What Constitutes Locally Advanced PCa?

The most widespread nondermatologic malignancy in men, prostate cancer afflicts one in six American males and kills one in 34. Growing public awareness, an aging population, and increased PSA screening has led to greater detection of early-stage disease. Among newly diagnosed cases, for example, the proportion with T3 or T4 disease fell from 19% to 4% between 1988 and 1998, and those with metastatic disease at presentation fell from 14% to 3%, according to a study by Peter Carroll, MD, and colleagues at the University of California in San Francisco Comprehensive Cancer Center (J Clin Oncol. 2005;23:8146-8151). However despite improvements in screening, staging, and treatment modalities, mortality has remained high, causing 28,000 deaths in 2006. Improved definitions of locally advanced or subclinically metastatic prostate cancer and a better understanding of which forms are likely to progress are necessary to devise appropriate treatment strategies, such as adding neoadjuvant and adjuvant therapies to definitive localized therapy.

Locally advanced prostate cancer has traditionally meant the presence of extracapsular disease (stage T3 or T4) and logically has been a key variable relating to prognosis. However, with the expansion of diagnostic techniques and maturation of post-treatment cohorts in research, a more complete and complex picture has unfolded. In 1998, Anthony V. D'Amico, MD, published a landmark study in which patients were categorized as low, intermediate, or high risk based on multiple variables. Patients considered "high-risk" presented with stage T2c or higher, had PSA levels greater than 20 mg/mL, or had Gleason score (histopathologic types on biopsy) of 8 to 10 (JAMA. 1998;280:969-974). Clearly, some patients with evidence of aggressive tumor activity (either high serum PSA and/or high-grade tumor cell type) tended to progress despite low staging (i.e., intracapsular tumors, stage T1- T2).

There is little agreement in the scientific community as to what constitutes or predicts locally advanced disease. In the past 15 years, definitions have included combinations of tumor stage, Gleason score and PSA, MRI findings, nomograms, PSA velocity, PSA doubling time, and various combinations of these. At least 11 different inclusion criteria are being used in current clinical trials of "locally advanced prostate cancer." (http://www.clinicaltrials.gov). We should strive for a consensus in defining locally advanced disease. This would render future clinical research more comprehensible and allow us to better steer prostate cancer patients through their sometimes dizzying maze of therapeutic options.

Prognostic tables and nomograms

Pre- and post-treatment tables and nomograms allow for the integration of multiple clinical variables and prediction of pathological staging for prostate cancer patients. While these are useful tools to clinicians and patients for judging risk of progression and guiding treatment planning, clinicians should remember that every patient is unique. When it comes to an individual patient, odds calculations are educated guesses.

Originally based on the work of Alan W. Partin, MD, and Patrick C. Walsh, MD, the Partin Tables were most recently updated in 2001 and incorporate PSA, clinical staging, and biopsy Gleason grade to stage clinically localized disease. They are derived from data acquired from more than 5,000 patients who underwent radical prostatectomy at Johns Hopkins, and then validated by data from a similar cohort of more than 2,400 patients treated at the Mayo Clinic (www.urotoday.com). The original cohort on average had baseline pretreatment cancer stage of T1c, Gleason score of 6, and serum PSA level less than 10 ng/mL. Among those with a pre-surgical diagnosis of local organ-confined disease (64%), 36% had advanced disease postoperatively (30% with extraprostatic extension, 4% with seminal vesicle involvement, and 2% with lymph node involvement) (J Urol. 1993;150:110-114; Urology. 2001;58:843-848). Partin Tables are available at www.urology.jhu.edu/prostate/partintables.php.

Similarly, preoperative and post-operative nomograms, such as the Kattan nomograms developed at the Cleveland Clinic and Memorial Sloan-Kettering Cancer Center, offer accurate estimations of patient risk (J Clin Onc. 2005;23:7005-7012). Kattan nomograms may be found online at www.mskcc.org/mskcc/html/10088.cfm.

Biochemical studies

Assessment of PSA over time may be of greater predictive value than an isolated PSA level, making PSA velocity and PSA doubling time biochemical markers of interest. A pretreatment PSA velocity greater than 2 ng/mL/year has been shown to correlate with increased risk of locally advanced disease (J Clin Oncol. 2005;23:6157-6162) and in-creased prostate cancer-related mortality (J Urol. 2006;176:S11-S15). PSA doubling time has been used as a postoperative marker for cancer progression and risk of death (J Urol. 2005;174:2191-2196).

Implications

Tumor grade and extent of lymph node involvement inform the course of local-regional treatment and the timing of androgen ablative therapies. Lymph node invasion, however, does not always predict poor outcomes; therefore, more precise markers are needed to direct treatment and achieve cure. A re-cent study led by Edward M. Messing, MD, of the University of Rochester in New York, illustrates the point that lymph node metastasis in and of itself is not a death sentence. They showed that post-prostatectomy/pelvic lymphad- enectomy patients with positive nodes had better outcomes when given immediate androgen deprivation treatment (ADT) compared with ADT that was deferred until clinical progression (Lancet Oncol. 2006;7:472-479). Twenty-four percent of men in the deferred group never received ADT and were alive without clinical recurrence (with or without biochemical recurrence) at post-surgical year 12. Similarly, Patrick C. Walsh, MD, and his colleagues at Johns Hopkins showed that 17% of patients with positive lymph nodes who underwent radical retropubic prostatectomy and bilateral lymphadenectomy and had positive lymph nodes were progression-free at long-term follow-up without ancillary intervention (J Urol. 2004;172:1860).

Vincenzo Pagliarulo, MD, of the University of Southern California in Los Angeles, and his associates showed the prognostic importance of occult lymph node metastasis detection via immunohistochemistry and PSA among patients with pT3N0 grade tumors (J Clin Oncol. 2006;24:2735-2742). Patients with occult positive nodes had outcomes similar to those with overtly positive nodes and were at greater risk for progression and death compared with those without nodal metastases.

Basic research on genetic and molecular characterization of prostate tumor cells is underway and may lead to greater precision in patient stratification and more targeted treatments with fewer adverse effects. In the meantime, honing the definition of locally advanced prostate cancer should improve com-munication among physicians and lead to a more thoughtful approach to treatment.

Kenneth S. Koeneman, MD, is associate professor of urologic surgery, the Dougherty Family Endowed Chair in Urologic Oncology, and the Director for the Center for Prostate Cancer at the University of Minnesota in Minneapolis.