine Transporter Imaging

PurposeDopamine transporter single-photon emission computed tomography imaging utilizing iodine-123 ioflupane is accurate for differentiation of Parkinson disease from essential tremor. This study evaluates how reimbursement for I-123 ioflupane imaging changed between 2011 (year of FDA approval) and 2014 (year after loss of pass-through status for hospital-based outpatient imaging from CMS). MethodsI-123 ioflupane reimbursement data for our institution’s hospital-based imaging were compared between two periods: (1) July 2011 to October 2012, and (2) 2014. For each time period separately and in combination, averages and ranges of reimbursement for private insurance and CMS were analyzed and compared. A model to ensure recouping of radiopharmaceutical costs was developed. ResultsReview yielded 247 studies from July 2011 to October 2012 and 94 studies from 2014. Average reimbursement per study fell from $2,469 (US dollars) in 2011 to 2012 to $1,657 in 2014. CMS reduced average reimbursement by $1,148 in 2014 because of loss of radiopharmaceutical pass-through status. Average reimbursements from CMS versus private payors markedly differed in 2011 to 2012 at $2,266 versus $2,861, respectively, and in 2014 at $1,118 versus $3,470, respectively. Between 2011 to 2012 and 2014, the CMS percentage increased from 54% to 78%. Assuming that I-123 ioflupane cost $2,000, our model based on 2014 data predicts a practice with greater than 60% CMS patients would no longer recover radiopharmaceutical costs. ConclusionsReimbursement levels, payor mix, scanner location, and radiopharmaceutical costs are all critical, variable factors for modeling the financial viability of I-123 ioflupane imaging and, by extrapolation, future radiopharmaceuticals.

The AmpliChip: A Review of its Analytic and Clinical Validity and Clinical Utility.

In 2005, the FDA approved the Roche AmpliChip™ for clinical application. The AmpliChip is a microarray chip that has the capability to play an important role in clinical pharmacogenetics. Because of the possible influence the AmpliChip may have on patient medication management, the purpose of the review is to address the available evidence for the AmpliChip's overall performance at three key levels: analytic validity (genotyping accuracy, and prediction of the phenotype from the genotype) and clinical utility. We searched Medline, Embase and PubMed for studies of the AmpliChip. Limits were English language and 2005 (the year of FDA approval) and onwards, and we corresponded with authors for further papers of interest. 17 articles provided data for analysis in this review: 4 involving genotype accuracy, 7 involving genotype to phenotype prediction and 9 involving clinical utility. There is limited literature comparing AmpliChip results to gold standard tests and test-retest reliability when assessing genotype accuracy. Also, there is limited literature on the accuracy of AmpliChip predictions of phenotypes from genotypes and minimal evidence with appropriately powered studies whether the AmpliChip genotype to phenotype predictions result in clinical benefit. At all three levels there is significant evidence that the AmpliChip has the potential to be a robust clinical tool. However, more and adequately powered studies are required to determine fully whether the AmpliChip is a clinically effective tool.

Has Pharmaceutical Innovation Reduced Social Security Disability Growth?

This paper analyzes longitudinal state-level data during the period 1995–2004 to investigate whether use of newer prescription drugs has reduced the ratio of the number of workers receiving Social Security Disability Insurance benefits to the working-age population (the “DI recipiency rate”). All of the estimates indicate that there is a significant inverse relationship between disability recipiency and a good indicator of pharmaceutical innovation use: the mean vintage (FDA approval year) of Medicaid prescriptions. From 1995 to 2004, the actual disability rate increased 30%, from 2.62% to 3.42%. The estimates imply that in the absence of any post-1995 increase in drug vintage, the increase in the disability rate would have been 30% larger: the disability rate would have increased 39%, from 2.62% to 3.65%. This means that in the absence of any post-1995 increase in drug vintage, about 418,000 more working-age Americans would have been DI recipients.

The quality of medical care, behavioral risk factors, and longevity growth.

