ADT and radiation for first-line treatment of node-positive (N1) prostate cancer (STAMPEDE trial details)


In a previous commentary, we mentioned the early top-line results of the STAMPEDE trial, which demonstrated a benefit to whole pelvic radiation and ADT for treatment of high-risk prostate cancer when positive pelvic lymph nodes have been detected. We now have some additional details.

James et al. analyzed data from the control arm (Arm A) of the STAMPEDE trial. The control arm excluded patients with distant metastases and those who had previous treatment. All patients were high risk and were treated between 2005 and 2014 with a minimum of 2 years of ADT. At physician’s discretion, some were also treated with radiation therapy (RT) 6 to 9 months after the start of ADT. Patients with lymph nodes larger than 10 mm were typically staged as “node positive” (N1). Patient counts for this analysis were as follows:

  • N0 and RT — 121 patients (43 percent received whole pelvic radiation)
  • N0 and no RT — 46 patients
  • N1 and RT — 71 patients (82 percent received whole pelvic radiation)
  • N1 and no RT — 86 patients

Age, Gleason scores, and performance status were similar in all groups. Pre-treatment PSA was higher in patients who had RT, although the differences were not statistically significant. The planned radiation dose to the prostate and seminal vesicles was 74 Gy in 37 fractions or the equivalent hypofractionated dose. The planned dose to the pelvic lymph nodes was 46 to 50 Gy in 23 to 25 fractions or 55 Gy in 37 fractions. Increased doses were allowed if the physician was experienced in delivering nodal doses.

Although overall survival was measured, there was too little mortality as of this interim analysis to be worth reporting. Instead, the authors focused on 2-year failure-free survival (FFS), defined as no biochemical recurrence and no radiographically-detected progression among survivors. Patients would have been ADT-free for 12 to 15 months by that point, unless they showed early evidence of progressing.

Among the men with no detected nodal involvement ( N0):

  • The 2-year FFS was:
    • 96 percent among men who received RT
    • 73 percent among men who did not receive RT
  • Late gastrointestinal (GI) toxicity was:
    • Proctitis: Grade 2: 7 percent, Grade 3: 2 percent
    • Diarrhea: Grade 2: 3 percent, Grade 3: 1 percent
    • Rectal ulcer: Grade 3: 1 percent
  • Late genitourinary (GU) toxicity was:
    • Cystitis: Grade 2: 2 percent, Grade 3: 1 percent
    • Hematuria: Grade 2: 3 percent, Grade 3: 1 percent

Among the men with detected nodal involvement (N1):

  • The 2-yr FFS was:
    • 89 percent among men who received RT
    • 64 percent among men who did not receive RT
  • Late GI toxicity was:
    • Proctitis: Grade 2: 8 percent
    • Diarrhea: Grade 2: 6 percent
  • Late GU toxicity was:
    • Cystitis: Grade 2: 5 percent
    • Hematuria: Grade 2: 2 percent, Grade 3: 2 percent

Although this was a prospective study, patients were not randomized to receive RT or not, so there may be selection bias at work. The higher pre-treatment PSA in the patients who did not get RT suggests that they may have been considered to be too far progressed to benefit from radiation. However, the benefit of RT was maintained even after adjustment for pre=treatment PSA, age, and Gleason score.

The planned radiation dose, 74 Gy, is lower than the 80 Gy now considered to be curative. The dose delivered to the pelvic lymph nodes is still within the standard of care. Although almost half of those with no nodal involvement were treated with whole pelvic RT, there was no analysis of benefit in that subgroup.

RT clearly delayed the time to relapse among high-risk patients, regardless of nodal status. The FFS curves continued to diverge after 2 years, indicating a lasting effect of treatment, at least out to 5 years post-treatment. Long-term toxicity was low among all patients who received RT.

Subject to the above caveat on selection bias, this early analysis indicates that men with high-risk prostate cancer, whether they had detected nodal involvement or not, benefited from long-term ADT + RT. As there was little long-term toxicity attached to this decision, there seems little reason to withhold such treatment.

The questions mentioned in our earlier commentary continue to be important:

  • What is the most appropriate radiation dose?
  • Is there a limit to the number of infected nodes beyond which it is fruitless to use RT?
  • Should simultaneous integrated boost RT be used on infected nodes?
  • Can SBRT equal or improve the risk/benefit profile over IMRT?
  • What is the best timing for neoadjuvant/concurrent/adjuvant ADT?
  • Can outcomes be improved with docetaxel?
  • Can outcomes be improved with immunotherapy?
  • Is whole pelvic RT or extended pelvic lymph node dissection (ePLND) more effective?
  • Can staging be improved with new imaging techniques?
  • What are the patient risk factors that affect oncological control and toxicity?
  • How much of the improved survival is a delay due to cytoreduction, and how much is actual cure?

Editorial note: This commentary was written for The “New” Prostate Cancer InfoLink by Allen Edel.

One Response

  1. Another question is the length and type of ADT. I recall research indicating that 2 years of ADT is better for high-risk patients than 6 months, but there is a research gap for the zone > 6 months and < 2 years. My own decision for 18 months (my fourth round of ADT3) was aided by an expert's opinion that triple ADT brought something extra to the table and so was better than routine ADT for 18 months (and maybe better than routine ADT for 24 months). But at this point we must just place our bets and hope.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: