Higher RT Dose Lowers Failure Rate in Early Prostate Ca

ATLANTA—With the use of highly conformal radiation therapy (RT), men can safely receive a high dose of radiation for early-stage prostate cancer, Anthony L. Zietman, MD, reported at the 46th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO abstract 4). Compared with a conventional radiation dose, the high dose was associated with a lower rate of treatment failure.

"This trial was written in 1995, and the background at that time was the mounting evidence that conventional-dose radiation was associated with high rates of PSA failure and of locally persistent cancer, even in those patients with early-stage disease," said Dr. Zietman, professor of radiation oncology, Harvard Medical School, and a radiation oncologist at Massachusetts General Hospital. "There was also mounting phase II evidence that higher doses delivered by external beam could improve cancer outcome, but there were valid concerns that higher doses could be associated with higher morbidity."

Men were eligible for the randomized trial, PROG 95-09, if they had prostate adenocarcinoma of any grade with a T stage of 1b to 2b, and a PSA concentration of 15 ng/mL or lower. Men were ineligible if they had received prior hormonal therapy, radiation therapy, or chemotherapy, and if they had nodal involvement or distant metastases.

Patients in the conventional-dose arm received a proton boost of 19.8 gray equivalent (GyE), whereas those in the high-dose arm received a proton boost of 28.8 GyE. "The CTV [clinical target volume] was the prostate and a 5-mm margin. It is prescribed as gray equivalent after correction for an RBE [relative biological effectiveness] of 1.1, and patients were treated at 1.8 GyE per fraction, prescribed to isodose," he said.

Patients were positioned in the lithotomy or supine position, depending on the treatment center, and a rectal balloon was used daily to stabilize the prostate and displace the posterior wall of the rectum. After the boost, patients in both arms received the same 3D conformal photon therapy (50.4 Gy at 1.8 Gy per fraction). "The CTV was the prostate, seminal vesicles, and a 1-cm margin, again prescribed to isodose," Dr. Zietman said. The total doses to the prostate were 70.2 and 79.2 GyE in the conventional- and high-dose groups, respectively.

The 393 patients enrolled in the trial had a median age of about 66 years. Most had a PSA level between 4 and 10 ng/mL (74%), a Gleason score of 6 or lower (75%), and a T stage of 1c (61%). According to contemporary risk categories, 58% had low-risk disease.

Study Results

With a median follow-up of 5.5 years, the median time to PSA nadir after radiation was longer in the high-dose arm than in the conventional-dose arm (40 vs 28 months). "It is longer not because the PSA is falling slower, but because the PSA is falling lower," Dr. Zietman noted. The percentage of men with a PSA nadir of less than 0.5 ng/mL was significantly greater in the high-dose group (60% vs 45%), he said.

The 5-year rate of freedom from biochemical failure according to the ASTRO definition of biochemical failure (three consecutive rises in PSA level) was significantly higher in the high-dose group (79% vs 61%).

"Many criticize the ASTRO definition of failure because it incorporates backdating, which artificially flattens the curves and creates the illusion that all the action is finished," Dr. Zietman said. "So we reanalyzed using three PSA rises but without backdating, and this really just confirms the result."

When patients were grouped by their contemporary risk category and the analysis was repeated with the ASTRO definition, high-dose radiation remained associated with significantly higher rates of freedom from biochemical failure, particularly among patients at low risk (79% vs 55%) but also among those at intermediate or high risk (78% vs 61%). "And again, if we reanalyze using ASTRO without backdating, it only emphasizes . . . that the advantage appears even greater in the low-risk patients, those who have a lower risk of distant metastases," Dr. Zietman said.

Local failure was assessed directly (from clinical progression or a positive rebiopsy) or from a surrogate indicator (a PSA level of greater than 1 ng/mL after more than 2 years): 57% of patients in the conventional-dose group had a local failure vs 34% of those in the high-dose group.

The rate of salvage androgen-deprivation therapy in the conventional-dose group was about double that in the high-dose group, but Dr. Zietman cautioned that few patients have had this outcome to date. "At this early stage, overall survival is, of course, absolutely identical," he commented.

Morbidity was assessed with RTOG scales, which are not as sensitive as current tests, Dr. Zietman noted. Rates of acute grade 3 genitourinary toxicity and gastrointestinal toxicity were low (1% or less for each arm) and similar between the conventional- and high-dose groups. Likewise, rates of late grade 3 genitourinary toxicity and gastrointestinal toxicity were low (2% or less for each arm) and similar between groups.

"We conclude that dose escalation from 70 to 79 Gy can be achieved without any increase in grade 3 acute or late morbidity using highly conformal photon-proton techniques," Dr. Zietman said. "This escalation leads to significantly lower nadir PSA values, is associated with an improvement in 5-year biochemical disease-free survival for all risk groups (but the low-risk subgroup in particular), and is accompanied by an improvement in local control as assessed by surrogate markers."

He added that longer follow-up will be needed to assess any differences in rates of freedom from distant metastases and survival.