Expectant treatment with curative intent in the prostate-specific antigen era: Triggers for definitive therapy

Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

Seminar article

Expectant treatment with curative intent in the prostate-specific antigen era: Triggers for definitive therapy Christopher A. Warlick, M.D., Ph.D., Mohamad E. Allaf, M.D., H. Ballentine Carter, M.D.* Department of Urology, The James Buchanan Brady Urological Institute, Johns Hopkins Hospital, Baltimore, MD 21287, USA

Abstract Expectant treatment with curative intent for treatment of low-risk prostate cancer faces 3 challenges in the PSA era: (1) appropriate patient selection, (2) adequate surveillance strategies, and (3) identification of triggers for definitive intervention when cure is still possible. Men 65 years or older with T1c disease, prostate-specific antigen density ⬍0.15 ng/ml/cm3, and favorable biopsy characteristics per the Epstein criteria currently appear to be the safest candidates for expectant treatment. Changes in biopsy characteristics are the most objective trigger for definitive therapy currently in use. Outcomes data are still required to determine the safety of expectant treatment for localized disease. © 2006 Elsevier Inc. All rights reserved. Keywords: Expectant management; Prostate cancer; Curative intervention

Introduction Despite the fact that prostate cancer is the most common noncutaneous cancer in men, substantial controversy regarding its treatment persists. Much of this controversy stems from the fact that in reality, “prostate cancer” may be more accurately described as a group of very heterogeneous diseases with long and uncertain natural histories. Before widespread prostate-specific antigen (PSA) screening, we became aware of prostate cancer well along in its natural history because most men were diagnosed with advanced or metastatic disease, and few men were cured with definitive surgery. Furthermore, before the improvements in surgical therapy, such as the identification of the neurovascular bundles by Walsh and Donker [1], treatment was fraught with severe morbidities such as incontinence and impotence in the majority of patients, as well as severe bleeding and a mortality rate reported as high as 5% [2– 4]. In such an environment, observation with palliative intervention only with the presentation of symptomatic disease was a common and logical treatment strategy for prostate cancer, socalled “watchful waiting.” In the current era of PSA screening, most men are diagnosed with nonpalpable clinically localized disease, approx* Corresponding author. Tel.: ⫹1-410-955-0351; fax: ⫹1-410-6143695. E-mail address: [email protected] (H.B. Carter). 1078-1439/06/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2005.07.004

imately 10 years earlier in the natural history of the disease compared to digital rectal examination (DRE) detected tumors [5,6]. This has resulted in a prostate cancer incidence that is now 49% higher than when PSA testing was not available [7], and estimates of over-diagnosis (i.e., detection of cancer that would otherwise have not been detected in the absence of screening) are between 30% and 50%, depending on age [6,8]. This result suggests that there are many men who will be diagnosed with prostate cancer who may not require treatment and would ultimately die “with prostate cancer” and not “from prostate cancer” if left untreated. Furthermore, despite the advances in the definitive treatment of prostate cancer, both surgery and radiation therapy affect quality of life, thus conservative therapy for some men is still relevant in the PSA era. The concept of “watchful waiting” as practiced in the past (i.e., observation until patients become symptomatic and then the initiation of palliative therapy) has evolved into a more proactive strategy called “expectant management with curative intent” (EMCI) or “active surveillance.” Given the opportunity to diagnose men earlier in the natural history of the disease, our current challenge is to differentiate accurately between those patients who will require definitive therapy early enough to cure them and those in whom we can safely delay or avoid the morbidities of treatment. There are 3 fundamental questions that EMCI programs must address: (1) Which patients are appropriate candidates for EMCI? (2) How will patients in the EMCI program be

52

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

followed? and (3) What will be the triggers for intervention in these patients? The answers to these questions are not currently clear and, in fact, may vary with a given subset of patients with prostate cancer. The goal of this article is to review the literature on EMCI in the PSA era with particular emphasis on the triggers for intervention.

