The Gleason Score

The Gleason score is derived from low-power microscopic examination of cells on slides prepared from the core samples. Normal cells, also called well differentiated, are characterized by distinct clearly defined borders and clear centers. Johns Hopkins pathologist Jonathan Epstein describes them as "little round doughnuts" [37]. As differentiation is lost, cell borders become irregular, cells appear to clump together, and highly irregular shapes become more and more common until at the highest grade, the cells bear little or no resemblance to normal cells. Gleason divided this transition from normal to poorly differentiated into five patterns which generate the numbers one to five used in the pattern grading system.

Gleason dealt with the problem of more than one malignant pattern in a biopsy sample by having the pathologist characterize what is termed a primary or dominant and a secondary pattern, the latter representing at least 5% of the cancer. The dominant and secondary patterns are graded according to the Gleason 1-5 system, and then the two grades added to give the Gleason Score, with the first number the primary grade and the second the secondary. Thus the score runs from 2 to 10. When there is no secondary pattern, the primary pattern score is simply doubled. Some doctors will give the patient just the sum, whereas the individual numbers are also of interest. A description of each core provides the best information. A Gleason score of 6 could be 4+2 with the predominant finding being poorly differentiated cells, whereas 2+4 indicates that the poorly differentiated cells were present in smaller amounts. Patients with high Gleason scores are more likely to have cancer that has penetrated the prostate wall to the point where it perhaps cannot be removed surgically as well as cancer that is more likely to have spread to the seminal vesicles or lymph nodes, and even to other organs. Also, the presence of poorly differentiated cells that yield a high Gleason score suggests an aggressive cancer. Thus high Gleason score cancers may be less likely to respond to standard curative treatment.

At the opposite end of the Gleason scale is the 2 to 4 score. According to Hopkins pathologist Jonathan Epstein [97], most 2-4 scores obtained from needle biopsy are graded 5-6 when reviewed by experts in prostate pathology. In a study at Johns Hopkins Hospital, it was found that their pathologists agreed as to the 2-4 scoring in only 4 of 87 reviews of slides graded initially at other institutions. Sixty-eight were regraded as 5-6, thirteen as 7, and two as 8-10. He also remarks that the few cases actually graded 2-4 at Hopkins showed higher-grade tumor at RP radical prostatectomy). Epstein points out that in a study involving 10 expert pathologists, no agreement was found on what a 2-4 score looked like in a slide review. This is a serious matter since a true grade of 2-4 from a needle biopsy suggests that there is no need for immediate definitive therapy where in fact, the real score is probably higher and indicates the need for treatment. In a series of cases with Gleason scores of 2-4 from outside institutions who underwent RPs at Hopkins, 55% of the patients showed extraprostatic extension, including four cases with either lymph node or seminal vesicle invasion. Epstein quotes his experience with 2-4 grades at Hopkins [97]. He assigned a Gleason grade of 2-4 in only 1.1% of 2285 needle biopsies, whereas 24% of 83 pathological examinations of TURP tissue sent to Hopkins for consultation were given grades in this range. The low-grade cancers typically seen in TURP samples reflect the fact that cancers tend to be small when located anteriorly in the prostate within the TZ (transition zone) where tissue is removed during the TURP procedure.

Dr. Peter Scardino expands on the validity of the 2-4 grade in his recent book [2]. He points out that today many pathologists believe that Gleason patterns 1 and 2 may not be cancer at all, and thus it is rare to have a Gleason score of less than 6, i.e. 3 + 3. He suggests for anyone having a score of 5 or lower that a review of the slides by a pathologist considered an expert in prostate cancer is advisable. Another reason why it is wise to get a second opinion in any case is that pathologists in general differ in their interpretation of what they see through the microscope, and there is a variation in expertise as regards the Gleason grading system. He mentions three so-called referee centers where a second opinion can be obtained: the Armed Forcers Institute for Pathology, Memorial Sloan-Kettering, and Johns Hopkins, all of which have pathologists whose sole occupation is examining prostate tissue. If a man has a Gleason 2-4 result confirmed by expert review, he should raise the question of the merits of proceeding with treatment. The issue is confused by studies in the literature that include the 2-4 score in RP or RT series when in fact, the probability is high that an expert review of the slides would have upgraded these cases into a range where there would be less debate about treatment.

