FDA requested that Merck conduct an additional study in primates during its review of Januvia (sitagliptin) due to concerns of necrotoxicity that were based on similar findings seen with other DPP-4 inhibitors.

While Merck had not seen necrotic skin lesions in its preclinical studies of the first-in-class agent, FDA began requesting in November 2005 that all sponsors of dipeptidyl peptidase-4 inhibitors conduct three-month oral toxicity studies in monkeys to assess their respective compounds' potential for necrotic skin lesions. FDA asked Merck to submit an amendment to its NDA no later than July 16, 2006, three months before the user fee goal date. Januvia was approved Oct. 16 (¹ Also see "Januvia Clears FDA; Sitagliptin Has “Neutral Effects” On Weight, Agency Says" - Pink Sheet, 23 October, 2006., p. 6).

The agency notified manufacturers that it had received data indicating that administering DPP-4 inhibitors to monkeys results in dose-dependent and duration-dependent increases in necrotic skin lesions. The toxicity appears to be a class-related effect, medical reviewers said, adding that several DPP-4 inhibitors in development have been associated with vascular and necrotic skin lesions in preclinical models, particularly primates.

For example, Novartis' DPP-4 inhibitor Galvus (vildagliptin) had a November 2006 user fee date, but FDA extended the review period three months to examine data from a three-month primate dermatological study, Novartis said. In the study, primates dosed with Galvus experienced blistering and flaking skin with some necrosis at higher doses. The company also noted that no such phenomenon has been observed in humans.

Merck sent an e-mail April 25, 2006, to FDA requesting a teleconference to discuss results of its monkey study, as well as a second study in progress with a compound described as "L-000000826," in which the lesions were observed. But the agency refused the request, saying it preferred "to review the study report for both compounds before discussing the similarity of the findings to other DPP-4 inhibitors."

Merck found no signs of skin lesions in any of the Januvia-treated monkeys, Director of Regulatory Affairs Steven Aurecchia told FDA in the e-mail.

The Januvia findings "are in distinct contrast to those evident in a recently initiated 14-week oral toxicity study with [L-000000826]," Aurecchia explained. "Findings to date include distinct changes in the skin of animals in all drug-treated groups. These changes include red discoloration of the skin and swelling in the face, limbs, abdomen and inguinal region," the exec stated, noting that the adverse events "appear to have partially or completely reversed in most of the affected monkeys."

"Are the findings described above with L-000000826 consistent with the division's observations thus far with DPP-4 inhibitors as a class?" Aurecchia asked the agency. "Additionally, subject to your review of the final reports for these two studies, are the contrasting findings with [Januvia] sufficient to support a waiver of the anticipated class labeling with respect to sitagliptin?"

FDA concluded that sitagliptin, unlike other unnamed DPP-4 inhibitors, does not cause skin lesions and that the effects could be due to other DPP subtypes.

"The mechanism for these lesions is not known; however, there is speculation that the risk may be increased with less selectivity of the drug for DPP-4 and some cross-reactivity with other serine proteases such as DPP-8/9," Division of Metabolism and Endocrinology Products Director Mary Parks concluded.

Medical reviewer Ilan Irony concurred: "The sitagliptin toxicity study conducted in monkeys was unrevealing. ... The main issue related to the toxicity of this new class of drugs relates to the specificity of inhibition of DPP-4. Sitagliptin is a highly selective inhibitor of DPP-4 and has not been associated with the known toxicities encountered by this applicant and sponsors of other DPP-4 inhibitors that are less selective."

Irony said that five other DPP-4 agents tested in monkeys were associated with dose- and duration-dependent necrotic skin lesions in preclinical studies.

Januvia labeling does not include any mention of necrotoxicity, although additional DPP-4 inhibitors could carry class labeling on the adverse event.

Merck conducted four pivotal studies to support approval: two in monotherapy, one in combination with metformin and one in combination with a peroxisome proliferator-activated receptor gamma agonist (thiazolidinediones). More than 2,300 patients with type 2 diabetes participated in the studies. FDA noted the robustness of the safety and efficacy data.

"There are relatively few controversies for this signatory to address," Office of Drug Evaluation II Director Robert Meyer remarked.

