Gilead Sciences Inc.’s submission late in the Sovaldi (sofosbuvir) review process of interim data from two Phase III trials overcame FDA reviewers’ initial skepticism about the appropriate breadth of the hepatitis C treatment’s indication, giving the sponsor the broad claim it sought for use in combination with other (unspecified) HCV treatments.

FDA review documents show an agency clinical team eager to approve the first-in-class nucleotide analog NS5B polymerase inhibitor in chronic hepatitis C patients with viral genotypes 1, 2, 3 and 4, with strong confidence in efficacy and no major safety or risk-management concerns.

“The review team was in agreement with a broad indication after supplemental information was reviewed from VALENCE and PHOTON-1 trials and after reconsidering the fact that treatment response varies and greatly depends on baseline host and viral factors,” Birnkrant said.

However, the appropriate breadth of the drug’s indication was a continuing source of internal deliberations, with some reviewers believing it should be limited to use only in combination with the treatments – ribavirin and pegylated interferon – it was studied with in the Phase III trials supporting the NDA.

Ultimately, however, it was data from the ongoing VALENCE and PHOTON-1 trials, which FDA requested and went to great lengths to review within the PDUFA time clock, that won the sponsor a broad claim for combination treatment unrestricted by the type of concomitant therapy.

FDA signaled the approach was under consideration during the advisory committee review. Eliminating the specific reference to RBV and peg-IFN could position Sovaldi for easy integration of the drug with new HCV treatments making their way through the FDA approval process and better reimbursement prospects, rather than awaiting approval for each combination to be included in labeling ([A#00131209013]).

The indication also makes sofosbuvir part of the first all-oral treatment regimen for chronic HCV genotypes 2 and 3. With labeling that includes dosing recommendations for interferon-ineligible, genotype 1 patients, the approval opens the door to potentially broad use of sofosbuvir as part of an all-oral regimen in HCV genotype 1 patients regardless of interferon eligibility.

Treatment guidelines recently issued by three medical groups make sofosbuvir the gold standard for treating chronic hepatitis C, in some cases recommending its use in combination with Janssen Pharmaceutical Cos.’ HCV protease inhibitor Olysio (simeprevir), which cleared FDA two weeks ahead of sofosbuvir’s Dec. 6 approval ([A#00140203010]).

“I am in agreement with the conclusions of the multidisciplinary review team that the risk-benefit assessment favors approval of [sofosbuvir] as part of a combination regimen in adult patients with [chronic hepatitis C],” Division of Antiviral Products Director Debra Birnkrant wrote in a Nov. 27 summary review.

“The review team was in agreement with a broad indication after supplemental information was reviewed from VALENCE and PHOTON-1 trials and after reconsidering the fact that treatment response varies and greatly depends on baseline host and viral factors,” she said. “I am confident that with the use of this drug in combination with an antiviral regimen, we will be able to positively address the burden of chronic hepatitis C.”

Interim data from the PHOTON-1 trial of HCV patients with HIV-1 co-infection also enabled the inclusion of this special population in the main indication statement, even though Gilead previously had decided not to request an HCV/HIV co-infection claim in the initial NDA due to the limited nature of data available at the time of submission (see box).

Sofosbuvir’s Indication

Sovaldi is a hepatitis C virus (HCV) nucleotide analog NS5B polymerase inhibitor indicated for the treatment of chronic hepatitis C (CHC) infection as a component of a combination antiviral treatment regimen.

Sovaldi efficacy has been established in subjects with HCV genotype 1, 2, 3 or 4 infections, including those with hepatocellular carcinoma meeting Milan criteria (awaiting liver transplantation) and those with HCV/HIV-1 co-infection.

And although treatment-experienced genotype 1 infected patients are not mentioned specifically in the indication statement, labeling reflects FDA reviewers’ exploratory analyses suggesting sofosbuvir is an effective treatment option for this group.

