Prostate cancer diagnosis: even the experts disagree
Article Outline
The increasing acceptance of opportunistic prostate cancer screening by the community at large, coupled with the introduction of the procedure of thin core biopsy of prostate, has led to an enormous growth in the number of prostate biopsies submitted to laboratories for histologic assessment [1]. Thin core biopsy, which has the advantage of permitting a comprehensive sampling of the gland while minimizing the risk of infection does, however, presently the reporting pathologist with occasional diagnostic challenges because of the limited volume of tissue sampled in each core [2]. Although pathologists would agree that in most cases, the reporting of core biopsies, whether benign or malignant, is relatively straightforward, it is also well recognized that in a small proportion of cases, no conclusive diagnosis is possible.
The study by Egevad et al [3], reported in this issue of Human Pathology, investigated the handling and reporting habits of a large group of genitourinary pathologists when dealing with prostate biopsies. In addition to providing an insight into the practice of the so-called experts, the study sought to provide guidelines through consensus, that it was hoped would inform best practice for general anatomical pathologists. Somewhat surprisingly, rather than indicating consensus, the study discloses some diversity, both in aspects of tissue handling and in the criteria used by the respondents to diagnose malignancy.
The optimal assessment of prostate biopsies is dependent on the processing of tissue samples, and the protocols of the genitourinary pathologists surveyed showed some variation in this regard. All but one of the respondents reported that they examined more than one level from each block, with the range being from 1 to 12 sections. This result is similar to that obtained from a survey of pathologists from 130 histopathology departments in the United Kingdom, where 81% of the pathologists routinely examined 3 levels and a further 10% examined between 3 and 6 levels [4]. The importance of examining more than one level per block is well recognized as, if only one level is taken, it has been demonstrated that approximately 23% of the core length will not be sampled [5], and 3% to 8% of malignant foci will be missed [6], [7].
Despite the reported heavy usage of immunohistochemical studies by genitourinary pathologists to confirm or refute a diagnosis of malignancy, only 47% of the respondents routinely retained sections for subsequent staining if required. This figure was reduced to 29% for genitourinary pathologists practicing outside North America, which compares unfavorably with 42% of the United Kingdom laboratory pathologists who have reported that they routinely retain interval levels from serial sections of prostate cores [4]. This result is somewhat surprising because it is well recognized that foci suspicious for malignancy are often cut out when subsequent sections are taken for immunohistochemical staining [8], [9].
Although the constellation of features considered to be diagnostic for prostate adenocarcinoma on light microscopy has shown some evolution over recent years, the detection of abnormal glandular architecture, nucleolar prominence, and the absence of a basal cell layer, first described by the Francis Delafield Hospital group in New York 53 years ago [10], remain the lynchpin on which the diagnosis of prostate cancer is based. In some biopsies that appear atypical, these features may not be clearly demonstrable for one of several reasons, such as the small size of the atypical focus, the poor orientation of the specimen leading to tangential sectioning of acini, the presence of inflammation, or the distortion of acinar architecture and/or cellular detail. So problematic is the diagnosis of malignancy in a small proportion of biopsies, the convention has arisen that the pathologist describes the tissue as showing atypical small acinar proliferation [11] or focal glandular atypia [12], and these terms have been proposed as a diagnostic category for prostate biopsies. This is an unfortunate development because rather than being diagnostic, these labels are indicative of an inability to diagnose, and as such, should serve only as a useful addition to the terminology of diagnostic uncertainty, alerting the clinician to the fact that because malignancy cannot be confirmed or refuted, repeat biopsies should be considered.
In an attempt to increase the diagnostic yield of thin core biopsy, the validity of additional histologic features has been investigated with varying success, and this has led to the identification of major and minor diagnostic criteria for prostate carcinoma by the workgroup participants at a consensus conference held in 1995 [13]. The major criteria adopted by the conference were virtually unchanged from those formally defined over 50 years ago, and whereas the minor criteria were themselves not considered diagnostic of cancer, it was anticipated that they should serve to signal the possibility of the diagnosis in the absence of major criteria. In addition to those features detailed in the consensus diagnostic guidelines, the utility of a variety of other cancer markers has been investigated, and of these, the identification of circumferential perineural infiltration by tumor [14], the presence of glomerulations (glomeruloid structures) [14], [15], and mucinous fibroplasia [14], [16], [17] have been proposed as features pathognomonic of malignancy.