Patient selection The ability to identify accurately men with small volume disease is critical for the implementation of EMCI. To be able to intervene and “cure” men after an initial period of observation, they must first have curable disease to begin with. Herein lays one of the most challenging issues related to EMCI. In the era before PSA, when most patients were diagnosed with locally advanced or metastatic disease, many older men were candidates for watchful waiting because their cancers were too far advanced to cure with locally directed therapy. However, in current EMCI programs, in which the goal is to cure with definitive intervention at the appropriate time if necessary, “success” will largely be determined by selecting appropriate patients with low-volume, low-risk disease. The potential criteria for selection of men for whom expectant treatment would be safe include age, stage, needle biopsy findings (grade and extent), and PSA criteria. Watchful waiting programs have typically involved older men. However, in the PSA era, with more younger men being diagnosed, EMCI is being offered to younger men as well. As such, dissimilarities in the likelihood of different aged men in expectant treatment programs to undergo treatment have been noted. Younger aged men tend to be treated with definitive therapy more often than older men. Meng et al. [9] found that men older than 75 years were less likely to receive treatment than men younger than 75 years when examining the Cancer of the Prostate Strategic Urological Research Endeavor database. Likewise, Zietman et al. [10] found that men younger than 75 years were more likely to undergo definitive treatment in their cohort of patients with a mean age of 71. In addition, men older than 75 years were more likely to undergo hormonal therapy. Similarly, ElGeneidy et al. [11] found that age ⬍75 years was predictive of curative intervention on both univariate and multivariate analysis. Furthermore, Johansson et al. [12] noted a decrease in the progression-free survival, survival without metastases, and prostate cancer-specific survival when patients with T1⫺T2 disease were followed beyond 15 years as compared to the rates observed up to 15 years, suggesting that age should be an important factor in selecting patients for expectant treatment because the length of follow-up is predictive of progression. In addition to age, comorbidities that may decrease life expectancy are also important to consider before embarking on expectant treatment. Although Albertsen et al. [13] found that comorbidities influenced the decision to pursue watch-

ful waiting, Wu et al. [14] failed to note any effect of comorbidities on secondary treatment-free survival for patients in their cohort when observed to 5 years. The differences in observations between these 2 studies may be related to the difference in the length of follow-up; the study by Wu et al. [14] looked at men to 5 years, while the study by Albertsen et al. [13] had a mean follow-up of 15.5 years. Because there is no uniformity in how EMCI programs are conducted, the predilection to treat younger, otherwise healthy men versus older men or those with significant comorbidities is likely reflective of the biases of both physicians and patients, many of whom are less comfortable withholding definitive treatment in younger healthy men who may have a longer lifespan and, thus, may be more likely to be affected clinically by their prostate cancer. Until we are able to more confidently predict who is likely to have progression and show the efficacy of the EMCI strategy, this bias seems prudent. Clinical stage is a very important criterion in selecting patients for expectant treatment. Bill-Axelson et al. [15] found in their cohort of untreated men (nearly 3/4 of whom had palpable disease) that at 10 years of follow-up, more than 44% of the untreated men had evidence of local progression, while 25% had evidence of metastases. These rates were significantly lower in patients who received surgery. Similarly, the Partin tables predict that 75% of patients diagnosed with stage T1c disease, Gleason 5⫺6, and PSA from 6⫺10 ng/ml are likely to have organ confined disease [16]. For T2a disease, Gleason 5⫺6, and PSA 6⫺10 ng/ml, organ confined status decreases to 58%, while in those patients with T2b disease, the rate decreases to 50%, suggesting that clinical stage T2 disease is significantly less likely than T1c disease to be organ confined. Thus, definitive treatment seems logical for those patients with palpable disease because surgery has proved to reduce prostate cancer death in these men. The effect of tumor grade on outcome is also profound and is, in fact, the most important aspect of biopsy characteristics. Johansson et al. [12] showed that patients with grade 3 disease (similar to Gleason sum 8⫺10) had a 56% chance of distant metastases developing compared to a 24% chance in patients with grade 2 disease (similar to Gleason sum 5⫺7). Furthermore, Albertsen et al. [13] showed in a recent study on watchful waiting with 20-year follow-up that patients with Gleason 8⫺10 disease had a mortality rate of 121/1000 patient-years; those patients with Gleason 6 disease died at one fourth that rate, with only 30 deaths per 1000 patient years. It is known that the extent of tumor found on biopsy generally correlates to the amount of tumor found at prostatectomy. Epstein et al. [17] established a set of PSA and needle biopsy findings (the Epstein criteria) that were found to be predictive of small volume disease in patients diagnosed with clinical stage T1c prostate cancer. Studies applying these criteria found that 79% of patients with PSA density ⬍0.15 ng/ml/cm3 and favorable needle biopsy char-