Duration An Issue In Genotype 3 Patients

Gilead’s NDA, submitted in April 2013, sought approval in combination with other agents for treatment of chronic HCV infection in adults (see timeline,Also see "Sovaldi Clinical Development" - Pink Sheet, 19 March, 2014.).

The original submission was based on data from three Phase III trials in HCV genotypes 2 and 3 – FISSION, POSITRON and FUSION – and the Phase III NEUTRINO trial in HCV genotypes 1, 4, 5 and 6 (see chart). The primary endpoint for all studies was sustained virologic response 12 weeks after discontinuation of active treatment (SVR12), which is the key efficacy endpoint recommended in a recently revised FDA draft guidance on new direct-acting antivirals ([A#00131104005]).

In the sofosbuvir treatment arms across FISSION, POSITRON and FUSION, significantly higher SVR12 rates were seen in HCV genotype 2 subjects compared to those with genotype 3 infection ([A#14130913005]). In FUSION, extending the treatment duration from 12 to 16 weeks increased SVR12 rates in genotype 2 subjects from 82% to 89%, and in genotype 3 subjects from 30% to 62%. These findings led the FDA review team to conclude that 12 weeks of sofosbuvir in combination with ribavirin may not be the optimal regimen for genotype 3 patients and that a longer duration may be beneficial.

The initial submission also included SVR4 data on 31 HCV/HIV-1 co-infected patients enrolled through September 2012 in the Phase III PHOTON-1 trial.

Due to slower-than-expected enrollment in the trial, Gilead informed FDA at an October 2012 meeting that it no longer planned to seek an indication for HIV/HCV treatment-naïve genotype 2 and 3 patients in the original NDA submission, although it would include the limited data available at that time in the filing. “Gilead expects to submit a supplemental NDA to support the HIV/HCV co-infection indication following the approval of the original NDA,” the meeting minutes state.

In a Sept. 6, 2013, memo, clinical reviewer Poonam Mishra included a discussion of the then-limited PHOTON-1 data and concluded, “available efficacy and safety data in HIV/HCV co-infected subjects is limited at this time and precludes a full indication in this patient population.”

In that same memo, Mishra addressed the breadth of Gilead’s proposed indication. “After in-depth discussions about the regulatory implications of proposed indication (for sofosbuvir use) – ‘in combination with other agents,’ the review team has decided to propose the use of sofosbuvir in combination with ribavirin in subjects with genotype 2 or 3 HCV infection and sofosbuvir in combination with pegylated interferon and ribavirin in treatment-naïve subjects with genotype 1 or 4 HCV infection,” Mishra said.

Sovaldi Phase III Data
Study	Population	Investigational Treatment	SVR12
Studies In Original NDA Submission
FISSION	HCV GT 2 or 3 treatment-naïve patients	Sofosbuvir and ribavirin (SOF+RBV) for 12 weeks	GT 2 – 95% GT 3 – 56%
POSITRON	HCV GT 2 or 3 patients intolerant, ineligible or unwilling to take interferon	SOF+RBV for 12 weeks	GT 2 – 93% GT 3 – 61%
FUSION	HCV GT 2 or 3 treatment-experienced patients	SOF+RBV for 12 or 16 weeks	12 weeks: GT 2 – 82% GT 3 – 30% 16 weeks: GT 2 – 89% GT 3 – 62%
NEUTRINO	HCV GT 1, 4, 5 or 6 treatment-naïve patients	SOF + RBV + pegylated interferon alfa for 12 weeks	GT1 – 89% GT 4 – 96%*
* Seven subjects with GT 5 or 6 infection are not included in the table.
Results From Ongoing Phase III Studies Submitted During The NDA Review
VALENCE	HCV GT 2 or 3 treatment-naïve or treatment-experienced	SOF+RBV for 12 or 24 weeks	GT 2 (12 weeks) – 93% GT 3 (24 weeks) – 84%
PHOTON-1	HCV GT 2 or 3 treatment-naïve or treatment-experienced patients with HIV-1 co-infection HCV GT 1 treatment-naïve patients with HIV-1 co-infection	SOF+RBV for 12 weeks (GT 2/3 treatment naïve) SOF+RBV for 24 weeks (GT 2/3 treatment-experienced) SOF+RBV for 24 weeks (GT 1 treatment-naïve)	GT 2/3 treatment-naïve – 75% (GT 2 – 88%, GT 3 – 67%) GT 2/3 treatment-experienced – 93% (GT 2 – 93%, GT 3 – 92%) GT 1 – 76%

