Prostate Cancer Diagnosis: Do you need a second Opinion?
Epstein, Jonathan I.
Prostate Cancer Diagnosis: Do you need a second Opinion?
THE IMPORTANCE OF OBTAINING A SECOND OPINION ON PATHOLOGY BIOPSIES WITH EMPHASIS ON PROSTATE NEEDLE BIOPSIES
Author: Jonathan I. Epstein, M.D.
Professor of Pathology, Urology, and Oncology
The Department of Pathology, The Johns Hopkins Hospital, Baltimore, MD.
When a patient receives news of what their biopsy showed, there may be several misconceptions. First, you as the the patient may assume that the surgeon who performed the biopsy was also the person who rendered the diagnosis. This misperception may be perpetuated when surgeons do not tell patients that diagnoses are made by someone other than themselves. In fact, when your tissue is biopsied, the samples taken are studied under the microscope by a specialized doctor with many years of training called a pathologist. The pathology report tells your treating doctor the diagnosis in each of the samples to help manage your care.
The second misconception is that the diagnosis on your tissue is analogous to the result of other laboratory tests, for example a blood glucose (sugar) level. These other lab tests, typically taken from your blood, are processed in calibrated machines that typically give straightforward and reproducible results. In contrast, when a specimen is analyzed under the microscope, the analysis is made by a person whose performance is subject to all the variability associated with human tasks. Some pathologists have extra training or expertise that make them particularly skilled at making certain diagnoses. For instance, there are pathologists that specialize in diseases of the prostate, breast or skin. These pathologist are the ones that are sought out by their colleagues when the interpretation of the biopsy sample is complex or confusing. This is analogous to surgeons in that you would not chose a colon surgery if you had problem that required a neurosurgeon. . However, it should be emphasized that virtually all practicing surgical pathologists in the United States have passed difficult National tests (American Board of Pathology examinations) such that, in general, there is a minimum high level of competency amongst the profession.
In addition to variability amongst pathologists, the analysis of tissue sections is much more fraught with difficulties than the analysis of blood samples by a machine. Often tissue samples are very small, can show distortion from the biopsy, and there are, in almost all organs, various mimickers that can be confused with the correct diagnosis. There are special studies that can be done on tissue lesions that can, in some cases, help arrive at the correct diagnosis, but in other cases can lead to misleading information and potentially the incorrect diagnosis.
In order, to circumvent these pitfalls with analyzing tissue sections, pathologists often show difficult cases to other pathologists at their institution for second opinion. In some cases this consists of a certain percentage of all cases (ie. 10% of all cases are reviewed by 2 pathologists). In others, only types of tissues that historically are more challenging cases are reviewed (ie. all prostate needle biopsy specimens are reviewed by 2 pathologists). Another common quality control mechanism is that all first time malignant (cancerous) diagnoses are reviewed by at least 2 pathologists. Finally, pathologists may elect to send microscope slides from challenging cases to an outside institution for consultation with a nationally/internationally recognized expert in the field for a second opinion.
There have been multiple studies looking at the incidence of diagnoses that differ amongst pathologist looking at the same case (1-7). In a study from our institution (The Johns Hopkins Hospital), of 6,171 cases reviewed, second opinion surgical pathology resulted in 86 “major” changed diagnosis (1.4%) (2). This is virtually identical to another study from our institution which found a 1.3% rate of change in diagnosis from malignant to benign in prostate needle biopsies (1). In most of these changed 1.4% diagnoses amongst all organ sites, the difference was also between one pathologist calling the specimen “benign” and the review pathologist calling the specimen “cancer” or vice versa. We found that no type of organ was less likely to have a change in diagnosis, such that all types of biopsies had a risk of an incorrect diagnosis. Both this study and a prior study from our institution demonstrated that there were discrepant diagnoses in cases from community hospitals, commercial laboratories and large teaching institutions, such that a patient can’t say just because his or her case was reviewed at, for example, a major medical center then the diagnosis is accurate (1,2).
The 1.4% change in diagnosis did not factor in changes associated with tumor stage and grade. However, these changes can influence patient management although their assessment may be more subjective. For example, some men with nonpalpable prostate cancers diagnosed by needle biopsy (stage T1c) may be candidates for watchful waiting, as they have relatively "insignificant" tumor (10). Preoperative identification of these men requires accurate grading and quantification of cancer on needle biopsy. At the other end of the clinical spectrum, accurate grading of poorly differentiated tumor is necessary since these men are likely to fail systemically and may not benefit from surgery.
We did another study analyzing prostate needle biopsy specimens to assess the overall change in diagnosis when an expert reviews cases that have been diagnosed by a generalist who does not specialize in prostate pathology. In 241 (35.2%) cases upon expert review, a change in diagnosis was rendered (7). The expert agreed with the majority of outside cancer, benign and precancer diagnoses. In contrast, the expert agreed in only about 1/3 of outside cases where the nonexpert pathologist called the biopsy uncertain or “atypical”. An atypical diagnosis can be rendered on needle biopsy for several reasons. Most often, there are just a few atypical glands present that have insufficient changes under the microscope for the pathologist to render a definitively malignant diagnosis. Cancer may not be diagnosable since the pathologist cannot exclude mimickers of cancer, such as atrophy, adenosis, nonspecific granulomatous prostatitis, and high grade PIN (a precancer). The presence of associated inflammation can result in an atypical diagnosis, as inflammation can cause benign glands to resemble cancer. Another reason for an atypical diagnosis relates to glands or cells with crush artifact as a result mechanical distortion from the needle biopsy procedure. Almost 2/3rds of nonexpert cases were changed to a cancer or benign diagnosis upon review by an expert pathologist. The difference between an atypical and cancer diagnosis often reflects the experience and comfort level of the pathologist in making a diagnosis of a small focus of cancer.
