Early, Earlier, Earliest (Docetaxel)

Chemotherapy improves survival when given to patients prior to radical prostatectomy.

It is the latest, and potentially the last, piece of data in the decades-long march of this important and interesting (but much-maligned) therapy. Will this news change practice?

I am talking about the results of CALGB 90203, which was presented by James Eastham of Memorial Sloan Kettering at the 2019 AUA Annual meeting. And while there is a bit of controversy in how we are going to interpret the data, the bottom line is that a treatment effect from docetaxel was observed in high-risk patients – data which should not be ignored.

In brief, here is the story. Seven hundred and eighty-eight patients were randomized to receive androgen deprivation therapy (ADT) plus docetaxel followed by radical prostatectomy (RP) versus immediate radical prostatectomy. The ADT was discontinued after the RP in all patients.

This was a study for high risk patients, as a Gleason score of 8 or an estimated risk of recurrence of 60% or greater based on a pretreatment nomogram, was required for eligibility. Almost 90% of the patients enrolled had Gleason scores of 8 or higher.

The primary endpoint of the study was not overall survival (OS), it was PSA recurrence, defined as a PSA measurement of 0.2ng/dL or higher 3 years following prostatectomy. Survival and freedom from treatment failure were secondary endpoints.

The study did NOT meet its primary endpoint although it was fairly close – 13% of those in the chemotherapy arm vs 19% in the immediate RP had recurred per this definition and the p value was 0.067. Interestingly, looking 5 and 8 years out from RP the curves continued to separate and yes, the OS figures support a benefit. The hazard ratio for survival is 0.66 in favor of the docetaxel arm, and the p value is 0.06.

Wait, the p value is 0.06, so this isn’t statistically significant, right?

Therein lies the rub. The treatment effect is clear. The survival curves are separate and it is not subtle. But, this is a secondary endpoint, and it is not below the magic p value threshold 0.05, so it will be viewed with skepticism

The data will also evolve, as there are many patients on both arms who are still alive, and as they are followed we are likely to see the curves continue to separate over time. So stay tuned. What will this mean for the next patient you see with a localized disease but a Gleason score of 9 who is a good surgical candidate?

I should note that there were no deaths due to chemotherapy toxicity in the arm and there was no signal that surgery was complicated in any way by excess blood loss or other complications.

So what does this mean about the increasingly broad applicability of docetaxel in this disease? The history of docetaxel's use in prostate cancer is now in its third decade, and we continue to see new areas of benefit. Although it wasn’t approved by the FDA until 2004, it was used in clinical trials as well as “off label” in the mid 1990s, after it showed benefit in breast cancer. Now, with the 20+ year view, we are able to see the trend that earlier really is better, but what about the earliest?

Recall that the Tax 327 study was conducted in metastatic castration resistance prostate cancer (mCRPC)  and approximately 45% of the patients treated had such advanced disease that they required, at baseline, opiate use for analgesia. You may not recall, however, that the 55% of patients who did not require opiates had a significant overall survival advantage compared to the 45% who did AND that the margin of benefit over mitoxantrone, the control arm, was NOT statistically significant in the opiate requiring arm. So, if you consider ‘opiate requiring’ to be a later form of the disease than ‘asymptomatic’ as I do, then even Tax 327 had embedded in it, a study of ‘early vs late’ the OS better for those without pain (early CRPC). This was followed by CHAARTED and STAMPEDE, demonstrating that docetaxel exerted an OS benefit when given in combination with initial ADT in men with untreated metastatic disease, and now there is CALGB 90203, the study detailed above.

We will wait for the data to evolve, and certainly for the peer reviewed publication, but these data are provocative to say the least.

So how could docetaxel be improving the OS of men when given before ADT?

  1. Elimination of micromets
  2. Elimination of potentially lethal clones in the prostate itself.
  3. Downsizing of tumor/pulling back invasive disease, reducing positive margin rate or ECE, thus reducing recurrence.

This is also a story about the power of persistence. I have called this study “90203” since its launch, and this is a code number based on the old Cancer and Leukemia Group B nomenclature, the “02” in 90203 is an indicator of the year that the study was conceived and began development. Yes, that's 2002, before docetaxel was approved. The study launched in 2004 or thereabouts and it completed its accrual over the next 8 years or so, with survival being monitored ever since. As the data emerge it will be important to know that, despite the high risk requirement, most of the men enrolled in the study are still alive. Discussing the use of chemotherapy prior to radical prostatectomy (not to mention going on ADT) takes patience and an understanding of the potential benefits in terms of systemic control, downsizing, etc, as mentioned above. It would always be easier to put a patient on the OR schedule for next week and just proceed with surgery, but in this case, dozens of urologists and medical oncologists took the time to explain the high-risk nature of this cancer, the reality that there is a high failure rate in radical prostatectomy and that a study such as this would be required to demonstrate, positively or negatively, the contribution of early ADT plus docetaxel.

There’s a lot more to come from this study. We don’t yet know that p0 rate, and how that contributed to the outcome. We don’t have data yet on surgical margins, extracapsular extension or subgroup analysis of who did vs did not benefit.

Further, many events happened which resulted in censoring a particular patient's outcome. Censoring is when the patient is not “counted” as a data point due to some aberration or because no such event has occurred. The primary endpoint of the study was biochemical (PSA) relapse at three years. If a patient took ADT and salvage XRT when his PSA hit 0.1ng/dL before three years he was censored, and not counted in the primary analysis – and surprisingly that was the case with over 200 patients!

Nevertheless, it suggests that the march of docetaxel, that began in the 1990s and has moved ever “leftward” in the progression of disease may have now landed at a point where we can consider this in our highest of risk individuals. The big question is will people believe these data? I do. Will urologists start routinely referring their patients for chemotherapy prior to prostatectomy? We’ll see. I look forward to further data, and the conversation that those of us in the field need to be having about these data.

Written by: Charles Ryan, MD, B.J. Kennedy Chair in Clinical Medical Oncology, Director and Professor of Medicine, Division of Hematology, Oncology and Transplantation, University of Minnesota, Minneapolis, Minnesota

Published Date: December 16th, 2019