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

acteristics (biopsy Gleason sum ⱕ6, no more than 2 cores positive on biopsy, and no more than 50% of any single core involved with cancer) had tumor volumes ⱕ0.5 cm3 [18]. In contrast, 83% of patients with PSA density ⱖ0.15 ng/ml/ cm3 or any unfavorable biopsy characteristic (more than 2 cores or ⬎50% of any core positive for cancer, or biopsy Gleason sum ⬎6) had a tumor volume ⬎0.5cm [3,18]. This retrospective study was validated in a prospective study by Carter et al. [19], confirming the ability of these criteria to predict small volume disease. Other important criteria to examine in men considering expectant treatment as a treatment option include PSA and PSA kinetics. PSA level is related to the amount of cancer that is seen after radical prostatectomy [20]. Furthermore, Epstein [17] and Allan [21] et al. have shown that PSA density can help predict the amount of tumor found at radical prostatectomy, while Kattan et al. [22] factored in PSA level and prostate volume as separate variables in their predictive nomogram, underlining the importance of considering PSA levels in the context of the size of the gland. It is interesting that PSA doubling time (DT) ⬎2 ng/ml per year during the year before diagnosis has been associated with cancer-specific death within 5 years after surgery [23], suggesting that these men are less likely to have curable disease and, thus, would not be appropriate candidates for EMCI programs. However, there has been no established PSA kinetics threshold below which it is considered absolutely safe for men to undergo EMCI. The most appropriate patients for EMCI programs are men with low-volume, low-risk disease. The challenge is to determine pretreatment criteria that allow us to identify such men. D’Amico et al. [24] devised a system to stratify men into low, medium, and high-risk of posttreatment biochemical failure based on clinical stage, PSA level, and biopsy Gleason sum. Low-risk patients (i.e., ⬍25% biochemical failure at 5 years after therapy) were defined as men with clinical stage T1c or T2a, biopsy Gleason sum ⱕ6, and PSA ⬍10 ng/ml. Patients with a high risk of biochemical failure (i.e., ⬎50% at 5 years after therapy) were defined as having clinical stage T2c, or PSA ⬎20 ng/ml or a biopsy Gleason sum ⱖ8. Patients with a PSA between 10 and 20 ng/ml, or Gleason sum 7 on biopsy, or clinical stage T2b were at intermediate risk (25% to 50%) for biochemical failure at 5 years after therapy [24]. Thus, older men who are classified into the low-risk category would be most appropriate for EMCI programs. Similarly, it is our opinion that men older than 65 years with a PSA density of ⱕ0.15 ng/ml/cm3, with stage T1c disease, and Gleason sum ⱕ6 on biopsy who fulfill the Epstein criteria [17] after undergoing adequate sampling by at least a 12-core biopsy (with sampling of the far lateral peripheral zones) are the safest candidates for EMCI programs at this time. By applying the stringent Epstein criteria and PSA density requirements to men already in a low-risk category as defined by their clinical stage, and biopsy Gleason sum, we endeavor to select the lowest risk men from a

53

category of men with low-risk disease to begin with. As our experience with EMCI grows and we are more confidently able to identify those men unlikely to have disease progression, we may become more enthusiastic about broadening the inclusion criteria, especially age, because younger men may in fact be the patients who benefit most from EMCI and the maintenance of current quality of life.

Triggers for intervention The goal of EMCI programs is to avoid or delay treatment of men with low-risk disease and to provide definitive treatment to men at the first sign of disease progression, effectively individualizing treatment to the patient. Therefore, a fundamental challenge is to identify appropriate markers of disease progression that are detectable at a time when disease is still curable. Many different parameters such as PSA levels or PSA kinetics, changes in clinical stage as detected by DRE, and biopsy findings have been used singly or in combination to define disease progression, although without consensus. These questions and challenges have become evermore pressing as we attempt to move from a phase describing feasibility of EMCI programs, toward attempting to define the efficacy of this as a treatment strategy for prostate cancer.

