Amgen Needs To Show Xgeva Does More Than Delay Bone Metastases, ODAC Says

Measurement of a downstream endpoint, such as delay in skeletal-related events, could have been better than radiographic scans to assess efficacy of denosumab in delaying bone metastases in high-risk castration-resistant prostate cancer patients, several ODAC members contend.

To prove the value of treating patients without bone metastases with Xgeva (denosumab), Amgen Inc. should look at a downstream endpoint, such as skeletal-related events, that would be more meaningful than prevention of bone metastases themselves, some panelists suggested at the Feb. 9 Oncologic Drugs Advisory Committee meeting.

ODAC recommended 12 to 1 against approving an indication for prolonging bone metastasis-free survival in men with castrate-resistant prostate cancer who are at high risk for such progression, citing lack of clinical benefit and a high risk of osteonecrosis of the jaw (Also see "Xgeva Comes Up Short On Clinical Benefit, Has Too Much Risk, ODAC Tells FDA" - Pink Sheet, 8 Feb, 2012.). Denosumab currently carries an approval based on a prevention of skeletal-related events endpoint; it was cleared to delay SREs associated with bone metastases from solid tumors, such as prostate cancer (Also see "Amgen Obtains Broader Label For Denosumab With Oncology Indication" - Pink Sheet, 22 Nov, 2010.).

Advisory Committee Vote

12-1 that denosumab does not have a favorable risk/benefit for treating patients with castrate-resistant prostate cancer at high risk for bone metastases

An important question is whether delaying bone metastases adds a clinical benefit beyond that already seen by the biologics’ ability to delay SREs, Brent Logan, Medical College of Wisconsin, noted. FDA had raised that issue in briefing materials for the meeting (Also see "Xgeva Prostate Cancer Claim May Not Benefit Beyond Existing Cancer Indication, FDA Says" - Pink Sheet, 6 Feb, 2012.).

“This question cannot be addressed using bone metastasis-free survival. It needs a study to look at some kind of overall survival or skeletal-related event endpoint, something’s that’s measurable after the late intervention. Otherwise, there’s no way of knowing whether the slowing of bone metastases from denosumab, which we’ve seen here, is actually added to the effect of denosumab on the skeletal related- events after metastasis.”

A hard endpoint, such as SRE prevention, would have been better, committee chairman Wyndham Wilson, National Cancer Institute, agreed.

Clinical Benefit Missing

The clinical trial to support the proposed indication employed an endpoint of bone metastasis-free survival as determined by radiographic imaging. Patients receiving denosumab were met-free 4.2 months longer than those on placebo. Data on overall survival, progression-free survival and patient-reported outcomes on pain and functional abilities did not show a benefit for denosumab over placebo. Patients were determined to be at high risk by PSA levels of 8 or higher or a doubling of PSA levels every 10 months or less.

The clinical meaning of the radiographic finding is called into question by data showing that 73% of patients with positive bone scans were asymptomatic, FDA suggested to the panel.

“We’re looking for a clinical benefit endpoint defined as how a patient feels,” FDA's Pazdur said.

That finding is a problem for the agency, Office of Hematology and Oncology Products Director Richard Pazdur said. Symptomatic metastasis has a much different meaning than a radiographic impact, he pointed out. “We’re looking for a clinical benefit endpoint defined as how a patient feels. … We’re not simply treating a bone scan, we’re treating a patient.”

Without symptoms, “patients had no immediate consequences” from developing metastases, Division of Oncology Products II Director Patricia Keegan explained. What is the value to the patient of delaying the detection of bone metastases by an X-ray? she asked.

Panel members failed to find an answer. James Liebmann, University of Massachusetts, suggested that “simply delaying the appearance of a positive bone scan by four months does not provide a significant [improvement in] quality of life as opposed to preventing … skeletal-related events.”

The findings would have been more convincing if they showed “patients living longer or living better,” Deborah Armstrong, Johns Hopkins University, agreed.

“At the end of the day … a symptom is much more clinically meaningful than simply a radiographic finding,” Wilson contended.

Too Little Symptom Information

The findings could have been more relevant if the trial had provided information on what the symptoms associated with bone metastasis were, according to Pazdur.

Investigators provided case record logs that included a check box of whether the patient was or was not symptomatic, he noted. “It doesn’t really characterize what those symptoms are. … If we were going to put weight in this as a clinically beneficial endpoint, I think we would want a greater degree of elaboration of what the symptoms were, what they led to, etc.”

Logan suggested there were more symptomatic patients than reported. Once bone metastases occurred, patients were dropped and there was no follow-up, he pointed out. Making inferences in this area when data collection ended prematurely “is very difficult to do.”

A Matter Of Context

FDA’s Keegan also pointed out that the new indication changes the risk-benefit ratio; longer use certainly increases the exposure to the side effects associated with Xgeva, while it has not been shown that early treatment with denosumab provides more benefit compared to waiting for metastases (the scenario cleared under the current labeling).

The assumption tends to be that the effects are additive, that the delay of bone metastases can just be added to the delay in SREs for a cumulative benefit. “That is in fact a major unknown,” Keegan said.

What is known, she continued, is that data from the trial used to support Xgeva’s indication for delaying SRE indicates that there is a 1% annual cumulative risk for osteonecrosis of the jaw in patients receiving denosumab. Because of this, patients may drop off treatment due to toxicity from longer term use of denosumab before they derive benefit from delaying SREs.

The trial does not answer the question of whether using denosumab to delay bone metastasis is better than using the therapy at the time of progression, Wilson agreed. “We just do not know that. This trial does not show that. … It could be better, the same or worse.”

In light of the risk of ONJ, Howard Sandler, Cedars-Sinai Medical Center, Los Angeles, urged caution. “In the prophylactic setting, where we’re taking men who are asymptomatic and only have a risk in PSA and giving them therapy to try to delay something down the road, I think there’s an extra burden that we as physicians should carry in minimizing potential toxicity from a prophylactic therapy.”

Including cases of ONJ confirmed in an open-label phase of the trial, FDA calculated a 6.6% incidence of ONJ in denosumab-treated patients, compared to zero in the placebo arm.

The lone vote in favor of the new indication for Xgeva was patient representative James Kiefert, who argued that the possibility of delaying bone metastases could be a clinical benefit for men with CRPC and Xgeva would provide oncologists with another tool to treat the disease. All prostate cancer therapies have significant side effects, he pointed out.