Although it is well recognized that benign prostate acini may abut on intraprostatic neural elements, the presence of circumferential infiltration, regardless of the degree of cellular atypia, is considered diagnostic for cancer. Despite this, 16% of the respondents chose to reject this as satisfactory criterion of malignancy. The presence of mucinous fibroplasia or glomerulations have similarly been shown to have a 100% specificity and positive predictive value for malignancy [18], and yet somewhat surprisingly, this was not supported by 36% and 42% of the respondents, respectively. The reasons for the apparent lack of confidence in these features as diagnostic markers of malignancy is unclear, but may be a reflection of the relative novelty of the criteria and the limited number of corroborative studies. In addition to this, the validity of the presence of glomerulations as an absolute diagnostic criterion has been questioned, as occasional glomeruloid-like projections have been observed in benign glands [19].
The last decade has shown a considerable advance in our understanding of the behavior of prostate cancer and in the development and refining of diagnostic and prognostic criteria. From the results of this survey, it would appear, however, that further evidence is required before these more recent findings are universally accepted in clinical practice.
References
- . Prostate cancer screening in Australasia. Clin Oncol. 2005;17:231–233
- . Thin core biopsy of prostate. Pathology. 1998;30:247–256
- . Current practice of diagnosis and reporting of prostate cancer on needle biopsy among genitourinary pathologists. Hum Pathol. 2006;37:292–297
- . Variations in the processing of prostatic needle cores in the UK; what is safe?. J Clin Pathol. 2003;56:341–343
- Needle biopsies of the prostate: what constitutes adequate histologic sampling?. Arch Pathol Lab Med. 1998;122:833–835
- . Adequate tissue sampling for prostate needle core biopsies. Am J Clin Pathol. 1997;107:26–29
- How often are diagnostic features missed with less extensive histological sampling of prostate needle biopsy specimens?. Am J Surg Pathol. 1999;23:257–262
- . Needle biopsy technique, tissue sampling, and processing of needle biopsies and transurethral resection specimen. Chapter 2 In: Prostate biopsy interpretation. 3rd ed.. Philadelphia: Lippincott, Williams, & Wilkins; 2002;p. 7–10
- . The utility of basal cell-specific anti-cytokeratin antibody (34beta E12) in the diagnosis of prostate carcinoma. A review of 228 cases. Am J Surg Pathol. 1995;19:251–260
- . Microscopic differential diagnosis of latent carcinoma of prostate. Arch Pathol. 1953;55:131–141
- . The focus of atypical glands suspicious for malignancy in prostatic needle biopsy specimens: incidence, histologic features and clinical follow-up of cases diagnosed in a community practice. Am J Clin Pathol. 1997;108:633–640
- . Diagnostic effect of complete histologic sampling of prostate needle biopsy specimens. Am J Clin Pathol. 1998;109:416–422
- Assessment of prostate cancer in core needle biopsy—definition of minimal criteria for the diagnosis of cancer in biopsy material. Cancer. 1996;78:376–381
- . Diagnosis and reporting of limited adenocarcinoma of the prostate on needle biopsy. Mod Pathol. 2004;17:307–315
- . Prostatic adenocarcinoma with glomeruloid features. Hum Pathol. 1998;28:543–546
- Mucinous differentiation in prostatic adenocarcinoma. Hum Pathol. 1991;22:979–988
- . Collagenous micronodules in prostate cancer. A specific but infrequent diagnostic finding. Arch Pathol Lab Med. 1995;119:444–447
- Morphologic criteria for the diagnosis of prostatic adenocarcinoma in needle biopsy specimens. A study of 250 consecutive cases in a routine surgical pathology practice. Arch Pathol Lab Med. 2002;126:554–561
- . Adenocarcinoma. Chapter 14 In: Prostate pathology. Chicago: American Society for Clinical Pathology; 2003;p. 258–337
PII: S0046-8177(06)00024-4
doi:10.1016/j.humpath.2006.01.004
© 2006 Elsevier Inc. All rights reserved.
