NEW YORK (TheStreet) -- Incyte (INCY) has been a source of investor contention since the company first announced Jakafi was effective in increasing overall survival in pancreatic cancer patients. The controversy centered on Incyte's analysis, which found a survival benefit favoring Jakafi only in a subset of pancreatic cancer patients with elevated levels of C-reactive protein (CRP), a biomarker for inflammation.
At this year's American Society of Clinical Oncology (ASCO) annual meeting, Incyte addressed the Jakafi controversy. The company asserted the drug's mechanism of action -- the inhibition of the JAK 1/2 pathway -- is known to reduce inflammatory cytokines, which in turn, leads to a more robust anti-cancer response in patients. In addition, the presence of systemic inflammation in patients with advanced cancer is a negative prognostic factor, so if one could reduce that inflammation, survival would increase, the company said.
To further explicate the biomarker analysis, the Jakafi ASCO presentation described a well-established clinical measure of inflammation called the modified Glasgow Prognostic Score (mGPS) which look at levels of CRP and albumin. For mGPS, patients with CRP below 10 mg/L have the best prognosis and are scored 0. Patients with CRP greater than 10 mg/L and albumin greater than 35 g/L are slightly worse and scored 1; those with CRP greater than 10 mg/L and albumin less than 35 g/L have the worst prognosis and are scored 2. This measure has been used in over 50 clinical studies that have independently validated its utility as a prognostic indicator across a range of tumor types.
Having a validated scale correlated with survival adds a level of comfort to the Jafaki pancreatic cancer subset analysis, with one caveat: Incyte examined the effect of Jakafi on patients with CRP above 13 mg/L, a different level than what's defined in the mGPS. Moreover, Incyte disregards albumin levels in its analysis.
Putting aside the question of the CRP cut-off, in pancreatic cancer patients with CRP levels above 13 mg/L, the addition of Jakafi lead to a statistically significant overall survival benefit, says Incyte. (OS hazard ratio was 0.47 with a p-value of 0.01.) Oddly, the progression-free survival benefit in this patient subset was not statistically significant. (PFS hazard ratio of 0.62 and p-value of 0.10.)
Incyte also analyzed the Jakafi pancreatic cancer data to parse the overall survival benefit based on mGPS-defined scores of 0, 1, 2, and those patients scored 1 or 2. These analyses provide a hint of what would happen to the Jakafi results with slight changes in the CRP cut-offs. Similar results would bolster confidence in Incyte's method of defining the subset of pancreatic patients benefiting from Jakafi.
There was no statistically significant overall survival benefit in any of the mGPS groups, although survival time trended higher with lower inflammation. This finding doesn't necessarily invalidate Incyte's use of a 13 mg/L CRP cut off but confidence in the analysis should be lessened. Incyte is conducting a phase III study of Jakafi in pancreatic cancer, recruiting patients based on mGPS. The uncertain survival finding from the phase II study based on the mGPS analysis is concerning.
Treating pancreatic cancer patients with Jakafi might have a minor benefit, but the totality of the results in the phase II trial do not inspire high confidence in the phase III program.
Sobek has no position in Incyte.