The guide to due diligence in early-stage prostate cancer

Nancy Ferrari

Senior editor, Harvard Health

In a business, due diligence means doing your homework, exploring all the options, and taking reasonable steps to protect yourself. When trying to make a decision about how to treat early-stage prostate cancer, taking time to conduct due diligence is absolutely vital for three reasons:

  • No expert consensus exists about which treatment for early-stage prostate cancer is best (or when treatment is best postponed).
  • Evidence indicates that you can wait up to a year after a prostate biopsy to make a treatment decision without jeopardizing your health.
  • Rushing into treatment without fully considering the consequences of side effects like impotence and incontinence may have an enormous impact on your life and your relationships.

In this article, you will learn how to conduct due diligence for early-stage prostate cancer: what questions to ask and what information to consider. The advice provided in these pages is based on information I share with my own patients.

This advice may be helpful even if you were diagnosed several years ago and have already gone through treatment. Researchers are uncovering new information about prostate cancer all the time: To be truly informed, talk with your doctor every few years and review the information gathered during your initial diagnosis and treatment. You may want to adjust your initial decisions based on more current information.

Questioning your numbers

Chances are, you already know your “numbers” — your PSA level, the Gleason score of your prostate biopsy, and the clinical stage of your tumor as determined by the physical exam that was performed with the digital rectal exam (DRE). If you need a brief refresher, don’t worry: We’ll provide basic information on each of these subjects before going into the types of questions to ask to gain more advanced knowledge.

Before recommending a treatment, your doctor uses this information to help determine how likely it is that your cancer is the type that might progress or metastasize, manifesting again in the future. After treatment, your doctor will continue to monitor your PSA levels in order to gauge the effectiveness of therapy.

Clearly, you’re going to be faced with the challenge of digesting a lot of information. But rather than just accepting your numbers at face value, it’s wise to consider asking questions that will help you make an informed decision about treatment. In the following pages, I’ll suggest questions to ask regarding each test result or finding, based on the advice I’d give if you were my patient.

Table 1: Questions to ask about your PSA

Before deciding on a course of treatment, ask: Why this is important to know
How has my PSA changed over time? The rate of change over time, or PSA velocity, provides an indication of how advanced your cancer is and how likely it is to metastasize. PSA levels that double from one year to the next may signify an aggressive cancer; a PSA that has gone up by more than 2 units in the past year may be worrisome. PSA tests that have risen more slowly probably indicate a slow-growing cancer.
Does my current PSA indicate that the cancer is likely to progress? It’s important to obtain your doctor’s opinion about this question, but understand that opinions on the matter vary simply because there’s such a variation in PSA levels. Generally speaking, PSA levels of 5–10 ng/ml probably indicate localized disease that is unlikely to become active, while those greater than 20 ng/ml indicate you are at higher risk.When used with results from other tests, your PSA level may help you avoid unnecessary diagnostic or therapeutic procedures. For instance, I had one patient with a PSA of 7.6 ng/ml who underwent a bone scan that revealed an abnormality on one of his ribs. But given his relatively low PSA value, as well as his Gleason score and clinical stage, it was unlikely that this single bone scan abnormality was due to metastatic cancer.
When monitoring your PSA to assess the effectiveness of your treatment, ask: Why this is important to know
What is my PSA level? If you’ve had surgery:After surgery for prostate cancer your PSA level should drop to undetectable levels (0 ng/ml) because your prostate gland, and the source of PSA, has been removed.

If a follow-up PSA test reveals PSA levels even as low as 0.2 ng/ml, this indicates that cancer remains in the body or has recurred.

If you’ve had radiation therapy:

Here the situation is more complex. Remember that PSA is specific to the prostate gland and is not just produced by cancer cells. After radiation therapy, your PSA value may not drop completely to zero — and yet treatment is considered successful. This is because your prostate gland has not been removed and continues to produce some PSA.

