The Oncologist, Vol. 10, No. 10, 799-805, November 2005; doi:10.1634/theoncologist.10-10-799 © 2005 AlphaMed Press
When Prostate Brachytherapy Fails: A Case Report and Discussiona University of California, Berkeley, California, USA, and Harvard Medical School, Boston, Massachusetts, USA; b Department of Radiation Oncology, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts, USA; c Department of Radiation Oncology, Beth Israel/Deaconess Medical Center, Harvard Medical School, and Boston Veterans Health Affairs Medical Centers, Boston, Massachusetts, USA Key Words. Prostate cancer • Brachytherapy • Prostate-specific antigen failure • Local recurrence Correspondence: Richard H. Matthews, M.D., Ph.D., Beth Israel/Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave., Boston, MA 02215, USA. Telephone: 617-232-9500, ext. 4457 or 5628; Fax: 617-278-4478; e-mail: rhmatthews{at}comcast.net Received March 1, 2005; accepted for publication September 7, 2005.
For appropriately selected brachytherapy patients, prostate-specific antigen failure is uncommon. Our patient experienced biochemical failure after 125I brachytherapy treatment for low-risk prostate adenocarcinoma. We suggest neoadjuvant/adjuvant hormonal therapy combined with pelvic external-beam radiation therapy as a reasonable salvage treatment. At the 2-year follow-up, he is apparently doing well. With limited data available, salvage management for this situation is presently investigational.
Prostate cancer affects more than 180,000 men each year and is the second-leading cause of cancer death in men. Approximately 85% of all cases diagnosed yearly in the modern prostate-specific antigen (PSA) era present as clinically localized prostate cancer, treated with curative intent by means of external-beam radiation therapy (XRT), brachytherapy, or radical prostatectomy. Transperineal ultrasound-guided prostate brachytherapy is a viable option in the management of early stage cancer of the prostate gland, targeting high radiation doses to the prostate alone or, in some techniques, the prostate and adjacent portion of the seminal vesicles. Prostate low-dose rate 125I brachytherapy, first reported by Grimm et al. [1], used to treat probable organ-confined disease has been a treatment of choice; the concept of delivering a high dose to the prostate itself with rapid dose fall-off in surrounding normal tissues should presumably diminish complications. Beyer [2] has suggested that this is the case with significant complications occurring in <5% of a cohort of 734 prostate implant patients with long-term follow-up. In clinically low-risk patients, the 125I monotherapy reportedly achieves an 87% rate of biochemical and clinical control, as demonstrated by Grimm et al. [1]. 125I implantation offers an advantage over radical prostatectomy in that it is a less invasive outpatient procedure, has a higher rate of sexual potency preservation, and causes less acute morbidity. In terms of long-term clinical local control and freedom from biochemical failure, the results of conformal brachytherapy are as favorable as the most successful radical prostatectomy data, despite the differences in patient selection criteria. However, some comparative 5-year follow-up data indicated that PSA failure was increased for intermediate- and high-risk patients when compared with XRT, which gains an advantage in treating surrounding tissues when there is a need for their treatment [3]. Causes for local failure after brachytherapy are speculative but might likely be attributed to difficulties encountered in achieving a geometrically appropriate distribution of seeds to achieve a satisfactory dose distribution within the prostate, or to inappropriate patient selection for the procedure in a monotherapy management plan [4]. Patients considered for interstitial 125I brachytherapy are generally those classified as low risk, more specifically, those with stage T1cT2a (tumor confined to the prostate gland), PSA less than 10 ng/ml, and Gleason score of six or less. In addition, the patient should not present with any significant obstructive symptoms or perineural/extracapsular extension, the prostate volume should be less than 60 ml, and three or fewer of the six sextant biopsies should be positive. The brachytherapy approach effectively treats the prostate gland only, or perhaps a minimal margin of periprostatic tissue, and should be selected only for those patients at low risk for direct extracapsular extension of disease or microscopic seeding of the pelvic lymph nodes. Patients selected to be at low risk for extraprostatic disease treated with appropriate attention to the technical details should uncommonly experience treatment failure (a review of 400 cases treated at Massachusetts General Hospital [Boston, MA] showed only two failures; DAmico, unpublished data). Locally recurrent cancer of the prostate may lead to significant morbidity, metastasis, and death, as therapeutic options with curative intent are limited at this point. According to Grimm et al. [1], modern techniques of 125I brachytherapy achieved an 87% rate of biochemical and clinical control in low-risk patients and 100% cause-specific survival at 10 years. Johnstone notes that the question of an appropriate response to apparent local treatment failure draws variable and diverse opinions with a limited basis in evidence-based medicine [5]. In fact, PSA recurrence after treatment by radiation in general predicts for subsequent clinical failure and cause-specific death in 10 years, with the spread for biochemically controlled and biochemically failing patients depending somewhat upon the definition of biochemical failure. Classifying failure as an increase to at least 3 ng/ml PSA above the nadir leads to a spread in 10-year cause-specific death rates of 49% [6]. Here, we report the case of a patient from the minority group of implant patients who suffered PSA failure after brachytherapy treatment, discuss the salvage management offered the patient, and offer some of our thoughts on this problem area.
