Spiriva Safety Debate Pits Clinical Study Results Against Meta-Analysis

FDA also asks advisory panel to assess mixed efficacy results for BI's drug to reduce exacerbation of chronic obstructive pulmonary disease.

Safety data from a 6,000-patient clinical trial and a safety signal from a 2008 meta-analysis will be a focus of discussion at the Nov. 19 Pulmonary-Allergy Drugs Advisory Committee meeting as FDA tries to resolve safety questions surrounding Boehringer Ingelheim's Spiriva HandiHaler (tiotropium bromide inhalation powder).

The deliberations are part of FDA's review of an sNDA to expand current labeling, which allows use of Spiriva as maintenance therapy for bronchospasm associated with chronic obstructive pulmonary disease.

One claim in the sNDA is for reducing exacerbations in COPD patients. Only GlaxoSmithKline's Advair Diskus (fluticasone/salmeterol inhalation powder) now has that specific indication, although the COPD space is becoming increasingly crowded. AstraZeneca's Symbicort received a COPD indication in February and companies including Forest Laboratories and Novartis are developing drugs for the market (Also see "Spiriva Finds COPD Exacerbation Studies Can Be An Exasperating Experience" - Pink Sheet, 22 Sep, 2009.).

The meta-analysis was published in the Journal of the American Medical Association in September 2008, muddying the safety waters for Spiriva just prior to FDA's Oct. 7, 2008, announcement that preliminary results from BI's UPLIFT study showed no increased risk of stroke with the drug (Also see "Spiriva’s “All-Clear” On CV Safety From FDA Could Raise Bar For COPD Candidates" - Pink Sheet, 8 Oct, 2008.). That safety signal had been the subject of an agency Early Risk Communication.

The meta-analysis, conducted by researchers at Wake Forest and the University of East Anglia, however, found a relative risk of 1.58 (95% CI 1.21, 2.06) for the primary combined outcome of cardiovascular death, myocardial infarction or stroke in patients receiving inhaled anticholinergics.

These include Spiriva and ipratropium bromide, which was approved as a bronchodilator for COPD in 1986. Seventeen randomized, controlled clinical trials were assessed in the meta-analysis.

The final results for BI's Understanding Potential Long-term Impacts on Function with Tiotropium study showed a rate ratio for death due to CV disorder of 0.81 (95% CI 0.48, 1.36), and for death due to MI of 1.00 (95% CI 0.43, 2.30), according to FDA briefing material.

Adjusting for exposure in patient years, overall cardiac serious adverse events in UPLIFT were less common in the tiotropium group, with a rate ratio of 0.84 (95% CI 0.73, 0.98), FDA says. MI occurred at a rate ratio of 0.71 (95% CI 0.52, 0.99), and angina at 1.44 (95% CI 0.91, 2.26).

OSE Finds Flaws In Meta-Analysis

In a review of the meta-analysis, the Office of Surveillance and Epidemiology concluded it had study selection bias, lacked discontinuation information and failed to use person-time data. OSE also found limitations with other published observational studies on tiotropium and ipratropium, according to the briefing material.

"Currently available data implicating tiotropium and ipratropium in increasing risk of cardiovascular death, myocardial infarction and stroke is not compelling," OSE concluded.

With regard to stroke, the subject of the 2007 Early Risk Communication, FDA reports that UPLIFT found, after adjusting for exposure in patient years, hazard ratios for stroke adverse events at 0.95 (95% CI 0.70, 1.29), stroke serious AEs, 0.97 (95% CI 0.69, 1.37) and fatal strokes (adjudicated) of 0.85 (95% CI 0.39, 1.87). While this does not suggest an increase of stroke events with Spiriva HandiHaler, FDA notes the uncertainty of the confidence intervals.

FDA plans to ask the advisory panel to assess whether data from UPLIFT address the potential safety signals of both stroke events and adverse cardiovascular outcomes.

The third voting question before the committee will be whether UPLIFT and another clinical trial, the VA study, are sufficient to support the claim for reducing COPD exacerbations. Boehringer Ingelheim also wants to add to labeling a description of long-term effects on lung function of Sprival Handihaler and a claim for reduction in respiratory failure.

The primary endpoints of UPLIFT dealt with lung function, as measured by FEV1 (forced expiratory volume in one second). While the results for these measures are sufficient to support the description of lung function BI seeks to add to labeling, they are not statistically significant, FDA says.

Ordinarily, this would eliminate consideration of the secondary endpoints, which included a number of measurements of COPD exacerbation. But since the results for these secondary endpoints "are quite robust," FDA says "it may be reasonable to consider the COPD exacerbation claim."

Based on an estimated hazard ratio using the Cox model, FDA says there is evidence that Spiriva HandiHaler reduced the risk of COPD exacerbation by 14 percent compared to placebo. The median time to first COPD exacerbation for those on the drug was 16.7 months, compared to 12.5 months in the placebo group.

However, FDA notes the number of exacerbations leading to hospitalization per patient year and resultant days in the hospital/patient year were not statistically significant.

In the Phase IV VA study, the co-primary endpoints were the proportion of patients who experienced a COPD exacerbation and the proportion with a hospitalization associated with exacerbation during the six-month trial period. The latter was not statistically significant and secondary endpoints related to exacerbation were numerically supportive but not uniformly statistically significant, FDA concludes.

BI withdrew its request for a fourth claim - reducing mortality - citing the desire to maintain consistency with global labeling.

- Cathy Dombrowski ([email protected] )