After Gleason score 6, a score of 7 is as the most common found with needle biopsy [98]. There are two ways to get this score, i.e. 3 + 4 or 4 + 3. In the former the lower grade Gleason pattern predominates, whereas in the latter it is the higher grade. Thus the patient should ask to see a breakdown of a score of 7. The score 4 + 3 is generally considered more ominous than 3 + 4, although the evidence is as yet not compelling [99]. Scardino points out that the real issue is the presence or absence of poorly differentiated cancer (patterns 4 and 5), and if so, how much, since poorly differentiated cancer should be treated promptly [2].

Both the Gleason scores of 2-4 and 8-10 are relatively rare in current needle biopsies. The former have an incidence of about 2%, the latter about 8% [37,99]. Authentic scores of 2-4 are generally considered harmless, but as discussed above, they may actually be of higher grade. Many experts feel that scores of 2-4 should not be reported. Grade 5-6 tumors are considered slow-growing, whereas 7 is considerably more dangerous and 8-10 very dangerous. Scores of 7 or greater are more likely to involve extra capsular extensions, seminal vesicle involvement, and lymph node infiltration. The higher the score, the poorer the expected outcome of treatment [99]. But the score is rarely used in isolation, but rather combined in the calculation of prognostic factors using the number of cores containing cancer and how much, tumor stage (see below) and PSA at the time of biopsy. Examples are given below. An obvious question relates to the correlation in general between the Gleason score derived from needle biopsy and the corresponding score found upon pathological examination of the prostate itself after a RP is performed. Sved et al [100] investigated this question for men with a biopsy score of 6. Out of 451 patients, 41% had a score of 7 or greater when the prostate itself was examined, whereas 8% had a lower score and 51% retained the same score. Those who were undergraded at biopsy were more likely to have extraprostatic extension, seminal vesicle invasion and recurrence post-RP. Lattouf and Saad [101] also found significant undergrading in a study of 390 patients with biopsies who went on to have a RP. Undergrading was found in 38.2%, and overgrading in 32.6%. The discrepancies were somewhat smaller when the patients were grouped into categories (Gleason score 2-4, 5-6, 7 and 8-10) with 48.5% of the patients remaining in the same group after pathologic grading. There was also an improvement when only a subset was considered where a single pathologist examined both the cores and the RP pathology. Nevertheless, it is clear that the Gleason score is only an approximation, if it is assumed that the histological examination of the sectioned prostate is really a gold standard. Perhaps this is not surprising considering that the tissue collected in the needle biopsy represents only 0.04% of the average prostate volume. Finally, there is a problem with both intra-observer and inter-observer variability. For the former, studies find only 43-78% exact agreement on blinded reexamination of slides, whereas the disagreement between observers of plus or minus one score unit has been reported to range from 72-87% [99]. Perhaps more confidence and reliance is placed on the Gleason numbers than is actually deserved.

Atypical Results – Intraepithelial Prostatic Neoplasia

Aside from the unfortunate fact that the initial biopsy may fail to detect a cancer present in the prostate, the results of the needle biopsy may not be clear-cut (i.e., benign or PC). The report may come back with terms such as "inconclusive" or "atypical hyperplasia" or prostatic intraepithelial neoplasia (PIN). PIN is sub-classified as low- or high-grade. The low-grade is generally considered clinically insignificant. High-grade PIN (HGPIN) is thought to be the earliest stage in prostate carcinogenesis and is regarded as a premalignant lesion that has the potential of progressing to adenocarcinoma [102]. Thus HGPIN is viewed as an early warning sign. The incidence of a HGPIN diagnosis in needle biopsies averages 9% (range 4-16%). Thus somewhat over 100,000 new cases of HGPIN are diagnosed each year in the US. Low-grade PIN is sufficiently hard to distinguish form normal cells that Epstein, for example, suggests it not be included on pathology reports [103].

