Atrasentan is a selective endothelin A receptor antagonist, which was found to have cytostatic properties in prostate cancer (reviewed here) . Due to promising early results, atresantan was initially developed for the treatment of hormone-refractory prostate cancer (see here). However, a phase 3 trial of atrasentan in non-metastatic HRPC, showed no statistically significant difference in time to progression between atrasentan and placebo (see ASCO abstract and Cancer 2008). Furthermore, the incidence of peripheral edema, nasal congestion, headache, dyspnea, and heart failure was greater with atrasentan than with placebo (Cancer 2008). However, the dose used in oncology trials was 10 mg/day compared to 0.75 mg/day for the planned phase 3 study in diabetic nephropathy (see below). In September 2005, the Oncology Drug Advisory Committee (ODAC) of the FDA recommended against approval of atrasentan (assigned the brand name Xinlay at the time) (see here). Another study of atrasentan added to standard chemotherapy in advanced prostate cancer also failed to show any benefit and was terminated by the DSMB for futility (see here and here). Due to these unfavorable results, Abbott changed its focus to the use of atrasentan in the treatment of diabetic nephropathy.
In a recently completed dose ranging phase 2 study in 89 subjects with diabetic nephropathy (eGFR >20 mL/min/1.73 m2; UACR 100 to 3000 mg/g), atresantan at doses of 0.75 or 1.75 mg/day resulted in significant reduction in UACR (see JASN article). Following the positive outcome of the phase 2 trial (see here), Abbvie has initiated a pivotal phase 3 trial of atrasentan in patients with diabetic nephropathy (see News release from Abbvie). If successful in reaching the pre-defined endpoints, this study will form the basis of a New Drug Application (NDA) submission to the FDA. The phase 3 clinical trial, called SONAR (Study Of Diabetic Nephropathy with Atrasentan) is a large, multinational, double-blind, placebo-controlled clinical study that is expected to enroll more than 4,000 patients with diabetic nephropathy. The inclusion criteria include estimated GFR (eGFR) of 25 to 75 mL/min/1.73 m2 (CKD stage 2 to 4), UACR >300 and <5,000 mg/g, and systolic blood pressure within 110 and 160 mgHg (see Abbvie News release). Based on the phase 2 study, the dose for this trial will be 0.75 mg per day.
The primary endpoint will evaluate the effect of atrasentan on time to doubling of serum creatinine or the onset of ESRD, as defined by need for chronic dialysis, transplant or death due to renal failure. Secondary endpoints will assess the effects of atrasentan on urine albumin excretion, eGFR and cardiovascular events including cardiovascular death, heart attack and stroke. Quality of life evaluations also will be conducted. More detailed information about the trial is available at (Abbvie News release).
Commentary: Very few drugs have been successful in slowing the progression of renal disease. Currently, only ACE inhibitors and ARBs are approved for this indication. Bardoxolone methyl, co-developed by Reata Pharmaceuticals and Abbott (now Abbvie) had shown promise in a phase 2 trial in patients with diabetic nephropathy, but its phase 3 trial had to be halted due to excess mortality in the bardoxolone arm (see KDDD posting and Reata news release). studies aimed at slowing progression of kidney disease in patients with diabetic nephropathy or other types of chronic kidney disease are expensive, requiring large numbers of patients and long treatment durations. Unfortunately, no surrogate endpoints exist to predict the clinically relevant outcomes (need for dialysis or death), so currently only composite hard endpoints, including doubling of serum Cr or a significant (usually >25%) decrease in eGFR, combined with time to dialysis or death are used in clinical trials. Changes in proteinuria or albuminuria are not acceptable as endpoints for registration purposes but are often included either as part of the composite endpoint or as secondary endpoints.
Another issue in performing phase 3 trials in CKD is the need to include patients who show a decline in eGFR, ascertained either from historical values or during a run-in period. Inclusion of patients with early stages of CKD (e.g. stage 2) in a trial aimed at slowing the rate of CKD progression will require very long (4 or 5 years) treatment and observation and large numbers of patients, in order to see a statistically significant change in eGFR. Patients at more advanced stage of CKD (e.g. stage 4) will progress more rapidly and require a shorter treatment and observation period, but in late stages, glomerulosclerosis, interstitial fibrosis, and other “irreversible” changes have already occurred, making it unlikely that the treatment will be effective. So, selecting patients for such trials is a very difficult task indeed. Some of these issues and proposed solutions have been discussed in a recent review by Formentini et al. that was published in NDT.
Until a validated surrogate endpoint, able to predict long-term outcomes in CKD patients has been identified, clinical trials aimed at assessing progression to ESRD will require long observation periods and thousands of patients to have a chance to be successful.