The rate of increase of longevity has varied considerably across U.S. states since 1991. This paper examines the effect of the quality of medical care, behavioral risk factors (obesity, smoking, and AIDS incidence), and other variables (education, income, and health insurance coverage) on life expectancy and medical expenditure using longitudinal state-level data. We examine the effects of three different measures of the quality of medical care. The first is the average quality of diagnostic imaging procedures, defined as the fraction of procedures that are advanced procedures. The second is the average quality of practicing physicians, defined as the fraction of physicians that were trained at top-ranked medical schools. The third is the mean vintage (FDA approval year) of outpatient and inpatient prescription drugs. Life expectancy increased more rapidly in states where (1) the fraction of Medicare diagnostic imaging procedures that were advanced procedures increased more rapidly; (2) the vintage of self- and provider-administered drugs increased more rapidly; and (3) the quality of medical schools previously attended by physicians increased more rapidly. States with larger increases in the quality of diagnostic procedures, drugs, and physicians did not have larger increases in per capita medical expenditure. We perform several tests of the robustness of the life expectancy model. Controlling for per capita health expenditure (the "quantity" of healthcare), and eliminating the influence of infant mortality, has virtually no effect on the healthcare quality coefficients. Controlling for the adoption of an important nonmedical innovation also has little influence on the estimated effects of medical innovation adoption on life expectancy.

Effect of introduction of paclitaxel on survival among women with stage III and IV ovarian cancer

5536 Background: The objective of this study was to evaluate survival over time in relation to FDA approval of paclitaxel (P) for second- and first-line treatment in a population-based cohort of women with stage III and de novo stage IV ovarian cancer. Methods: The Surveillance, Epidemiology, and End Results (SEER) program was searched to identify 8,267 and 10,746 women with stage III and stage IV epithelial ovarian cancer diagnosed between 1988–2004. Women were divided according to their year of diagnosis and year of FDA approval of P for second- (1992) and first-line(1998) treatment of ovarian cancer: Group1 (1988–1991; before P approval); Group2 (1992–1997; P approved for second-line); Group3 (1998–2003; P approved for first-line). Overall (OS) and ovarian-cancer-specific survival (OCS) were estimated using Kaplan-Meier product method and compared across groups with log rank statistic. Cox-proportional hazards models were fitted to determine the association of group year of diagnosis and survival after adjusting for patient/tumor characteristics. Results: Median age was 66 years. Median OCS was 44 and 18 months among women with stages III and IV disease, respectively. With stage III disease, 2-year OCS was 64%, 68%, and 70% for groups 1, 2, and 3, respectively (p < 0.0001). With stage IV disease, 2-year OCS was 39%, 41%, and 42% for groups 1, 2, and 3, respectively (p = 0.19). In the multivariable model for stage III disease, women in group 1 (HR = 1.4, 95% CI 1.2–1.5, p < 0.0001) and group 2 (HR = 1.2, 95% CI 1.1–1.3, p = 0.0003) had an increased hazard of ovarian-cancer-specific death vs. group 3. For stage IV disease, women in group 1 (HR = 1.2, 95% CI 1.12–1.3, p < 0.0001) had a significantly increased hazard of ovarian cancer-specific death, but no significant difference in group 2 (HR = 1.0, 95% CI 0.9–1.1, p = 0.88) vs. group 3. Similar trends were observed for OS. Conclusions: The survival of women with stages III and IV ovarian cancer has significantly improved with the introduction of P over the last two decades. However, the incremental improvement in survival with stage IV disease is clinically minimal and indeed not significant in the univariable analysis in the SEER patient cohort analyzed, suggesting a desperate need for new and more active drugs in these patients. No significant financial relationships to disclose.

Response to Baker and Fugh-Berman's Critique of My Paper, 'Why Has Longevity Increased More in Some States than in Others?'