The patient was a 64-year-old man who presented at a screening clinic on September 21, 1998, with no palpable nodule but a PSA of 4.85. This led to a biopsy on December 18, 1998, yielding a Gleason score of 3 + 3 = 6 in 5% of the tissue on the right side and 30% of the tissue on the left side of the prostate; thus, a T1c adenocarcinoma of the prostate. No persistent new bone pains were reported. He reported an International Prostate Symptom Score (IPSS) of 7/35, with mild urinary frequency and nocturia one or two times per evening. The patient also had a history of hypertension treated with amlodipine 10 mg daily and gouty arthritis treated with allopurinol 300 mg daily. The patients brother had been diagnosed with prostate cancer and had undergone radical retropubic prostatectomy. A physical exam confirmed the absence of prostate nodularity and was otherwise unremarkable. The patient decided against radical surgery and opted for prostate 125I brachytherapy, which appeared to be appropriate, as his PSA was less than 10, his Gleason score six or less, and his stage identified as T1c, therefore classifying him as low risk. 125I brachytherapy was completed on April 16, 1999, delivering a peripheral dose of 144 Gy. The technique in use at the time involved a preplanning ultrasound procedure in the Radiation Therapy Department rather than an intraoperative planning procedure, and the stabilization equipment then in use may have been less ideal than equipment that would subsequently be introduced. Approximately 1 month later, the patient reported residual urinary obstructive symptoms and, in follow-up, reported an IPSS of 10/35. Terazosin 2 mg was prescribed to improve bladder function. No hematuria, incontinence, erectile dysfunction, serious diarrhea, or hematochezia was reported in immediate follow-up. From 2000 to 2002, a gradual increase in PSA ensued; doubling time appeared to be approximately 1 year. The PSA pattern was indicative of biochemical failure (American Society for Therapeutic Radiation and Oncology [ASTRO] definition). From late 2002 through April 2003, the rise in PSA accelerated and doubling time appeared to be approximately 6 months. This prompted a re-evaluation of management options. A bone scan in February 2003 was negative for metastatic disease. In April 2003 (4 years after the date of implant), a biopsy was positive for locally recurrent adenocarcinoma of the prostate. One might guess that failure was local-regional in nature with a negative metastatic work-up, but this did help confirm the presence of disease recurrent in the prostate. It also gave us some indications of the degree of aggressiveness of the cancer at this time. The patient had perineural invasion and a Gleason score of 3 + 3 = 6 in 2/3 cores from the right side of the gland, with about 25% of tissue examined involved. An IPSS of 8/35 was reported along with moderate symptoms, including nocturia twice. Nilutamide 150 mg was administered to block the effects of testosterone at the androgen receptor level, as prostate cancer usually responds to androgen ablation. At this time, pelvic radiation was considered using a moderated dose of XRT after hormonal therapy (Nilutamide) had been ongoing for approximately 2 months. The patient consented to this management option and received 4,500 cGy of XRT in 180-cGy daily fractions of 18-MV photons delivered by a shaped four-field technique to a moderate pelvic field. (Tumor dose was calculated at an isodose line encompassing the target area, the 95% isodose line being chosen. This would then make the stated dose comparable to 4748 Gy at the isocenter.)
