Diana Zuckerman, National Center for Health Research, February 26, 2020

Thank you for the opportunity to speak today.  I am Dr. Diana Zuckerman, president of the National Center for Health Research. Our center analyzes scientific and medical data to provide objective health information to patients, health professionals, and policy makers. We do not accept funding from drug or medical device companies, so I have no conflicts of interest.

Although active surveillance is recommended for many patients with low-risk prostate cancer, some patients prefer treatment and that can be a reasonable choice as long as the treatment has a meaningful benefit and does not increase patients’ harms. However, the clinical data for TOOKAD that we are discussing today do not clearly demonstrate that its benefits outweigh its risks – efficacy is uncertain, particularly for U.S. patients, and the drug has the well-known very negative side effects typical of prostate cancer treatments.  

There are 5 major problems with the data, and the FDA did an excellent job of expressing their concerns:

#1.  There was only 1 clinical trial. Replication is the key to science, and effective treatments are available for prostate cancer, so the FDA should not approve a new treatment based on just one study.

#2.  This one trial took place in Europe, and 99.8% of participants were white.  Five patients – not 5% but only 5 men — were not white. Given the higher percentage of African American men who die of prostate cancer, this lack of diversity should not be considered acceptable, unless the FDA is considering approving this product for white men only.

#3. The accuracy of detecting cancer is one major problem. The data showed a high false negative rate. 13.5% of those in the active surveillance arm had a negative biopsy after 2 years. Since it is unlikely that the cancer would spontaneously disappear, this probably reflects false positives in the original diagnosis or false negatives in the post-test. This high false negative rate raises major concerns about the accuracy of biopsies in the treatment arm and at baseline. 

#4. Further, there was a large amount of missing data regarding biopsies at 24 months. Unfortunately, 18.4% of the patients in the treatment arm and 41.5% of those in the active surveillance arm were missing these crucial pieces of information regarding efficacy.  That makes it impossible to draw conclusions based on this one study. 

 #5. The trial was open-label, meaning patients as well as those collecting the data knew whether a particular patient was receiving the treatment or active surveillance. The decision to pursue definitive therapy  — such as surgery — was a subjective decision. Wouldn’t active surveillance patients be more likely to choose treatments later. Also, when biopsy results were uncertain, pathologists may have been biased by their knowledge of whether or not a patient was assigned to the treatment group. 

The FDA held a workshop in September 2018 to discuss issues related to clinical trials of novel treatments for localized prostate cancer.  They concluded that treatment endpoints might be clinically meaningful only if 1) the patients using the new therapy were less likely to undergo subsequent treatment (such as surgery), 2) there was an overall reduction in adverse events, and 3) there was no reduction in long-term cancer control.  However, in this trial, these criteria were not met. 95% of those treated with TOOKAD had adverse event in the weeks after treatment compared to 55% of the men on active surveillance. For many men these problems continued 2 years after treatment, when 34% continued to have urinary problems compared to 16% with active surveillance. Two years after treatment, 38% of patients treated with TOOKAD had erectile dysfunction, compared with 12% of those assigned to active surveillance. In other words, erectile dysfunction was 3x as prevalent among those treated with TOOKAD.

The long-term complications of the treatment that are unclear.  In addition to the urinary problems and ED that might be worse for those who are subsequently treated with surgery, it may be more difficult to obtain accurate biopsies. 

The American Urological Association recommends active surveillance as the best available treatment for low-risk prostate cancer patients, and the data clearly support that recommendation.  We understand that the fear of cancer can persuade men to seek treatment, but if so they should not be bamboozled by ads to choose a product that may have no benefits for them but that does have clear risks of erectile dysfunction and urinary problems. If the FDA thinks this product might have a useful benefit, they should require at least one double-blind study – preferably two – since the treatment does not involved surgery and is therefore easier to blind.  The studies should also include a patient population that represents the diversity of men with prostate cancer. There is no need to rush to approve this particular treatment based on a single, open label trial with endpoints of questionable clinical relevance and problems with inaccurate diagnoses. Better research needs to be completed before approval because the sponsor did not comply with FDA’s recommendations for the study and was even less likely to do so after the product is approved.   

The Oncologic Drugs Advisory Committee voted 13 to 2 against approval of TOOKAD to treat low-risk prostate cancer.