By no means should you consider a PSA level of under 4 ng/ml to be “normal,” however, as it would have been before radiation treatment. For optimal results, following radiation treatment, your PSA level should be less than 1 ng/ml and even less than 0.5 ng/ml. Levels that are above 1–2 ng/ml 12 to 18 months following radiation treatment may indicate that some cancer remains.

What to ask about your PSA test

This blood test measures the level of a protein called prostate-specific antigen (PSA), which is manufactured by both normal prostate cells and prostate cancer cells. It is important to understand that PSA is specific to the prostate gland — but not to cancer cells — which is why some PSA is detected even in men whose prostate cancer is treated with radiation (see Table 1). Prostate cancer cells allow more PSA to get into the bloodstream than normal cells, however, which is why an abnormally high or rising level of PSA may indicate cancer.

A PSA level of 4 ng/ml or higher may trigger a biopsy. (A nanogram is 1 one-billionth of a gram, a remarkably small amount, as it takes 454 grams to make up one pound.) After a biopsy reveals cancer, however, your preoperative PSA level becomes one of the factors that you and your doctor can use in determining how advanced the cancer is and what your treatment options are.

After a cancer diagnosis, your PSA levels will also be monitored periodically to gauge the effectiveness of whatever treatment you choose, or to assess whether the tumor is growing if you have opted for a strategy of active surveillance. Once again, it’s important to know what questions to ask.

Table 2: Gleason score risk assessments

Gleason score Simple risk assessment Points to consider when assessing risk
2–4 Lower risk
  • Pay attention to the first value assigned, as this is based on the preponderant area of cancer. That is why a Gleason score of 7 has a worse prognosis if it is based on the values 4 + 3 than if it is based on the values 3 + 4.
  • Most cancers detected as a result of PSA screening are Gleason 6 (3 + 3) or 7 (3 + 4)
5–67 (if 3 + 4) Moderate risk
7 (if 4 + 3)8–10 Higher risk

What to ask about your Gleason score

After your prostate biopsy, a pathologist examines your core tissue samples under a microscope to determine whether cancerous cells are present and how different they look from surrounding healthy cells. The pathologist then assigns a Gleason score that provides an estimate of the pathologic grade of the cancer. This is one of the most important factors in determining whether the tumor is confined to the prostate and how aggressive the cancer is. (The Gleason score is named for Dr. Donald Gleason, a pathologist who developed this grading system in the 1960s, basing it on the thousands of prostate biopsies he evaluated while working as a pathologist with the U.S. Veterans Administration. The VA conducted an important series of early studies about prostate cancer.)

The Gleason score is a numerical value that grades cancer cells according to how they appear compared to normal prostate tissue and how active they appear under the microscope (see Figure 1a). The Gleason score is actually the sum of two values. The pathologist looks at the tissue samples obtained during biopsy and assigns a Gleason value first to the area where most cancer cells are present (the preponderant cancer area) and then to the next most preponderant area. A value of 1 is assigned to cancer cells that are most differentiated (meaning that they most resemble normal cells, which differentiate to assume particular roles in the body) and least aggressive (not dividing rapidly), and a value of 5 to cells that are least differentiated and most aggressive. The two values are then added to produce a Gleason score.

The Gleason score ranges from 2 to 10; the higher the number, the more aggressive the cancer (see Table 2). It’s important to understand, however, that the Gleason score is only as good as the pathologist examining the tissue — and the tissue obtained to begin with. Pathologists also meet periodically to reach consensus on how best to modify the Gleason scoring system to accommodate new understandings of how prostate cancer develops and advances (see Figure 1b). For now, the Gleason score is best used in conjunction with PSA levels and tumor staging to get a better overall view of how aggressive a cancer is.

Figure 1a: The Gleason score

The Gleason score is a numerical value that grades prostate tumor cells according to how they appear compared to normal prostate cells (a quality known as differentiation). Because tumors often consist of multiple types of cells, the pathologist assigns two values: the first to the predominant cell type, and the second to the next-most-prevalent cell type. These two values are added to come up with the Gleason score.

Well differentiated

  1. Glandular cells are small, of fairly uniform shape, and tightly packed together.
  2. Cells display more varied and irregular shapes and are loosely packed.