Studies of PC risk subsequent to or associated with the observation of HGPIN have produced widely variable results. When both PC and PIN are found at the initial biopsy, the classification is PC because for diagnostic purposes the presence of PIN is now irrelevant. However, until recently the sextant biopsy was the standard protocol and missed approximately 20-25% of cancers in the initial biopsy. Finding HGPIN but not PC generally indicates one or more repeat (immediate) biopsies, and a number of large studies [104] found the repeat biopsy yielded 23 and 35% PC. This range is close enough to that for normally missed cancer in the sextant biopsy that the risk of PC being present in newly diagnosed cases of PIN was not clear. However, Lefkowitz et al [105] reported in 2001 that when HGPIN but no cancer was found in an initial 12-core biopsy, the cancer yield of a repeat 12-core biopsy within one year was only 2.3%. However, this result was inconsistent with that of Rosser et al [106] who also used an extended biopsy protocol. The reason is not clear. Nevertheless, in a subsequent study, Lefkowitz et al [107] used the observation that only 2.3% of patients undergoing repeat 12-core biopsy within 1 year had cancer as validating their particular 12-core protocol and they then examined the incidence of PC after three years in 31 men who had only HGPIN initially and no cancer as determined by this initial 12 core biopsy. They found 25.8 % had developed PC.

Scardino quotes [2] the statistic that 50% of men with high grade PIN will be found to have prostate cancer on a subsequent biopsy over a 5-year period, but it is not clear how effectively undiagnosed cancer at baseline was excluded. Only additional studies will resolve the important question of the real risk of progression of HGPIN to PC if one has HGPIN but no cancer present initially. HGPIN and cancer frequently coexist. For example, 85% of patients diagnosed with PC will also have areas of high grade PIN on examination of the prostate itself [2].

There is no consensus on when a repeat biopsy should be done when HGPIN is found, with some experts recommending six month, others recommending as soon as possible [103]. As regards atypical needle biopsy results in general, Epstein recommends a repeat biopsy within 3 months [103]

Staging of Prostate Cancer

There are two types of "stage" of prostate cancer, the clinical and the pathological stage. The first is an estimated stage based on what the physician believes the patient's cancer to represent, which in turn is based on the DRE (digital rectal examination) and needle biopsy. It is to some extent an educated guess. The pathological stage is determined after surgery when reliable information is available regarding the extent, not only of the cancer within or outside the gland, but also whether or not it has spread to the lymph nodes, seminal vesicles or regions adjacent to the prostate. In cases where advanced disease is suspected, the information available for the clinical staging can be enhanced by bone scans or knowledge of metastasis found in distant sites. Staging information can be obtained when an operation is performed to examine the lymph nodes. Information gathered from the DRE is of necessity incomplete since only part of the prostate can be explored. Nevertheless, staging based just on the DRE and the biopsy is commonly done and is a very important guide with regard to treatment. It is however important to realize that all the information required might be unavailable when the cancer is not organ confined. The 1997 so-called TNM staging system is given by the following (after Walsh [37]).

TNM Staging

T1a: Not Palpable with DRE. Cancer found incidentally during TURP; 5% or less of removed tissue cancerous.

T1b: Not Palpable with DRE. Cancer found incidentally during TURP; more than 5% of tissue cancerous.

T1c: Not Palpable with DRE. Cancer found with needle biopsy prompted by elevated PSA. Most common stage presently.

T2a: Palpable with DRE. Involves one lobe.

T2b: Palpable with DRE. Involves both lobes.

T3: Palpable with DRE and penetrates the wall of the prostate and/or involves seminal vesicles.

T4: Spread to adjacent muscles, bladder neck, external sphincter and/or rectum.

N+: Spread to lymph nodes.
M+: Spread to other sites including bone

The TNM stage is central to therapy decisions [37]. T1 and T2 cancers are viewed as favorable for treatment with RP or RT (radiation treatment) with the intent to cure, whereas T4 cancers have traditionally been considered suitable only for systemic treatment, e.g. hormone therapy, as are M+ and N+ stages, although the prognosis associated with positive lymph nodes and the merits of surgery in this case are debatable and changing with time [108]. T3 tumors are in the gray area. However, there is growing evidence that for clinically staged T3 tumors, RP offers a number of advantages over other forms of treatment [109,110].

As indicated in the table, individuals with a non-suspicious DRE and an elevated PSA who have a positive biopsy are staged as T1c. Approximately 25% of the radical prostatectomies performed for this stage disease reveal potentially insignificant tumors [51]. Epstein et al [51] and Walsh [37] provide guidance regarding the prediction of tumor significance in T1c grade disease (adapted from [37]).