New Data Spur Teleconference, “Breakthrough” Request

However, in the weeks leading up to the Oct. 25 Antiviral Drugs Advisory Committee review of sofosbuvir, the agency received top-line data abstracts from the VALENCE and PHOTON-1 trials that were to be presented in early November 2013 at the American Association for the Study of Liver Diseases annual meeting.

The VALENCE data appeared to support a longer, 24-week sofosbuvir regimen in genotype 3 patients, while the PHOTON-1 results provided support for a 24-week regimen in genotype 1 patients who are unable to take interferon, according to a memorandum of a Sept. 27 FDA teleconference with Gilead to discuss the new data.

“Based on this information, the agency requested all available VALENCE and PHOTON-1 data to be submitted. The submission should include interim study reports, raw data, some derived datasets, topline safety data, and clinical investigator information,” the memorandum states. “Additionally, Gilead was instructed to submit a breakthrough therapy designation request for sofosbuvir in combination with ribavirin for the treatment of chronic genotype 1, 2 and 3 HCV patients.” That designation was granted in early October within days of the formal request (see related story, [A#06140301004]).

In a Nov. 20 addendum to the clinical review, Mishra explained the regulatory reasoning behind deciding to consider so much new clinical data just two months before the user fee goal date for the priority review application.

“From a public health perspective, approving a suboptimal regimen when the emerging data is already available for 24-week treatment duration would not be beneficial for patients and would unduly expose patients to a suboptimal therapy,” Mishra said. “Taking these factors into consideration, a decision was made to review the currently available data from VALENCE trial during current review cycle rather than waiting for the trial completion and subsequent submission of data as an efficacy supplement.”

In addition, the PHOTON-1 top-line data “supported a treatment option (24 weeks of SOF+RBV) for interferon-ineligible genotype 1 patients and provided safety and efficacy data in the HCV/HIV co-infected patient population,” Mishra said.

On Oct. 9, Gilead submitted the VALENCE and PHOTON-1 data with efficacy results for 323 and 210 patients, respectively.

Extra Meetings Give Glimpse Of Evolving Agency Opinion

Since sofosbuvir was being reviewed under the PDUFA V “program” for new molecular entities and novel biologics that began Oct. 1, 2012, the NDA was the subject of more extensive interactions between the agency and sponsor. These included a mid-cycle communication on July 17 and a late-cycle meeting on Oct. 10.

At the late-cycle meeting, it was decided that Gilead would present the VALENCE data at the advisory committee meeting two weeks later. However, the company was directed not to present any data from PHOTON-1 because committee members had not been screened for conflicts for the HCV/HIV co-infection trial.

FDA also informed Gilead at the late-cycle meeting that internal agency discussions were still ongoing regarding the sponsor’s proposed broader indication for use in combination with other agents “versus how the sofosbuvir regimens were studied.”

At the advisory committee meeting, the panel unanimously endorsed sofosbuvir’s approval, in combination with ribavirin, for treatment of chronic hepatitis C in adults with genotype 2 and 3 infection, and in combination with pegylated interferon and ribavirin for treatment-naïve patients with genotypes 1 and 4 infection ([A#00131028022]).

However, FDA hesitation over Gilead’s requested broad indication was on display at the sofosbuvir advisory committee meeting and during the advisors’ review of Janssen’s simeprevir one day earlier ([A#00131104003]).