Patient anxiety Several groups [9,25–27] have reported freedom from treatment as an end point, instead of or in addition to “progression” as an end point when assessing their watchful waiting programs. This process appears to be because a significant fraction of patients undergo treatment, not for any objectively documented evidence of progression (however defined), but because anxiety about their untreated disease prompts them to seek treatment. El-Geneidy et al. [11] noted that nearly half their patients from the Veterans Administration Medical Center in Portland, Oregon, requested treatment without documented evidence of progression, attesting to the significant anxiety that living with prostate cancer can produce. Similarly, Patel et al. [26] reported that 14 of 31 patients who went on to treatment in their study did so because of a significant anxiety component. It is interesting that Zietman et al [10]. reported that in response to a telephone questionnaire administered to their patients who went on to treatment, 81% of the respondents stated they did so because it was desired by their physicians; however, in only 24% of the patients was there physician documented clinical or biochemical progression. The most common objective reason observed for progression in this study was a small increase in PSA (median increase 2.9 ng/ml). These observations lead the investigators to conclude that improved communication and discussion of ex-

54

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

pectations between patient and physician regarding the significance of increasing PSA levels and treatment decisions are needed [10]. These studies suggest that assessment of the patient’s level of anxiety might be another important criterion for proper patient selection. As more data become available on the effectiveness of EMCI programs, some of this patient (and physician) anxiety may be alleviated.

PSA PSA levels or PSA kinetics have been the most common objective parameter associated with proceeding to definitive therapy in a variety of studies [9-11,14,26-29]. When examining the Department of Defense Center for Prostate Disease Research Database, Wu et al. [14] found that PSA at diagnosis was a significant predictor of receiving secondary treatment. Patel et al. [26] found that PSA levels higher than the median were predictive of secondary treatment compared to initial PSA levels less than the median. The study by Meng et al. [9] found that men with higher baseline or follow-up PSA levels were more likely to undergo active treatment. Furthermore, men who had PSA increase from 2 to 4.9 ng/ml were 2.6 times more likely to receive treatment than those with an increase of less than 2 ng/ml. Those patients with an increase of ⱖ5 ng/ml were more than 3 times as likely to receive treatment [9]. These results from the Cancer of the Prostate Strategic Urological Research Endeavor database may represent common practice today, with most patients in this cohort older than 75 years and receiving hormonal therapy when active treatment was initiated. The phenomenon of changes in PSA prompting therapy was also observed by Zietman et al. [10] In this study, the changes in PSA that prompted therapy were relatively small. Median PSA increase in those patients receiving therapy was only 2.9 ng/ml compared to 0.9 ng/ml in those who did not receive therapy. This change in PSA was documented as the reason for intervention in 71% of the cases for which documentation existed. The levels of PSA changes used as triggers in these studies are arbitrary and not evidence based.

PSA DT Several studies have now shown a strong correlation between PSA DT and progression to treatment in expectant management studies [11,27-29]. In an early study, McLaren et al. [28] reported that on multivariate analysis, PSA DT strongly correlated with clinical progression, stage progression, and time to treatment. In fact, 50% of patients with PSA DT ⬍18 months had progression within 6 months. Progression was defined in this study as an increase in palpable disease on DRE or an increase in T stage, whereas changes in PSA or PSA kinetics were not used to define