T1c cancer significant if any one of the following characteristics is present:

• Cancer found in three needle cores or is present in greater than half of any one needle core
• Gleason score greater than or equal to 7
• PSA density > 0.1-0.15 ng/mL/cc (volume should be available from ultrasound directed biopsy)
• %fPSA < 15%

T1c cancer probably insignificant if all of the following characteristics are present:

• Cancer found only in one or two needle cores and makes up less than half of each needle core.
• Gleason score less than or equal to 6
• PSA density < 0.1-0.15 ng/mL/cc
• %fPSA > 15%

Walsh states that the above guidelines are only about 75% predictive [37]. If one is in the low risk group, gambles with watchful waiting (more recently called expectant management) and suffers progression, the cancer may still have a high probability of being curable [37]. The challenge is acting before it is too late.

Diagnosing Advanced Prostate Cancer

The needle biopsy can reveal spreading outside the gland, but only into areas accessible during the transrectal procedure. In particular, no information is obtained regarding lymph-node or bone metastasis aside from the enhanced probability of advanced disease inferred from a high Gleason score. Pelvic lymph-node metastasis has significant adverse prognostic implications and it presence impacts treatment decisions. While modeling techniques such as nomograms (see below) or probability tables are used to estimate the chances of lymph- node metastasis, surgical removal followed by histological examination has been the only satisfactory way to obtain a dependable answer. Traditionally, one common protocol was for an examination of the question of lymph-node (LN) metastasis to be done at the start of the RP operation. If lymph-node involvement was found upon frozen section analysis, some surgeons would not proceed with the RP but simply close the incision. This approach is now less common, partly because it is evident that surgery can benefit some patients even if there is lymph-node metastasis [37,111].

Nevertheless, it is generally desirable to know the LN status. The performance of a lymphadenectomy to obtain lymph nodes for pathological examination as a stand alone procedure, as would be required if RT was the planned mode of treatment, is only justifiable if the results are going to significantly impact treatment. If hormone therapy is planned anyway as an adjunct to RT, knowledge of the LN status is unnecessary. The removal of the lymph nodes associated with the prostate at present appears to have no therapeutic value even if metastasis is present, but knowledge of its presence is important in both judging prognosis and, in some cases, in treatment decisions. If a patient is of very low risk of LN metastasis, as determined from tables or nomograms, some surgeons will omit the removal of these nodes. A recent study [112] justifies this approach for low-risk patients, but merits of omitting pelvic lymph node dissection have also recently been questioned by Weckermann et al [113] who suggest not dispensing with surgical LN staging even in apparently low-risk disease. Some surgeons routinely remove the easily accessible nodes, while some remove all the relevant ones in a so-called extended pelvic node dissection. There is no doubt that patients with LN metastasis are at high risk of not remaining progression free, and the more extensive the LN involvement, the poorer the prognosis [114].

Non-invasive procedures such as CT scans and traditional MRI would be attractive as procedures to acquire information on lymph node status, but both lack the sensitivity, especially for clinically occult disease where there has not been as yet an enlargement of the infiltrated nodes. In what appears to be a major diagnostic breakthrough, Harisinghani et al [115] have recently reported the results of a study using a special imaging agent which collects in lymph nodes but does not image areas where metastasis has occurred. MRI with superparamagnetic nanoparticles correctly identified all patients with nodal metastases, and node-by-node analysis had a significantly higher sensitivity than conventional MRI or nomograms. Only in the case of nodes measuring less than 5 mm was there a decrease in sensitivity. This approach has the advantage of being non- invasive, does not require hospitalization for surgery, is free of side effects, and has 100% sensitivity. The reference standard for the presence of metastasis was based on surgical lymph-node resection or a biopsy.

Metastasis frequently involves the spread of the cancer to bones, and thus if this is suspected, a bone scan may be offered or recommended prior to treatment decisions. There have been a number of studies that attempt to establish guidelines for when a bone scan is appropriate. The most recent appears to be that of Lee et al [116] from Columbia University. They studied 631 patients with bone scans for whom they had a Gleason score, PSA value and DRE based clinical staging. From this study they developed a guideline for identifying patients for whom a bone scan would have a low probability of being positive. The guidelines were a Gleason score of 2-7, a PSA of 50 or less, and a clinical stage of T2b or less. For patients with clinical stage less than or equal to T2b and Gleason score 2-7, and PSA less than or equal to 15, zero out of 237 patients had a positive bone scan. Raising the limit to PSA less than or equal to 50 resulted in 3 out of 545 patients with a positive bone scan. Positive bone scans were obtained in 49.5% of patients with PSA > 50. They recommend a bone scan in all patients not meeting the above criteria and especially those with a PSA > 50. The authors review the literature up to 1999-2000 and find similar but not identical results. The studies reviewed looked only at a PSA cut-off. Patients with a PSA of less than or equal to 10 always had negative bone scans, whereas less than or equal to 20 produced a small percentage of positive results. The guidelines presented by the authors are similar to that given by Scardino [2] where he states that a bone scan may provide useful information if the Gleason score is 8- 10, PSA > 20 , and the clinical stage is T3 or greater. For those who do not meet these criteria, he suggests that the scan offers little value and has considerable potential to generate unnecessary problems due to false positives and possible recommendation of a bone biopsy, which itself is not necessarily conclusive.