Moderately differentiated

  1. Cells are even more irregular in size and shape and are more dispersed; some cells are fused, and cell borders are less distinct.

Poorly differentiated

  1. Many cells are fused into irregular masses; some cells (see those darkly shaded) have begun to invade the connective tissue that separates cells.
  2. Most of the tumor consists of irregular masses that have invaded the connective tissue.

Figure 1b: Modified Gleason grading system

Pathologists met at the 2005 International Consensus Conference of Urological Pathologists to develop guidelines about how to modify certain aspects of the Gleason score, especially when attempting to grade tissue samples that contain more than two types of cancer cells show unusual patterns (see below). The Gleason scoring system has remained useful precisely because pathologists have made adjustments to it over the years.

This drawing shows proposed modifications to the Gleason grading system. The most significant changes are to Gleason patterns 3, 4, and 5, which in the proposed model show a greater mixture of different cell types and less uniformity in the patterns seen by a pathologist. These modifications reflect a better understanding of the complexity and variety of prostate cancer made possible by advances in diagnostic technology — such as the use of narrower biopsy needles to remove multiple tissue cores from different areas of the prostate — that occurred after Dr. Donald Gleason created the grading system in the 1960s.

Source: Epstein JI, Allsbrook WC Jr., Amin MB, et al. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. American Journal of Surgical Pathology 2005;29:1228–42. PMID: 16096414.

Why a second opinion — and even a second biopsy — may be necessary

It’s always wise to question how the pathologist arrived at your Gleason score (see Table 3). Here’s why: Often the findings are unclear. If there’s any question about the pathology of your tumor, you want to know that before you decide on a treatment plan rather than afterward.

Table 3: What to ask about your Gleason score

The questions listed below will help you better evaluate whether you should accept your Gleason score at face value or ask for a second opinion. (As a rule, I always advise patients to ask for second opinions, simply because different pathologists may interpret the same findings in different ways.)

Even if you were diagnosed several years ago and have already undergone treatment, it may be helpful for you to ask your doctor to review your biopsy pathology report with you again. Medical practice is dynamic, changing in response to evidence from newly reported studies. It’s always a good idea to review your prostate cancer treatment plan periodically to see if it’s time to make adjustments.

Questions to ask Why this is important to know
How many cores were taken during my biopsy? Six cores were once standard, but concern about whether this led to undersampling of the prostate — and missed cancers — has led to changes in how biopsies are done. Controversies remain.Some medical centers do a modified sextant technique, in which six cores are taken from various parts of the prostate, including peripheral areas. The consensus among experts, however, is that it’s better to take 10 to 12 cores during a standard biopsy, in order to sample more of the prostate. When a patient is considered high risk for cancer, either on the basis of his PSA level or from the results of a previous biopsy, some medical centers — including the one where I practice — do “saturation” biopsies, involving 20 or more cores. (Of course, any abnormality detected on a physical examination should be biopsied.)

Some doctors argue that as more cores are sampled, we may be increasing the likelihood of finding insignificant cancers. Frankly, I’m not convinced. And if you are like most of my patients, you want to know if you have cancer so you can then decide what to do.

If you have any doubts about your diagnosis (and especially if the pathology findings from an initial biopsy are equivocal), consider having a second biopsy at a medical facility that performs saturation biopsies.

What areas of the prostate were cores taken from? Ideally, prostate biopsies take cores from both sides of the prostate (bilateral sampling) and from particular regions on each side of the prostate. Not only does this increase the likelihood that the biopsy will sample enough areas of the prostate to provide a definitive diagnosis but, if cancer is found, it will also be possible to say whether it is unilateral (confined to one side of the prostate) or bilateral, and to estimate how likely it is for the cancer to become active.
What percentage of the cores were positive (had cancer cells)?Of the positive cores, what percentage of the tissue consisted of cancerous cells? These two questions are important because they provide information about how extensive the cancer is and the likelihood that it may have spread beyond the prostate. (The raw data are fed into statistical models, or nomograms, that predict probability of progression.)The answers to these questions will also determine whether you are eligible for active surveillance or whether you need to choose some type of treatment.