progression. The study cohort included men with moderate and high-risk disease (stage T1a⫺T3c), which may explain why PSA kinetics was predictive of progression. Although PSA DT was not correlated with clinical stage or tumor grade, it did correlate with initial PSA values [28]. An interesting analysis performed in this study looked at PSA values excluding those obtained between 2 and 12 months before the date of progression to determine if progression could be predicted at this early time. However, PSA DT calculated from these early times failed to predict future progression [28]. The study by Carter et al. [27] also found that PSA DT was the most significant factor associated with secondary treatment. Of these patients, 22% had a PSA DT of ⬍2 years, and another 17.6% had PSA DT between 2 and 5 years. Of the men in this subgroup, 81% went on to definitive treatment [27]. Despite rigid inclusion criteria for the analysis (i.e., biopsy Gleason sum ⱕ6 with no pattern 4, ⱕ3 positive biopsy cores, clinical stage ⱕT2, PSA ⱕ20), the significant number of men observed to have fast PSA DT prompted the investigators to conclude that this may reflect men in the cohort with occult higher grade cancer, who would not be suitable candidates for expectant treatment to begin with. This point again emphasizes the use of confirmational biopsy for preventing understaging because only 24.6% of the patients in this study underwent a repeat biopsy. The studies reported by El-Geneidy [11] and Panagiotou [29] et al., involving the same cohort of patients, both looked at multivariate models with and without PSA DT as a variable to predict progression to treatment. Age and percent of positive biopsies ⱖ34% were the most significant predictors of progression to definitive therapy in the study by El-Geneidy et al. [11] when PSA DT was not included in the model. Similarly, Panagiotou et al. [29] found the percent of positive biopsies between 34% and 50%, and Gleason score 8⫺10 to be predictive. When PSA DT was added back into the models, it became the most significant predictor in the study by Panagiotou et al. [29], while age remained the most significant predictor (PSA DT was also correlated) in the study by El-Geneidy [11]. Using PSA DT and age subgroups, El-Geneidy et al. [11] stratified risk groups with further analyses. Patients with PSA DT ⬍3 years were at highest risk, while men with PSA DT 3 years or longer and age 65 years or younger, 66⫺74 years, and ⱖ75 years defined groups of decreasing risk for proceeding to definitive therapy. PSA DT helped to stratify patients with Gleason 7 disease in the study by Panagiotou et al. [29], so that patients with Gleason ⱕ7 and PSA DT ⬍3 years were at significantly lower risk for progression to treatment than patients with Gleason ⱕ7 and PSA DT ⬎3 years. In contrast to the aforementioned studies, Patel et al. [26] failed to note a correlation between PSA DT higher than versus less than the median and progression on univariate analysis. In fact, 46% of the patients in the cohort had a

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

negative PSA DT. The patients in this cohort had stage T1⫺T2 disease, a mean age of 65.3 years, and all but 8 patients had only 1 positive core on initial biopsy, with a median number of 6 cores taken. Of these patients, 78% had stage T1c disease or lower. The investigators suggest the failure to see a correlation between PSA DT and progression in this cohort was because of the limited impact of these low volume cancers on serum PSA [26]. Two prospective studies of expectant treatment cohorts have looked at PSA and its derivatives in men who have progression and those who did not have progression to treatment [25,30]. Carter et al. [30] found that initial PSA, PSA at progression, and PSA velocity were no different between the men who had progression and those who did not. (Progression was defined by the presence of Gleason 7, ⬎2 cores positive, or ⬎50% of any core involved with cancer on annual surveillance biopsy.) There was a statistically significant difference noted in percent-free PSA and PSA density between patients who had progression and those who did not [30], despite PSA density being used as a selection criterion for the study. However, the overlap in percent-free PSA between those patients who did and those who did not have progression suggested limited usefulness for free PSA in selecting appropriate candidates for expectant treatment [30]. Khan et al. [31] examined further whether PSA derivatives could be used to predict progression in an expanded cohort of patients, including those used in the study by Carter et al. [30]. The investigators found that PSA velocity, percent-free PSA, and gland volume did improve the predictive capacity to determine which men might safely continue with expectant treatment and those who were more likely to have unfavorable biopsy characteristics on their next follow-up biopsy. By using such biomarkers in combination, it may be possible to better predict future progression and avoid annual surveillance biopsies in some men. Choo et al. [25] used PSA parameters as one definition of progression in their cohort. They defined biochemical progression as PSA DT ⬍2 years, and final PSA ⬎8 ng/ml and P ⬍ 0.05 on regression analysis of log PSA on time. Of 69 patients who had progression, 16 met this definition, second in frequency only to patient request as a reason for active therapy. PSA DT failed to correlate with any of the baseline variables in this study, including age, initial PSA, clinical stage, or Gleason score [25]. Median PSA DT in this study was 6.68 years, higher than any previous reports. In fact, 42% of the patients had PSA DT ⱖ10 years, reflective, perhaps, of the indolent nature of many of the tumors in this series. The use of PSA and its derivatives to predict progression may be relatively limited in the context of prospective trials, such as those by Carter [30] and Choo [25] et al. and the data presented previously by Patel et al. [26] because the participants in these programs are a much more homogeneous group than those typically reported in the retrospective studies discussed previously. Furthermore, with more

55

restrictive criteria for inclusion in EMCI programs, patients are likely to have very low-volume, low-risk disease and would not be expected to have large or rapid changes in PSA. These parameters, particularly PSA DT, will likely prove to be more useful in watchful waiting programs that recruit men with more advanced disease (moderate and high risk) as compared to those programs populated by low-risk disease. It is currently unknown whether the use of PSA kinetic thresholds will allow intervention at a time when cure is still possible. In addition, the introduction of new biomarkers with increased discriminatory powers or accurate imaging for localized prostate cancer may help to stage better these patients initially as well as provide better surveillance in the future.

