PHILADELPHIA - The full data set for Human Genome Sciences' pivotal BLISS-52 study - presented at the American College of Rheumatology meeting on Oct. 20 - gives some indication that the firm's systemic lupus erythematosus candidate Benlysta (belimumab) could have a broader than expected commercial reach.

Topline results from the BLISS-52 trial were presented in July (¹ Also see "GSK's Option Strategy For Human Genome Sciences' Lupus Drug Looking Smarter After BLISS-52 Win" - Pink Sheet, 20 July, 2009.). As no lupus therapy has been approved in more than 50 years, and clinical development has proved challenging, the positive findings garnered much attention from Wall Street analysts and physicians.

But an FDA approval also is dependent on results from a nearly identical second pivotal study - BLISS-76, which evaluated patients primarily in North America and Europe. This study will likely be more heavily weighed by FDA, given that it includes a U.S. patient population (unlike BLISS-52). Those results are expected to be announced Nov. 2.

BLISS-52 was conducted primarily in Asia, South America and Eastern Europe. The study results showed that among 865 patients randomized and treated, belimumab met its primary efficacy endpoint by achieving a superior SLE patient response rate at Week 52 versus placebo plus standard of care.

Assuming the BLISS-76 data are positive too, Human Genome Sciences and development partner GlaxoSmithKline plan to submit marketing applications for Benlysta in the U.S., Europe and other regions in the first half of 2010. HGS and GSK entered into a co-development and commercialization agreement in August 2006 for the drug.

In the meantime, the fully analyzed BLISS-52 data provides some greater insight into the eventual commercial prospects for the therapy. Investigators presenting the findings at the ACR meeting noted some favorable outcomes for the drug that were not as evident when the initial results were announced.

Population Could Be Broader Than Planned

How Benlysta is used ultimately will depend on the labeling FDA agrees to. Still, at this point it is likely that the drug will be limited to the group enrolled in the clinical trial: seropositive (HEp-2 ANA > 1:80 and/or have anti-dsDNA antibody levels > 30 IU/mL) patients with SLE.

Michelle Petri, a Johns Hopkins rheumatologist who presented BLISS-52 data at ACR, believes the drug can be given to even more patients, if some of the achieved secondary endpoints make it into the label.

Specifically, if Benlysta receives "a steroid sparing label, a quality of life label, or a prevention of flare label," she told analysts during an Oct. 20 company-hosted call. "I think we could end up with the majority of lupus patients in the U.S., if there's a very broad label."

HGS estimates that about five million people globally, including 1.5 million in the U.S., suffer from various forms of lupus, including SLE.

But an FDA approval also will likely depend on an FDA arthritis advisory committee meeting, Petri predicted. "This will be the first biologic for lupus. I think the FDA is most certainly going to ask for an FDA advisory committee panel to answer some questions about labels," she told investors.

Benlysta Lessens Flares and Fatigue

Compared to the data available in July, Benlysta's impact on patient flares is one data nugget that's more pronounced now, Human Genome Sciences Exec VP R&D David Stump said in an interview.

"In July, the data did not show a lot of detail on flare outcomes," Stump said. He noted that controlling flare is important to physicians because it leaves patients with residual damage. "In Phase II studies we saw trends of flare reduction, but it was not robust enough to make flare a primary endpoint," he added.

The flare data likely is more robust because investigators had better guidance on concomitant medication and were able to implement learnings from past trials, Stump explained.

The full results show Benlysta "significantly delayed time to first SLE disease flare versus placebo" in both doses studied. On the SLE Flare Index, the median time to flare was 119 days for 10 mg/kg belimumab, 126 days for 1 mg/kg belimumab and 84 days for placebo. Compared to placebo, the risk of having severe SLE disease flares was reduced over 52 weeks by 43 percent in the 10 mg/kg treatment group and by 24 percent in the 1 mg/kg treatment group. And the risk of having one severe flare (BILAG A) or more than one moderate flare (BILAG B) in the organ domain score was reduced by 42 percent in the 10 mg/kg belimumab group and by 13 percent in the 1 mg/kg group vs. placebo.