Gilead and its experts said a broad label would allow practitioners to easily incorporate use of sofosbuvir with new antiviral treatments that are quickly making their way through the development pipeline. A broad claim also would have commercial implications, with the expectation of easier reimbursement for the drug even if it was not used with ribavirin and pegylated interferon.

With simeprevir, Janssen also sought a broad indication in chronic HCV genotype 1 infection. During its panel review, however, DAVP Deputy Director Jeffrey Murray suggested the agency was not ready to adopt for HCV the same kind of “mix-and-match” approach to broad labeling that has been employed in the HIV arena.

Murray noted that FDA is “still being continuously surprised by the results of studies” and differences between HCV drugs within the same class. “We’re really getting a grasp on how many drugs are needed for how long, for each genotype and subgenotype. So I think it’s just a lot more complex, especially right now, to just kind of move to the HIV model,” Murray said. “I think we need more data under our belts before we do that.”

Although FDA officials stressed that the exact wording of sofosbuvir’s indication remained under discussion, review documents show the review team was leaning toward a more limited claim.

In a Nov. 8 memo written before the primary reviewers completed their assessments of the VALENCE and PHOTON-1 data, Cross-Discipline Team Leader Sarah Connelly said the review division had proposed an indication for combination use with ribavirin, or with pegylated interferon and ribavirin, for the treatment of chronic hepatitis C in adults. This proposed indication, Connelly said, was accepted by Gilead.

FDA’s cross-discipline team leader said she did not “believe there are sufficient data to support a broader indication for sofosbuvir in combination with other agents for the treatment of [chronic hepatitis C], as the applicant originally proposed.”

“I agree with the above proposed indication and, at this time, do not believe there are sufficient data to support a broader indication for sofosbuvir in combination with other agents for the treatment of CHC, as the applicant originally proposed,” Connelly wrote. “Specifically, only the regimens of SOF+RBV and SOF+PEG/RBV were used in the Phase III and supportive Phase II trials for this application with clinical study reports and reviewable analysis datasets. The recommendation for the currently proposed indication, however, does not preclude re-evaluation in the future as more experience is gained with sofosbuvir used with other agents, particularly as part of an interferon and/or ribavirin-free regimen.”

In the Nov. 20 clinical review addendum, Mishra said the PHOTON-1 data support sofosbuvir’s use with ribavirin for 24 weeks for the treatment of genotype 1 infection. Mishra also cited Phase II data from a National Institutes of Allergy and Infectious Diseases-sponsored trial, and the retreatment arm of a Gilead-sponsored Phase II study, QUANTUM, which “further support the results obtained in PHOTON-1 with SVR12 response rates of 68% and 66% in genotype 1 HCV-infected patients who are not co-infected with HIV, and who were treated with SOF+RBV for 24 weeks, respectively.”

“These Phase II and III data collectively demonstrate HCV genotype 1 treatment-naïve SVR rates between 66% and 76% following SOF+RBV for 24 weeks,” Mishra said. These rates are similar to the current HCV genotype 1 standard of care of a protease inhibitor – either Merck & Co. Inc.’s Victrelis (boceprevir) orVertex Pharmaceuticals Inc.’s Incivek (telaprevir) – “plus PEG/RBV with SVR rates between 66% and 75%, after 24-48 weeks of therapy. SOF+RBV offers an all oral, interferon-free, low pill burden option to HCV genotype 1 patients who may be unable to take interferon.”

The indication statement reflects the populations for whom efficacy was established in clinical trials, including those with HCV genotypes 1, 2, 3 and 4 infection, patients with HCV/HIV-1 co-infection, and patients with hepatocellular carcinoma meeting specific criteria and awaiting liver transplantation. The latter two groups also are mentioned in the Use in Specific Populations section of labeling.

The VALENCE and PHOTON-1 results are summarized in the Clinical Studies section, which includes tables with SVR12 response rates in various subpopulations.