Evaluating the Pre-Treatment Picture – Crunching the Numbers

At the time of diagnosis the patient generally has a PSA value, a Gleason Score, and a stage based on the T1, T2 etc. system. In the last few years considerable effort has gone into developing mathematical models that allow predictions based on these clinical diagnostic results [16,117-120]. These models are based on pathological and medical outcomes from large databases, but to some extent contain institutional bias (the institutions specializing in PC may get the best treatment results). The results are presented in the form of tables (Partin Tables, www.urology.jhu.edu) or so-called nomograms (a diagram where points can be assigned to the various predictive factors and then summed to yield a total which is used to predict outcome). The pencil and paper analysis using nomograms or tables can now be avoided by using the internet based nomogram from the Memorial Sloan-Kettering Cancer Center (www.mskcc.org/mskcc/html/10088.cfm), which does all the calculations The patient inputs the PSA, Gleason score and tumor stage and the program calculates and displays the probabilities of organ confined disease, extra capsular penetration, seminal vesicle involvement, LN involvement, and the 5-year progression-free probability after RP or RT. This provides the patient with an easy to understand analysis of the probable significance of the set of parameters that characterize their diagnosis. The probabilities are based on databases that use clinical rather than pathological (post surgical) input, and thus the problem of under-grading is presumably taken into account. It is nomograms such as these that are used by some surgeons to decide if looking for LN metastasis can be omitted during surgery because the probability is very low.

As an example of the application of nomograms, the online Memorial Sloan-Kettering calculator, will be used to examine various outcomes for exceeding the two commonly suggested PSA thresholds for biopsy, 2.5 and 4.0. In this example, it will be assumed that the cancer stage is T1c, the most common situation today (T2a and T2b are next in incidence [38]). The results are given in the table below. The second, third and fourth items stratify the non-organ confined disease. Note that on the basis of the databases used to construct this model, Stage T1c and a "low" PSA did not completely rule out non-organ confined disease. The low biochemical failure rate (high % progression free) for these two typical screening scenarios for a Gleason score of 6, a common score with screened patients, is consistent with the view that early detection is highly curative. This table also illustrates that the lower cut-off for triggering a biopsy had a negligible influence on 5-year disease-free outcome as compared to the traditional cut-off. This was one of the principal arguments presented by H. Ballentine Carter from Johns Hopkins in an editorial discussing lowering the cut-off [43]. In fact, in a 1999 study by Carter et al [39] it was found that the probability of curable cancer was only slightly different within the range of PSA values from 2.5 to 6.0 for men with clinical T1c cancer. Increasing the PSA to 4.5 for a Gleason score of 6 does not change the figures in the above table. From these results it would appear that, for example, a PSA of 4.5 is not a crisis situation. In fact it can be argued that there is merit in repeated measurements at this point to determine the PSA velocity. It is even possible that the value would come back down below 4.0. However, recovery from PSA spiking above a cut-off does not rule out PC [121].