Clinical stage The data regarding the correlation between clinical stage and progression are conflicting. In the study by Carter et al. [27], which consisted of patients with clinical stage T1a⫺T2c disease (one third of which were stage T2), clinical stage was predictive of progression to treatment in both univariate and multivariate analyses. Similarly, Wu et al. [14] found clinical stage predictive of progression to treatment on multivariate analysis in their cohort of 8390 patients from the Department of Defense Center for Prostate Disease Research Database consisting of men with T1⫺T4 disease, roughly 10% of which had T3 and T4 disease. Choo et al. [25] found that the actuarial progression rate approached but did not attain statistical significance between T2 and T1 (evenly represented in the cohort) disease in their cohort of patients with T1⫺T2b disease. A nearly statistically significant difference was noted by Patel et al. [26] when comparing T1a versus all other stages by univariate analysis of their patients with T1a⫺T2c disease, of which 22% had T2 disease. The studies of Meng [9], El-Geneidy [11], and McLaren [28] et al. all failed to show a correlation between clinical stage and progression to treatment. These studies comprised patients with T1⫺T3 disease [9,28] or T1⫺T2 disease [11]. It is unclear why the effect of clinical stage on the risk of progression is conflicting, but this may reflect the differences in selection of candidates for expectant treatment programs.

Prostate needle biopsy findings The institutionally approved expectant treatment program at the Johns Hopkins Hospital requires participants to undergo a semiannual history, physical examination, including DRE, a percent-free and total PSA, with annual surveillance biopsies of at least 12 cores, including sampling of the far lateral peripheral zones, the importance of which has been recently reported [32]. Changes in PSA

56

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

may prompt a biopsy sooner than the planned annual biopsy but are not used in and of themselves as triggers for progression to definitive therapy. Decisions to intervene with curative intent are based solely on the appearance of adverse characteristics on needle biopsy (high grade cancer, or more than 2 cores with cancer, or more than 50% involvement of any core with cancer), as defined by the Epstein criteria [17]. The reported 31% progression rate based on biopsy findings has been attributed to underestimating disease extent in some men admitted to the program early on who may have only undergone a sextant biopsy and/or did not undergo a repeat (confirmation) biopsy before enrollment. In addition, the absence of cancer on follow-up biopsies has been shown to predict absence of future disease progression. Carter et al. [30] showed that 56% (22 of 39) of men who had cancer on their follow-up biopsy eventually had progression compared to 2% (3 of 42) of men with 1 or more negative follow-up biopsies. This observation was also noted by Patel et al. [26], who reported that a positive second biopsy was the most significant factor for progression in their cohort. Of patients who had a first repeat biopsy negative for tumor, 11% (5 of 43) progressed compared to 40% (12 of 27) of patients with a positive first repeat biopsy. Similarly, Khatami et al. [33] found in their case-control study that of 12 patients who underwent a second biopsy (after initial sextant biopsy), none had an increase in Gleason sum, but 10 of 12 had more cancer on the second biopsy than their initial biopsy, as measured by the length of cancer in the biopsy core (mean 1.6 mm at initial biopsy vs. mean 4.2 mm on second biopsy). The percentage of patients with cancer on only 1 core decreased from 75% on initial biopsy to only 19% on second biopsy, again supporting the importance of adequate initial sampling of the prostate with at least 12 cores, as well as confirmational biopsies to minimize the underestimation of disease extent. El-Geneidy [11] and Panagiotou [29] et al. both found biopsy information to be predictive of progression. These investigators found that percent-positive biopsy cores ⱖ34% and between 34% and 50% versus less than 34%, respectively, to be correlated to progression in their cohort. Assuming adequate initial sampling of the prostate, biopsy data are the most direct and objective methods available to document progression of disease and allow intervention, presumably when cure is still possible. Other criteria, such as DRE findings and PSA kinetics, are subject to either interobserver variability or biologic variability that may not accurately reflect disease status.