Dean Baker and Adriane Fugh-Berman have published a critique of a study I performed in 2007, entitled “Why has longevity increased more in some states than in others?” One of the conclusions I drew from that study was that medical innovation accounts for a substantial portion of recent increases in U.S. life expectancy. Baker and Fugh-Berman claim that my study was subject to a number of major methodological flaws. Many of their claims pertain to the role of infant mortality; the definition of drug vintage; the issue of age adjustment; and the appropriateness of controlling for AIDS, obesity, and smoking in the analysis of longevity. In this article, I make the case that their claims about my study are largely incorrect. I show that infant mortality was not an important determinant of the growth in U.S. life expectancy during the period that I studied, and that my estimates are completely insensitive to the inclusion or exclusion of infant mortality. Controlling for the age distribution of the population also has essentially no effect on the longevity equation estimates. I argue that my definition of drug vintage, based on the initial FDA approval year of a drug’s active ingredient, is quite reasonable, and it is consistent with the FDA’s evaluation of the therapeutic potential of new drugs. I argue that controlling for AIDS, obesity, and smoking in longevity analysis is entirely appropriate and consistent with the epidemiological literature. Baker and Fugh-Berman express deep skepticism about my study’s conclusion that medical innovation has played a very important role in recent U.S. longevity growth, but they offer no explanation of why life expectancy increased by almost a year during 2000-2006, a period of increasing poverty and obesity and declining health insurance coverage.

Rapid Identification of Hematology-Associated Serious Adverse Drug Reactions (sADRS) with Small Datasets: Findings from the Research on Adverse Drug Events and Reports (RADAR) Project.

Current FDA and pharmaceutical industry drug safety efforts rely on statistical review of large databases populated with voluntary reports of sADRs. These data are limited by low rates of complete reporting and high rates of underreporting. Improved post-marketing drug surveillance efforts are needed to improve patient safety. The RADAR project is an academic pharmacovigilance program that focuses on small numbers of thoroughly researched cases of sADRs. In the last decade, RADAR has utilized the resources of a comprehensive cancer center and a global network of collaborators to evaluate and disseminate information on sADRs, many of which are related to hematology. Between 1998 and 2007, RADAR investigated and reported 16 hematology-associated sADRs. RADAR disseminated information on over half of the sADRs within 2 years of FDA approval, compared to half of all new sADRs documented in the Physician's Desk Reference within 7 years of FDA approval. RADAR's novel pharmacovigilance methods identified previously undetected sADRs based on small numbers of completely described reports. The RADAR project's experience shows that private sector programs that utilize small yet comprehensive datasets can significantly contribute to pharmacovigilance networks in hematology. As the FDA develops public-private networks to evaluate safety concerns associated with new drugs, focused independent efforts can assume increasing importance to address unmet needs.Summary of 16 hematology-associated sADRs reported by RADAR since 1998.Drug (FDA approval)sADR (# of reports)Clinical Setting (yr)Data SourceTime to detection (yrs)FDA/Company alertPEG-rHuMGDF(unapproved)ITP (13)Healthy volunteers (2001)Phase I trialsPreclinical detectionNAPEG-rHuMGDF(unapproved)Lymphomas (3)Healthy volunteers (2003)Case reportsPreclinical detectionNATiclopidine (1991)TTP (13)CVA pts (1998)Case reports<1DHP letter; boxed warningGemtuzumab (2000)VOD (50)Leukemia or AML (2000)Phase II trials<1Boxed warningLenalidomide (2005)VTE (75)Multiple myeloma (2005)Phase III trial<1Boxed warningEnoxaparin (1993)Hemorrhage (5)PTCA (2003)Case reports1NAClopidogrel (1997)TTP (39)CAD pts (1998)Case reports1Company warningThalidomide (1998)VTE (91)Multiple myeloma (2002)Phase II trials1Black box warningNevirapine (1996)Stevens-Johnson syndrome (3)Healthcare workers (2002)Case series2Boxed warningNevirapine (1996)Hepatotoxicity (5)Healthcare workers (2002)Case series2Boxed warning; DHP letterEPO (1993)PRCA (9)ESRD (2000)Referral hematologist5Boxed warning; DHP letterEPO (1993)Loss of efficacy (2)Counterfeit drug (2000)Case reports7NAEPO(1993)/ Darb (2001)VTE (52)Cancer (2004)Phase III trial10(EPO)/3 (Darb)FDA alert; DHP letter; boxed warningEPO (1993)/ Darb(2001)Death (NA)Cancer (2006)Clinical trials10(EPO)/3 (Darb)FDA alert; DHP letter; boxed warningG-CSF (1991)AML/MDS (16)Breast Ca (2007)SEER-Medicare12NAPiperacillin (1981)Neutropenia (1)Infection (2006)Case reports25Precaution