Summary

One of the characteristic features of prostate cancer is the uncertainty surrounding practically every important aspect. Screening is a probability game, to some extent like handicapping horse races. Both the total PSA test and the modern variations have significant false positive and false negative rates. What is now quite clear is that while PSA testing can indicate the risk of PC, it cannot be used to rule out the presence of the disease. Intra-individual PSA variations from day to day and year to year are significant and confuse the issue of decision making with regard to biopsy advice. Drugs used to treat BPH (5-alpha-reductase inhibitors) and perhaps elevated cholesterol (statins) significantly lower PSA levels, the accuracy of corrections required before applying diagnostic benchmarks is unknown, and other drugs that influence PSA levels may turn up any time. The transrectal ultrasound guided needle biopsy only approaches 100% diagnostic accuracy when biopsies, if negative, are repeated several times. Pathologists reading needle biopsy generated slides exhibit alarming intra- and inter-observer variations when establishing Gleason scores. The DRE is notorious for false positives and false negatives. Imaging techniques still fail to provide definitive diagnosis. These are all complex issues. It is not for lack of research that these uncertainties exist. At present, nothing better appears available. A vast amount of careful and thoughtful research has brought us to this point. Clearly there are great challenges ahead. Knowledge of this rather unsatisfactory state of affairs is no doubt limited among the general public unless they have made a considerable effort at self-education.

However, it is also true that PSA screening has revolutionized the diagnosis of PC. The disease is being identified and treated at a much earlier stage than was possible before the PSA era. The PSA test offers the only approach to early diagnosis in general and routine use today. PSA screening is gaining popularity worldwide and is well established in the practices of urologists, internists and general practitioners. The problems and uncertainties are clearly not compelling enough to discourage this widespread use.

This review is concerned mainly with two issues. The first involves the PSA test—what does one need to know to decide whether or not to have the test or if variations in the PSA test should also be requested if not offered, especially the %fPSA, and the PSAV. The second issue concerns the basic knowledge deemed desirable prior to agreeing to proceed with the biopsy phase of diagnosis.

The pros and cons of getting the PSA test are those elaborated above in connection with the screening debate. To these must be added age, life expectancy, race and family history. It is not possible to resolve this debate since it depends not so much on disputing the facts as how each argument is weighted. The arguments of those opposed to screening are to some extent weakened by the studies that suggest that the %fPSA and the PSAV enhance the specificity of the tPSA test. There is still no escaping the fact that all three tests have their gray areas. At the extremes the indication of high or low risk has fairly high probability of being correct, and this supports either feeling good about the low risk of having PC or being forced to consider the merits of a biopsy if the risk is high.

It is the gray area that causes the most anguish. However, being in the gray area or at high risk also brings up the question of the merits of treatment vs. watchful waiting (WW), since there is little point is getting a biopsy or even worrying about screening if the decision has been made to reject conventional treatments, at least until the cancer, if present, causes major problems requiring palliative treatment. Such a rejection might be motivated by what the individual views as unacceptable adverse effects associated with either the radical prostatectomy (RP) or radiation treatment (RT) coupled with what the individual perceives as only marginal benefits of treatment over WW when it comes to prostate cancer specific mortality. As discussed above, such a rejection ignorers or rejects a considerable body of evidence supporting the advocates of screening which suggests a definite shift to favorable probabilities associated with having a PSA test and a DRE.

A compromise between the extreme options of aggressive treatment such as RP or RT or WW followed if necessary by palliation involves WW until evidence for progression suggest it is now or never for treatment with the intent to cure. This protocol is currently under study [122] and offers an approach which may be attractive to some men. It requires PSA and DRE monitoring and more biopsies. For men with diagnosed but possibly indolent cancers, such a compromise would have merit. It will be a number of years before long-term survival data associated with this protocol are available.

The discussion of the prostate biopsy in this review makes it clear that a single biopsy is by no means a gold standard for the presence or absence of PC. Quite the contrary, it misses a substantial percentage of cancers. Men should be aware that once one has agreed to a biopsy, one or more repeats may be suggested for good reason. Also men should be aware that the "classical" sextant biopsy is in general inferior to the so-called extended biopsies which utilize more needles, some directed to higher probability zones, with 12 or 13 needles now being common. While the biopsy is far from perfect, with all due respect neither are the pathologists who read the slides, and men should be aware of the advisability in some cases of a second opinion, preferably from someone with no potential conflict of interest (e.g. avoid having slides reviewed by an office or department associate). Also, if one elects to have a biopsy, there appears no reason to hesitate in asking for local anesthetic and in particular the nerve block, especially if the extended protocol is used. Being aware of the potential complications, their signs and symptoms, and what action is indicated is extremely important. Serious complications associated with the needle biopsy are rare but some can develop very rapidly and can be life- threatening. Finally, one probably should not have a biopsy unless the decision has been made beforehand as to what the next step will be if the result is positive or negative. However, some men who have no intention of being treated may still want to know in detail the nature of what they have decided not to treat, just for future reference.