Outcomes The most important question regarding EMCI as a treatment option is whether delay of definitive treatment diminishes the chance for cure in men who eventually have progression and undergo definitive therapy. Clearly, the

success of this strategy relies, to a large extent, on appropriately select patients. To date, very little outcomes data are available for EMCI programs in the PSA era. Carter et al. [30] reported that of the 13 men who had progression and underwent radical prostatectomy in their cohort, 12 (92%) had curable disease as defined by a ⬎70% chance of being biochemical recurrence free at 10 years (pathologic T2N0 disease, with a radical retropubic prostatectomy [RRP] Gleason sum of ⱕ7 [3 ⫹ 4], and negative surgical margins or T3aN0 disease and a RRP Gleason sum of ⱕ6 [3 ⫹ 3] with negative seminal vesicles, and negative surgical margins) [34]. Khatami et al. [33] published a case-control study in Sweden to determine if initial surveillance reduces the chance of cure by radical prostatectomy. A total of 26 patients with T1c⫺T2 disease, Gleason sum ⬍7, and PSA 3⫺13 ng/ml were treated by initial surveillance for a mean of 23.4 months (range 8 –55) before undergoing radical prostatectomy. Two controls were picked per case, matching for PSA, T stage, and biopsy Gleason score. There were no significant differences in tumor size, frequency of extraprostatic extension, Gleason sum, or time to progression. However, the follow-up in these patients was only 2 years, which is too soon to draw firm conclusions about cancer control. Similarly, Patel et al. [26] evaluated the 17 men in their cohort who underwent RRP after initial surveillance and found no evidence of clinical or biochemical recurrence in these men at a median follow-up of 15.2 months. One patient had seminal vesicle involvement and positive margins, and 2 had extracapsular extension but negative margins. Thus, although initial data are encouraging, a definitive answer as to the long-term safety and efficacy of this approach is not yet available.

Conclusion With the aging of the population and the continued use of PSA screening, we can anticipate an increase in prostate cancer diagnoses. A significant fraction of these cancers will be indolent cancers that may not become clinically relevant or require treatment during the lifetime of the host. There will also be another subset of men who will have aggressive tumors that will require definitive treatment to prevent sequelae from the disease. The changing paradigm from watchful waiting with palliative intervention to EMCI challenges us to find new ways to discriminate between these 2 populations. Another challenge will be to develop surveillance programs that provide the most thorough assessment of cancer burden in the least invasive and least expensive manner possible, while identifying disease progression when it is still curable. Advances in imaging and the development of new biomarkers may improve both selection and surveillance of men treated expectantly in the future.

C.A. Warlick et al. / Urologic Oncology: Seminars and Original Investigations 24 (2006) 51–57

References [1] Walsh PC, Donker PJ. Impotence following radical prostatectomy: Insight into etiology and prevention. J Urol 1982;128:492–7. [2] The Veterans Administration Co-operative Urological Research Group. Treatment and survival of patients with cancer of the prostate. Surg Gynecol Obstet 1967;124:1011–7. [3] Creevy CD. The mortality of transurethral prostatic resection. J Urol 1951;65:976 – 82. [4] Sakati IA, Marshall VF. Postoperative fatalities in urology. Trans Am Assoc Genitourin Surg 1965;57:132– 40. [5] Tornblom M, Eriksson H, Franzen S, et al. Lead time associated with screening for prostate cancer. Int J Cancer 2004;108:122–9. [6] Draisma G, Boer R, Otto SJ, et al. Lead times and overdetection due to prostate-specific antigen screening: estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 2003;95:868 –78. [7] Stephenson RA. Prostate cancer trends in the era of prostate-specific antigen. An update of incidence, mortality, and clinical factors from the SEER database. Urol Clin North Am 2002;29:173– 81. [8] Etzioni R, Penson DF, Legler JM, et al. Overdiagnosis due to prostate-specific antigen screening: Lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst 2002;94:981–90. [9] Meng MV, Elkin EP, Harlan SR, et al. Predictors of treatment after initial surveillance in men with prostate cancer: Results from CaPSURE. J Urol 2003;170:2279 – 83. [10] Zietman AL, Thakral H, Wilson L, et al. Conservative management of prostate cancer in the prostate specific antigen era: The incidence and time course of subsequent therapy. J Urol 2001;166:1702– 6. [11] El-Geneidy M, Garzotto M, Panagiotou I, et al. Delayed therapy with curative intent in a contemporary prostate cancer watchful-waiting cohort. BJU Int 2004;93:510 –5. [12] Johansson JE, Andren O, Andersson SO, et al. Natural history of early, localized prostate cancer. JAMA 2004;291:2713–9. [13] Albertsen PC, Fryback DG, Storer BE, et al. Long-term survival among men with conservatively treated localized prostate cancer. JAMA 1995;274:626 –31. [14] Wu H, Sun L, Moul JW, et al. Watchful waiting and factors predictive of secondary treatment of localized prostate cancer. J Urol 2004;171: 1111– 6. [15] Bill-Axelson A, Holmberg L, Ruutu M, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med 2005;352:1977– 84. [16] Partin AW, Mangold LA, Lamm DM, et al. Contemporary update of prostate cancer staging nomograms (Partin Tables) for the new millennium. Urology 2001;58:843– 8. [17] Epstein JI, Walsh PC, Carmichael M, et al. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994;271:368 –74. [18] Partin AW, Yoo J, Carter HB, et al. The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 1993;150:110 – 4.