Four months after pelvic XRT combined with neoadjuvant/adjuvant hormonal therapy completed in response to the brachytherapy failure, the patient reported a 6/35 IPSS, indicated no new focal bone pains, and was otherwise unremarkable. PSA was 0.03, continuing his declining pattern (Table 1
There is a basic importance to gaining control of prostate cancer in the local-regional area for any patient fundamentally healthy enough to receive treatment. Local persistence of disease will be expected to lead to a later wave of metastases [57]. The importance of local-regional disease eradication is further underscored by available dose-response data for 125I brachytherapy implant [8], XRT [9], or pelvic XRT combined with brachytherapy as a boost [10]. Brachytherapy treats only the prostate gland effectively. If patients are selected to have a low risk of disease extension outside of the gland and treated by optimum techniques, failures will be infrequent [1]. Due to the infrequent occurrence of unfavorable outcome and undefined treatment standards, there are limited data to guide us directly when a brachytherapy failure does occur, and it will be necessary to draw inferences from several directions to arrive at a reasonable salvage treatment regimen. In addition to assessing the probability of occult distant metastasis at the time of biochemical failure (mainly PSA doubling time and restaging by physical exam, bone scan, and computed tomography), the patients predicted survival and quality of life should be considered the major factors when deciding on a salvage treatment approach. The precise point in time at which treatment failure has occurred is an interesting question, as well as the nature or locus of the failure. Kuban et al. [11] have pointed out that the ASTRO consensus definition of failure as three successive PSA rises independent of the nadir value is based upon external-beam irradiation data only, and is strictly valid for only that management approach, although there have been attempts to extrapolate its use to brachytherapy or even radical prostatectomy. Critz [12] has argued that for a management approach in which pelvic XRT is combined with a prostate brachytherapy boost, a nadir PSA value of 0.2 or less better defines biochemically disease-free status; according to his analysis, 99% of patients attaining this nadir level will remain disease-free after combined pelvic XRT/brachytherapy boost management. Only 16% of his patients with a nadir in the 0.31.0 range remained disease-free.
The desire to intervene quickly after the formal definition of PSA failure has been met can be tempered by some evidence that PSA rise, even with a positive local biopsy, does not inevitably lead to progressive clinical disease. Smathers and colleagues reported cases of PSA rise and positive biopsy in the first 2 years post implant followed by a decline in PSA without additional treatment [13]. In the Cavanagh et al. study, it was noted that 35.8% of patients who had been judged as biochemical successes post brachytherapy had experienced an increase of 0.2 ng/ml or more, indicating that it would be best to hold off on beginning another form of therapy following observation of PSA rise of <23 ng/ml post brachytherapy [14]. Critz et al. [15] and Stock et al. [16] have also commented on the phenomenon of transient PSA "bounce" after brachytherapy as a boost with pelvic XRT or as monotherapy, respectively. Thus, observation alone may be an appropriate option in the low-risk, asymptomatic (status post [s/p] primary treatment) patient population. The morbidity and cost of secondary therapy should be carefully assessed and compared with the apparent risk of clinical progression of prostate cancer. Our issue is, then, can we identify patients who have a need for salvage treatment, and, if so, what is a reasonable form for it to take? It is evident that PSA rise is often considered a surrogate for distant metastasis and death by the cancer; more specifically, a rise in PSA of 3 ng/ml or greater is a strong indicator that clinical failure may follow PSA failure [6]. Thames and colleagues examined biochemical failure definitions for external beamirradiated patients, relative to a "gold standard" of clinical failure and found elevations of PSA to 2 or 3 ng/ml to be superior to the ASTRO definition of three successive rises in predicting clinical failure [17]. Our patient had encountered a rise of more than 5 ng/ml at the time salvage therapy was initiated (Table 1 Grossfeld et al. [24] studied 2,336 patients in the Cancer of the Prostate Strategic Urologic Research Endeavor treated initially with radical prostatectomy (1,744) or external-beam radiation (592). Twenty-five percent of these patients received secondary cancer treatment, although the type of management varied with the type of initial treatment. Patients initially treated with prostatectomy