Meyer noted that because renal impairment prolongs sitagliptin's elimination and exposure, and given the prevalence of renal impairment in long-term diabetes, "dose adjustments are needed for renal impairment/failure and have been appropriately addressed by the OCP reviewers and the labeling." Labeling recommends renal function testing prior to initiating Januvia treatment.

Merck pooled safety findings from all of the Phase III studies and from long-term exposure in a subset of Phase II and Phase III study populations and their extensions. The pattern of deaths and serious adverse events was similar between treatment and control groups, review documents state.

Common adverse events occurring in roughly three percent of patients included diarrhea, nasopharyngitis, upper respiratory tract infection, urinary tract infection, arthralgia and headache. Nausea was more prevalent in subjects treated with 200 mg sitagliptin. The frequency of hypoglycemia was similar in sitagliptin-treated subjects as in control-treated subjects, and was much lower than those treated with glipizide, Irony noted.

The agency approved of the pooled safety analysis, saying the pooled data were more likely to demonstrate patterns of adverse events that would not be readily recognized during the review of single studies.

Irony also recommended that Merck study co-administration of sitagliptin with insulin or sulfonylureas. Parks concurred, saying that given "the progressive deterioration in glycemic control over time in type 2 diabetics and the high likelihood of concomitant insulin or [sulfonylurea] use, these studies should be conducted to adequately assess efficacy and safety, especially risk of hypoglycemia." She noted that Merck had submitted a special protocol assessment for combined use with insulin, but the division is requiring that both insulin and sulfonylurea be formally studied with sitagliptin as Phase IV commitments.

FDA agreed with Merck's proposed pharmacovigilance plan, which includes routine monitoring of lab parameters that were more frequently abnormal in sitagliptin-treated subjects, such as decreased alkaline phosphatase, and increased uric acid, creatinine and neutrophil counts. Exposure to sitagliptin, which is designated as category B, during pregnancy will be monitored by a pregnancy registry.

The Division of Medication Errors and Technical Support recommended against use of the Januvia name, citing possible look-alike confusion with Genentech/OSI's Tarceva (erlotinib). DMETS noted that both drugs are available in 100 mg daily dosages and pointed to the potential for serious adverse events if Januvia were dispensed instead of Tarceva.

Merck disagreed, saying DMETS' assessment was not objective since "the true source of potential [confusion] is poor handwriting on the part of the physician and/or poor dispensing practices by the pharmacist."

The Division of Metabolism and Endocrinology Products also disagreed with the DMETS assessment, partly because Tarceva is considerably more expensive than Januvia, and because the two indications are so different. Also, since the drug does not cause hypoglycemia in normal patients, "Januvia dispensed for Tarceva would not be of important short-term consequence," Meyers said.

Merck sales reps were detailing Januvia within 36 hours of approval on Oct. 16, and 70 percent of target physicians were reached within the first eight days of promotion, the company said. The firm is awaiting approval for a fixed-dose combination of Januvia and metformin ( Janumet ), and expects FDA action by the end of March 2007.

Novartis is highlighting superior efficacy of Galvus compared to Januvia. The firm had previously said it would position Galvus against other existing treatments, not head-to-head against Januvia. Galvus may have an advantage over Januvia because it has no limitation or dose adjustments based on different populations, Novartis said. Novartis is also planning to submit an NDA for Galvus/metformin combo in the first quarter of 2007.

AstraZeneca and Bristol-Myers Squibb announced a collaboration Jan. 11 to develop and commercialize two investigational diabetes agents (the DPP-4 inhibitor saxagliptin and the SGLT-2 inhibitor dapagliflozin). The deal is potentially worth upwards of $1 billion (² Also see "Bristol/Astra Diabetes Deal Further Indicates Bristol Will Stay Independent" - Pink Sheet, 15 January, 2007., p. 16).

OSI's Prosidion unit announced in November that it would seek a partner for its DPP-4 inhibitor. Also in November, GlaxoSmithKline halted its Phase III trials for its DPP-4 inhibitor Redona (denagliptin) after receiving unfavorable preclinical data.

[Editor's note: This story appears courtesy of ³ Pharmaceutical Approvals Monthly . For a sample copy, contact customer service at 800-332-2181.]

- Tamra Sami (t.sami@elsevier.com)

Skin Toxicity Seen In Other DPP-4 Inhibitors Missing In Januvia – FDA Review

More from Archive

More from Pink Sheet