The drug's ability to reduce time to flares could significantly bolster the drug's commercial potential, because patients taking Benlysta may be able to avoid higher doses of steroids and have less exposure to other medicines, Stump said. This is important because steroids carry many risks and a hefty side effect profile.

Steroid-sparing is a clinical effect FDA highlighted in its 2005 guidance on experimental lupus therapies.

Combined Data Should Support Steroid-Sparing

Though BLISS-52 results alone do not appear strong enough to warrant a steroid-sparing claim in the labeling, Petri said that combining data from both Phase III studies will likely make an "even more powerful statement on predisone sparing."

Of patients in BLISS-52 who were receiving more than 7.5 mg per day of prednisone at baseline, a significantly higher percentage of patients in the 1 mg/kg belimumab treatment group compared to placebo had their average prednisone dose reduced by at least 25 percent from baseline during the last 12 weeks of study.

A higher percentage of patients in the 10 mg/kg belimumab treatment group versus the placebo group also had their average prednisone dose reduced by at least 25 percent from baseline to 7.5 mg per day or less during the last 12 weeks of study, but the difference did not reach a level of statistical significance.

In patients receiving less than 7.5 mg per day of prednisone at baseline, significantly fewer patients in the 10 mg/kg belimumab treatment group versus the placebo group increased their prednisone use during the last 20 weeks of study. Fewer increases in prednisone use also were observed in the 1 mg/kg group versus the placebo during the same time period, but the difference also was not statistically significant.

Additionally, patients taking Benlysta experienced less fatigue versus the standard of care, a benefit seen within four to eight weeks in the 10 mg/kg group. Patients on both doses of Benlysta achieved statistically significant improvement of fatigue by Week 52 (FACIT-Fatigue Scale).

Improvement in health-related quality of life (HRQOL), measured by the SF-36 Physical Component Summary score at Week 24 - a pre-specified secondary endpoint - was not significantly different among treatment groups. But HRQOL improvement at week 52 was significantly greater in both belimumab treatment groups versus placebo.

Similar Designs, But Will Findings Be Similar?

The design of the two trials is similar, but the duration of therapy is different - 52 weeks for BLISS-52 and 76 weeks for BLISS-76. The longer study aims to show durability of effect, HGS' Stump said.

A Phase II follow-up study of patients on Benylsta for four years also shows that the drug is well tolerated, since nearly 50 percent of those initially enrolled opted to continue taking the drug.

"The follow-up data is amazing; the patients have clearly voted," said Joan T. Merrill, a professor of medicine at the University of Oklahoma Health Sciences Center, after BLISS-52 was presented at ACR as a late-breaking abstract. "It's only out four years, but I'm cautiously optimistic." Merrill also is chair of clinical pharmacology research at the Oklahoma Medical Research Foundation.

Lazard Capital analysts are predicting that the BLISS-76 data should resemble the retrospective Phase II data, which had response rates of 46 percent and 29 percent for Benlysta and placebo, respectively. The analysts also believe the second Phase III study will be the "more important catalyst" in terms of boosting Human Genome Sciences shares, according to an Oct. 21 note.

In a note following the ACR meeting, Alex To, founder of Cross Current Research LLC, cautioned that: "So far, we see no reason from the BLISS-52 trial to predict a negative outcome on BLISS-76, although any clinical trial could end up with the unexpected."

Human Genome Sciences is optimistic that the second trial will be consistent. "If we confirm BLISS-52 with BLISS-76, then we'll be in a very good position," Stump said, looking ahead to Nov. 2.

- Carlene Olsen ( ² c.olsen@elsevier.com )

Outlook BLISSful For HGS' Benlysta After Full Data Released At ACR

More from Archive

More from Pink Sheet