Authors

  1. Nalley, Catlin

Article Content

NEW YORK CITY-Historically, overall survival (OS) has been considered the primary endpoint in randomized clinical trials of ovarian cancer.

  
ovarian cancer; clin... - Click to enlarge in new windowovarian cancer; clinical trials. ovarian cancer; clinical trials

This focus was justified by the toxicity of available regimens, the marginal survival benefit of subsequent therapies (second-line and beyond), and the belief that ovarian cancer for the most part was considered relatively homogenous in its biology, according to Maurie Markman, MD, President, Medicine and Science, Cancer Treatment Centers of America, who recently discussed this topic during the Chemotherapy Foundation Symposium, held Nov. 8-10.

 

"Because of advances in the treatment of patients with ovarian cancer and the effectiveness of multiple lines of therapy, we need to come up with a better approach towards evaluating effective therapies," Markman told Oncology Times. "Today, if a patient finishes a trial, they may then go on two, three, four, or more other lines of therapy, which is going to impact a survival end point."

 

Evolution of Treatment

To look to the future of ovarian cancer research and treatment, oncologists must not forget the lessons of the past.

 

"From a historical perspective, it is obviously quite clear that we have moved a long way. But in the world of clinical trials, recognizing in the 1970s, '80s, and even '90s, we were often talking about single agents with modest, but very real, activity," noted Markman, who is also Clinical Professor of Medicine at Drexel University College of Medicine, Philadelphia.

 

These agents, however, were often associated with severe side effects.

 

"With cisplatin, it was very clear that in combination therapy we were having a far greater impact on the natural history of the malignancy, but the toxicity was even greater," Markman said. "And, therefore, it was very appropriate at the time to say, 'If we cannot show an improvement in overall survival in this malignancy, how can we possibly say this should become a standard care option?' There was very little debate.

 

"To add to that, we had quite marginal activity for most second-, third-, and fourth-line strategies in that era, so the idea that a post-therapy survival outcome would be influenced by a second-line therapy was wishful thinking at best," he continued.

 

Historically, and even today, although this is changing rapidly, according to Markman, ovarian cancer has been viewed as a single entity and the management of clinical trials has reflected this perception.

 

"Well, the world has changed, very positively and very dramatically, in management of ovarian cancer," Markman remarked during his presentation. "Increasingly, there are both investigative and non-investigative, clinically active, antineoplastic agent regimens with improved toxicity profiles and associated supportive care medications, making those strategies increasingly acceptable not only for 3-4 cycles, but potentially for much longer use.

 

"Prolonged survival is a real possibility, even following initial disease progression," he emphasized.

 

"What does this have to do with ovarian cancer clinical trials?" Markman asked. With the continued development of more effective second-line therapies, as well as an increased understanding of the molecular basis of the disease, the overall survival endpoint can prove limiting, he noted.

 

Molecular Subsets

"Unfortunately, this idea of randomized trials does not deal with the very serious, increasingly relevant issue of rapid and, in my opinion, quite revolutionary changes in our understanding of the biology of cancer," Markman noted.

 

Citing the example of the recent approval of the checkpoint inhibitor pembrolizumab for patients with microsatellite instability-high (MSI-H) cancer, Markman emphasized the value of treatments that impact a small subset of patients.

 

"This approval includes ovarian cancer patients, and yet that is less than 2 percent of the patient population," he noted. "If that approval had not been granted, MSI-H would never have been looked at or approved in this population because you could not have done a trial."

 

Emphasizing this disconnect in ovarian cancer clinical trial development, Markman offered another example.

 

A study published in 2003 looked at trastuzumab in HER2 overexpressing ovarian cancer (J Clin Oncol 2003; doi:10.1200/JCO.2003.10.104). According to Markman, trial eligibility included 2+ and 3+ HER2 overexpression by immunohistochemistry. Researchers analyzed 837 tumors and there was a total of 14 patients who had 3+ overexpression.

 

"The trial had a 7 percent overall response rate and was thus labeled a negative trial," Markman reported. "Is it fair to say that trastuzumab, is inactive in ovarian cancer when 14 patients actually had the target as far as we can define it today? There was nowhere for this drug to go in ovarian cancer on the basis of this study, and I would argue that is a tragic conclusion.

 

"As we rapidly enter the world of precision cancer medicine, we are looking at very small subsets of patients with ovarian cancer. Perhaps less than 5 percent, maybe less than 2 percent," Markman reiterated. "We have got to figure out a way to evaluate the effectiveness without a phase III randomized trial requiring hundreds of patients, and taking years or maybe decades to complete to determine whether a drug is of value. We need to come up with new paradigms."

 

Alternative Endpoints

There must be a shift in the development and management of ovarian cancer clinical trials to accelerate progress in the treatment of this disease, noted Markman.

 

"We have to overcome barriers because there is a major impact on post-trial therapy," he explained. "Additionally, we have to find a way to decrease the required randomized trial size."

 

Progression-free survival (PFS) as a primary endpoint makes a lot of sense, according to Markman, because, depending on trial outcomes, this could permit earlier approval for non-investigative use as well as allow for permitted biological relevant crossover. However, he cautioned, "this idea of having PFS as a primary endpoint does not prevent one from still conducting what are biologically irrational trials."

 

With a growing focus on molecular targets and precision medicine, Markman emphasized the need to look for endpoints beyond OS. He suggested the following should be considered in the future: high objective response rates in small subsets of patients; prolonged time to disease progression; and longer time to disease progression.

 

"Of course, in each of these situations, we have to look at the question of toxicity and impact on quality of life," he concluded. "Simply saying there is an objective response does not make a drug valuable and beneficial; but from an efficacy endpoint, we have to look at new ways of evaluating benefit to our patients."

 

Catlin Nalley is associate editor.