57

[19] Carter HB, Sauvageot J, Walsh PC, et al. Prospective evaluation of men with stage T1C adenocarcinoma of the prostate. J Urol 1997; 157:2206 –9. [20] Stamey TA, Kabalin JN, McNeal JE, et al. Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. II. Radical prostatectomy treated patients. J Urol 1989;141:1076 – 83. [21] Allan RW, Sanderson H, Epstein JI. Correlation of minute (0.5 MM or less) focus of prostate adenocarcinoma on needle biopsy with radical prostatectomy specimen: role of prostate specific antigen density. J Urol 2003;170:370 –2. [22] Kattan MW, Eastham JA, Wheeler TM, et al. Counseling men with prostate cancer: A nomogram for predicting the presence of small, moderately differentiated, confined tumors. J Urol 2003;170:1792–7. [23] D’Amico AV, Chen MH, Roehl KA, et al. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med 2004;351:125–35. [24] D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969 –74. [25] Choo R, Klotz L, Danjoux C, et al. Feasibility study: Watchful waiting for localized low to intermediate grade prostate carcinoma with selective delayed intervention based on prostate specific antigen, histological and/or clinical progression. J Urol 2002;167:1664 –9. [26] Patel MI, DeConcini DT, Lopez-Corona E, et al. An analysis of men with clinically localized prostate cancer who deferred definitive therapy. J Urol 2004;171:1520 – 4. [27] Carter CA, Donahue T, Sun L, et al. Temporarily deferred therapy (watchful waiting) for men younger than 70 years and with low-risk localized prostate cancer in the prostate-specific antigen era. J Clin Oncol 2003;21:4001– 8. [28] McLaren DB, McKenzie M, Duncan G, et al. Watchful waiting or watchful progression? Prostate specific antigen doubling times and clinical behavior in patients with early untreated prostate carcinoma. Cancer 1998;82:342– 8. [29] Panagiotou I, Beer TM, Hsieh YC, et al. Predictors of delayed therapy after expectant management for localized prostate cancer in the era of prostate-specific antigen. Oncology 2004;67:194 –202. [30] Carter HB, Walsh PC, Landis P, et al. Expectant management of nonpalpable prostate cancer with curative intent: preliminary results. J Urol 2002;167:1231– 4. [31] Khan MA, Carter HB, Epstein JI, et al. Can prostate specific antigen derivatives and pathological parameters predict significant change in expectant management criteria for prostate cancer? J Urol 2003;170: 2274 – 8. [32] Epstein JI, Walsh PC, Carter HB. Importance of posterolateral needle biopsies in the detection of prostate cancer. Urology 2001;57:1112– 6. [33] Khatami A, Damber JE, Lodding P, et al. Does initial surveillance in early prostate cancer reduce the chance of cure by radical prostatectomy?–A case control study. Scand J Urol Nephrol 2003;37:213–7. [34] Epstein JI, Partin AW, Sauvageot J, et al. Prediction of progression following radical prostatectomy. A multivariate analysis of 721 men with long-term follow-up. Am J Surg Pathol 1996;20:286 –92.