had secondary treatment consisting of XRT as salvage in 52% of cases and androgen deprivation in 47% of cases. Ninety-two percent of the failures initially treated with XRT had androgen deprivation as their secondary management. Beyer [25] in his review notes that androgen deprivation (or in some cases, only observation) generally regarded as only palliative treatment may be appropriate for recurrence after XRT due to factors such as age and comorbid diseases, or recurrence following an indolent pattern, but then makes the case that a relatively young healthy man who has what may be a local recurrence only should be considered for potentially curative salvage treatment. He suggests brachytherapy implant as a reasonable salvage treatment after PSA failure following XRT, essentially the inverse of what we are suggesting here. Grado et al. [26], in fact, used brachytherapy as a salvage approach post external-beam PSA failure, so treating a group of 49 patients (three previous implant patients and 46 XRT only) over a period of 6 years; all had biopsy-proven local recurrence. At 5 years, they report 79% cause-specific survival and 34% biochemical disease-free survival (DFS). They point out that this rate of success at salvage is comparable with what is achieved by prostatectomy salvage, but with notably fewer complications. There was only one local failure detected after salvage brachytherapy in their series, but 27 patients developed distant metastases. The implications of PSA recurrence after prostatectomy without additional treatment have been laid out in the study by Pound et al. [27]. The median time to the development of clinically evident metastases was 8 years, and the median time to death after that was 5 years. Drawing upon the literature addressing salvage XRT versus adjuvant XRT for adverse outcome s/p radical prostatectomy, we know that delayed salvage XRT, or XRT after progression to higher PSA level, is less successful in attaining DFS than adjuvant XRT early and at low PSA levels for situations such as pathological T3 disease with positive margins in the post-prostatectomy setting [2830]. The adverse impact of a time delay could be due to seeding of the pelvic lymph nodes or the seeding of distant metastases, as to bone, in the extended time interval. If the failure involves microscopic seeding of the pelvic lymph nodes, then effective treatment of that area rather than just the prostate bed would offer a potentially curative salvage opportunity. The recently reported trial for moderately high-risk prostate cancer patients, RTOG (Radiation Therapy Oncology Group) 9413, indicated that there was an advantage in that patient population for treatment to a larger pelvic radiation field rather than a smaller periprostatic tissue volume, and that there was also an advantage to combining the large pelvic field with neoadjuvant and concomitant hormonal therapy [31], rather than subsequent adjuvant hormonal therapy. There are several significant implications to this important study. First, we may reasonably hypothesize that prostate cancer frequently follows a sequential pattern of progression: first, extension out of the gland; second, investing the pelvic lymph nodes; and, third, only subsequently spreading by hematogenous metastasis to generate bone metastases. Furthermore, it suggests a synergism for neoadjuvant hormonal therapy, making 4,500 cGy to a pelvic field a worthwhile dose of radiation and one that has a reasonable prospect for eradicating a small tumor burden in the pelvis. The synergism between XRT and neoadjuvant/adjuvant hormonal therapy is further underscored by a randomized Harvard trial of conformal radiotherapy to prostate cancer with certain risk features present (PSA at least 10, Gleason score at least 7, or radiographic evidence of extraprostatic extension) in which half the patients received XRT alone and the other half also received neoadjuvant/adjuvant hormonal therapy [32]. The latter group enjoyed significantly superior overall survival with diminished prostate cancerrelated mortality. One approach that has been suggested for locally advanced or high-risk prostate cancer at the outset is to combine pelvic radiotherapy with a brachytherapy boost, with a possible advantage to combining this with hormonal therapy also [33]. Ragde et al. noted a trend in favor of using an implant boost with pelvic XRT over implant alone in reporting an implant series at a 10-year follow-up with only 6% rate of distant metastases and 2% of deaths attributed to prostate cancer [34]. The approach taken in the present case might be thought of as a reversed boost implant, albeit with a significant time delay interspersed. Critz et al. [15] have also made it a regular practice to combine a brachytherapy procedure as a boost with pelvic XRT, including patients with poorly differentiated cancers or higher PSA levels than we