It is difficult to assess how commonly a benign diagnosis is changed to cancer. If one looked at the entire group of men, only 3/684 (0.4%) outside benign cases were diagnosed upon review as cancer. However, there is very scant information as to the incidence of missed cancer diagnoses on needle biopsy. In order to do these studies one would have to allow an expert to review an institution’s prior benign cases for missed cancer, which would have patient care and medicolegal ramifications.
The majority (73.5%) of changes affecting outside cancer diagnoses were due to changes in Gleason score (7). Overall, in about 20% of cases the expert review changed the grade made by the nonexpert pathologists. While the significance of a change in Gleason score may be argued, these changes may impact on management in terms of whether a patient is a candidate for watchful waiting, surgery or radiation, or the type of surgery or radiation recommended. For example, the largest number of cases with a grade change involve Gleason score 5-6 versus Gleason score 7. Whereas watchful waiting may be an option for a man in his 60s with a small Gleason score 5-6 cancer on biopsy, definitive therapy would typically be recommended for a comparably sized Gleason score 7 tumor. The finding of extensive Gleason score 7 tumor on multiple cores might tip the balance for radiotherapy over surgery, as opposed to extensive Gleason score 6 cancer. Nerve-sparing surgery might be the preference for a T1c (Nonpalpable) Gleason score 6 cancer of relatively limited quantity on biopsy, in contrast to a Gleason score 7 cancer on biopsy associated with a higher likelihood of extra-prostatic extension. For similar reasons, brachytherapy as the sole therapy is often performed for Gleason score 6 cancer on biopsy, yet considered potentially insufficient for Gleason score 7 tumor.
In summary, overall the vast majority of pathology diagnoses are accurate and rendered by competent pathologists who have received extensive specialty and, in some cases, subspecialty training and who continue to maintain their medical education through formal and informal processes. Nevertheless, there is a minority of pathology assessments that, when reviewed by experts in the field, will end up with a different diagnosis that could significantly affect the treatment and prognosis of the patient. Although the overall percentage of affected cases is not large, the consistent rate of discrepant diagnosis uncovered by second opinion surgical pathology may have an enormous human and financial impact. Given that the pathologic diagnosis is the foundation for all subsequent patient treatment, it is reasonable to consider obtaining a second opinion from an expert on your pathology specimen, especially for those with a malignant diagnosis or an equivocally malignant (ie. atypical) diagnosis. Atypical diagnoses have a relatively high likelihood of being more definitively diagnosed as either benign or malignant by an expert reviewing the case. Malignant diagnoses are very uncommonly changed to “benign” upon review, but the consequences of a misdiagnosis are so drastic that it is reasonable to get a review for these cases as well. For example, with prostate needle biopsies, even though there is only about a 1.4% chance of a prostate cancer diagnosis being changed to not cancer upon review by an expert, this means for every 200 men diagnosed with prostate cancer there are 3 men who may not have cancer. This number, while low, is not rare. More frequently, even for the cancer cases with an accurate diagnosis, the grade may be changed in about 1 in 5 cases by expert review which could have a major impact in treatment and prognosis.
FOR PATIENTS: FREQUENTLY ASKED QUESTIONS (FAQS) ABOUT YOUR PATHOLOGY REPORT CLICK OVER THE LINK BELOW TO THE JOHNS HOPKINS PATHOLOGY PATIENT CARE INFORMATION PAGE
JOHNS HOPKINS PATHOLOGY PATIENT CARE INFORMATION PAGE
1. Epstein JI, Walsh PC, Sanfilippo F. Clinical and cost impact of second-opinion pathology. Review of prostate biopsies prior to radical prostatectomy. Am J Surg Pathol. 1996 Jul;20(7):851-7.
2. Kronz JD, Westra WH, Epstein JI. Mandatory second opinion surgical pathology at a large referral hospital. Cancer 1999; 86: 2426-35.
3. Kronz JD, Milord R, Wilentz R, Weir EG, Schreiner SR, Epstein JI. Lesions missed on prostate biopsies in cases sent in for consultation. Prostate 2003 Mar 1;54(4):310-4.
4. Leslie KO, Fechner RE, Kempson RL. Second opinions in surgical pathology. Am J Clin Pathol. 1996; 106(4 suppl 1): 58-64.
5. Murphy WM, Rivera-Ramirez I, Luciani G, Wajsman Z. Second opinion of anatomical pathology: A complex issue not easily reduced to matters of right and wrong. J Urol 2001; 166: 1957-59.
6. Tsung JS. Institutional pathology consultation. Am J Surg Pathol. 2004 Mar;28(3):399-402.
7. Chan TY, Epstein JI. Patient and urologist driven second opinion of prostate needle biopsies. J Urol 2005: 174:1390-4.