Data from the two late studies also were used to inform dosing recommendations. Labeling states that sofosbuvir in combination with ribavirin for 24 weeks can be considered as a therapeutic option for patients with genotype 1 infection who are ineligible to receive an interferon-based regimen. “Treatment decision should be guided by an assessment of the potential benefits and risks for the individual patient.”

Notably, FDA also awarded a broad label to simeprevir, approving the protease inhibitor Nov. 22 for the treatment of chronic hepatitis C infection “as a component of a combination antiviral treatment regimen.” The indication is more limited, however, in that it states efficacy has been established only in combination with peginterferon alfa and ribavirin in genotype 1-infected subjects with compensated liver disease ([A#14131124001]).

Prior Nonresponders

While the Indications and Dosage and Administration sections of Sovaldi labeling do not specifically reference genotype 1 patients who were previous nonresponders to pegylated interferon and ribavirin, labeling nevertheless reflects FDA’s view that sofosbuvir treatment is expected to be effective in these patients.

Just as the review division went to great lengths to evaluate the VALENCE and PHOTON-1 data late in the current review cycle, it was similarly proactive with regard to the genotype 1 treatment-experienced population, conducting various exploratory analyses to support use in this patient group.

“During an End-of-Phase II meeting (June 5, 2012), when asked about sofosbuvir development plans in prior PEG/RBV treatment-experienced patients, the applicant stated they intended to identify the best regimen in HCV GT 1 treatment-naïve patients, then would proceed in trials in the more difficult to treat patients including those who failed prior PEG/RBV treatment,” FDA’s Connelly said in her cross-discipline review. “Therefore, HCV GT 1 patients who failed prior treatment with PEG/RBV were not specifically studied in the sofosbuvir development program supporting this NDA. During the review we recognized this population was in need of new therapies and, particularly due to the high overall SVR12 rate observed in NEUTRINO, attempted to estimate response rates in prior PEG/RBV nonresponders using existing data.”

FDA staff conducted exploratory analyses to try to estimate SVR rates in the genotype 1 treatment-experienced population using data on peginterferon/ribavirin treatment responses and baseline factors predictive of lower responses drawn from studies of other previously approved HCV drugs. FDA’s models resulted in predicted SVR12 rates ranging from 52%-78%.

”Together, these analyses provide supportive evidence that SOF+PEG/RBV for 12 weeks may be an effective treatment option in prior PEG/RBV treatment failures, with predicted SVR12 rates ranging 52%-78%,” Connelly said. “An ongoing trial offering sofosbuvir-based treatment to prior Gilead trial participants, GS-US-334-0109, has enrolled 59 HCV GT 1 subjects who did not achieve an SVR following treatment with PEG/RBV in combination with other [direct-acting antivirals]. These subjects have received retreatment with SOF+PEG/RBV for 12 weeks and the applicant states final SVR12 results will be available for submission mid-April 2014.”

A brief discussion of the exploratory analyses in the genotype 1 treatment-experienced group is found in the label’s Clinical Studies section following the presentation of the NEUTRINO results in treatment-naïve patients.

“It is estimated that the response rate in patients who previously failed pegylated interferon and ribavirin therapy will approximate the observed response rate in NEUTRINO subjects with multiple baseline factors traditionally associated with a lower response to interferon-based treatment,” labeling states. “The SVR rate in the NEUTRINO trial in genotype 1 subjects with IL28B non-C/C alleles, HCV RNA >800,000 IU/mL and Metavir F3/F4 fibrosis was 71% (37/52).”

With so much development concentrated in the HCV space, FDA is steadily amassing more and more data that should greatly enhance the understanding of how direct-acting antivirals treat the disease ([A#14140306004]). As the Sovaldi review shows, for such significant medical improvements the agency is willing to wield a great deal of flexibility, both in accepting early data and in drawing conclusions outside the initial dataset – the continuing submission of new agents should receive similar treatment, especially as the platform of knowledge, and FDA’s confidence, grows.

How Sovaldi Won A Broad Label: Emerging Data Swayed Reluctant FDA

More from United States

More from North America