would usually treat with implant as monotherapy. Various other salvage treatment strategies have occasionally been used, including such possibilities as radical prostatectomy or brachytherapy reimplantation. Amling et al. reported a Mayo Clinic (Rochester, MN) experience with salvage prostatectomy following recurrence after treatment by implant or XRT; there were 108 patients so treated in 30 years [35]. Rebiopsy was used to confirm the presence of a local recurrence in all cases. The authors suggested that the procedure has resulted in long-term DFS for some patients but "has not gained widespread acceptance due to the technical difficulty, and the potential for significant operative and postoperative complications." Bladder neck contracture rates were 21%, and incontinence rates were 50%. Most patients were treated with XRT, but the few treated by implant were not noted to be easy to approach surgically. This would seem a small number of patients to approach surgically over a 30-year time frame in a large surgical program; the authors estimate that only 2%5% of previously irradiated patients are good candidates for salvage surgery. Overall survival at 10 years was 60%, and cause-specific survival was 70%. Survival free of clinical progression was 61%, and the biochemical DFS rate was 43%. Rogers et al. reported a Baylor University (Houston, TX) series of 40 salvage prostatectomies over a span of 8 years; one value of this series is that 29 of these patients had been treated by staging lymphadenectomy and brachytherapy implant, the other 11 by external-beam irradiation only [36]. They found dissection to be more difficult in the 29 patients treated by lymphadenectomy and implant than in the 11 external-beam patients. The cancer-specific survival rate was 87% at 8 years, and DFS using PSA was 33%. Pre-treatment PSA more or less than 10 was prognostic in the two foregoing studies [35, 36]. Complications were worse than in primary surgery for prostate cancer, and 56% of the patients followed longer than 1 year were incontinent [36]. Schellhammer et al. also concluded that only 2%5% of patients experiencing recurrence after treatment by radiation methods would be considered candidates for salvage surgery; furthermore, these patients are the same ones who would have the most durable response to hormonal therapy [37]. Wallner et al. reimplanted 13 patients in 10 years with generally full-dose 125I implants; no XRT was used [38]. All 13 patients had palpable local recurrences and were then biopsy-confirmed. Freedom from local progression at 5 years was 51%, but by 6 years the rate of distant metastases had reached 100%. Two of the 13 had severe complications, and four had incontinence. Parker and Dearnaley in their review found little to recommend salvage surgery post XRT, noting difficulty of the surgical procedure, high complication rates, and poor DFS. They found little to recommend reimplantation in terms of effectiveness either [20]. Cryosurgical ablation has occasionally been used as salvage management after some type of radiotherapy failure, but in an M. D. Anderson Cancer Center (Houston, TX) study of this option, it was difficult to achieve complete ablation using salvage cryotherapy for recurrent prostate cancer (which entails freezing and completely ablating the proximal gland/tumor), and several complications occurred, such as 60% incontinence and 9% rectal fistula risks. This salvage treatment on the whole has not yielded a significant improvement in disease control [39]. A French group has recently suggested a novel salvage technique, high intensity focused ultrasound, for local failures after external-beam irradiation. The 71 patients included in their series all had biopsy-confirmed local failures prior to salvage treatment. Eighty percent then had a negative biopsy post treatment, and 61% achieved a PSA nadir <0.5 ngm/ml. Median follow-up at this time is just 14.8 months, and only 44% of the patients show no evidence of disease progression [40].
The best management of the small number of prostate brachytherapy patients encountering failure is unclear at this time. Hormonal therapy is the most commonly used secondary treatment modality but is unlikely to be curative. However, it may have value as a synergistic treatment used in a neoadjuvant/adjuvant fashion with a moderate dose of pelvic XRT, as used in RTOG 9413 [31] or the Harvard trial [32]. We attempted to use the synergism of moderate-dose pelvic XRT with neoadjuvant/adjuvant hormonal therapy and to outline what to our knowledge is an established combination treatment, but applied in a novel way as a salvage therapeutic approach that may deserve further exploration.
The authors indicate no potential conflicts of interest.
Textbook of Radiation Oncology (Leibel SA, ed., Philadelphia: W.B. Saunders Co., 2004).
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||