Consider the situation of a patient who sought me out for a second opinion about treatment. I’ll call him Joe Reilly. Mr. Reilly was seen by a respected urologist who has his own practice. The urologist sent Mr. Reilly to a small hospital for a prostate biopsy. During the biopsy, six biopsy cores were taken. The pathologist then read one of the cores as Gleason 3 + 3. In standard practice, this would have been enough for a recommendation of surgery to remove the prostate.

But Mr. Reilly wisely sought a second opinion. When he arrived at my office, he carried his medical records, including the pathologist’s report and the pathology slides. I reviewed the slides and sent them and the report to our own pathology department. That was when Mr. Reilly’s story got a bit more complicated.

Our pathologist looked at the tissue on the slides and saw cells that were atypical, but not abnormal enough to be classified as cancer. So I recommended that Mr. Reilly undergo a second biopsy. Our pathologists do saturation biopsies, which at our hospital involves taking 20 cores from different areas of each side of the prostate (see “Prostate biopsy guidelines”).

The second biopsy was done three months after the first. When our pathologist reviewed Mr. Reilly’s tissue, he found 14 positive cores, many of which had cancer cells in more than 50% of them. Our pathologist assigned a Gleason score of 3 + 3 or 3 + 4 to these samples. We also found evidence of perineural invasion — invasion around the nerves of the prostate gland — meaning that the cancer could already have begun to spread.

Based on these findings, it was clear that Mr. Reilly needed treatment for more advanced disease, and I recommended a combination of hormonal therapy and radiation to keep his cancer contained. This at least spared him the risks and discomforts of surgery, which is appropriate only for localized disease. Mr. Reilly may also participate in an investigational protocol involving chemotherapy.

There are two possible explanations for the difference between the first and second biopsies. It may be that the cancer had exploded in the three months between the first and second biopsies. However, this was unlikely, given that Mr. Reilly’s PSA levels remained stable during that time. The second possibility is that the initial biopsy had not been adequately done. This happens — sometimes due to lack of expert technique or incompetence, but more often due to chance. The evidence indicates that only 77% of prostate cancers are diagnosed during a first biopsy; another 14% will be found only after a second biopsy. (Additional cancers are found during third and even fourth biopsies.)

The point is that it’s important to ask detailed questions about your Gleason score and the pathologist’s findings. Understand that each pathologist may interpret the information differently, assigning a slightly different Gleason score to the same tissue. Sometimes these differences in interpretation are inconsequential, but it’s always wise to seek out a second opinion about the initial findings — and even to consider a second biopsy.

Prostate biopsy guidelines

Urologists continue to debate which type of prostate biopsy is best and how many cores to sample. Most doctors agree that at least 10–12 core samples should be removed during a first biopsy to provide the pathologist with enough tissue to examine. Less agreement exists about the need to remove even more cores by doing a saturation biopsy, which typically takes 12–14 cores (and sometimes as many as 20). Although saturation biopsies are often recommended for anyone undergoing a second prostate biopsy, several investigators have questioned whether they should be recommended for an initial biopsy.

For more information, see:

Jones JS, Patel A, Schoenfield L, et al. Saturation Technique Does Not Improve Cancer Detection as an Initial Prostate Biopsy Strategy. Journal of Urology 2006;175:485–8. PMID: 16406977.

Moon K and Theodorescu D. Does Saturation Biopsy Improve Prostate Cancer Detection? Nature Clinical Practice Urology 2006;3:468–9. PMID: 16964184.

How to evaluate cancer staging

Because staging is so important, and yet so complicated, it’s important for you to better understand both the basics of cancer staging and how new mathematical models known as nomograms are helping us to more accurately stage prostate cancer.

Clinical staging

Clinical staging is done based on the findings from your physical examination and DRE, your prostate biopsy, and — to determine whether the cancer has already spread — results of other tests. For example, a computed tomography (CT) scan of the abdomen and pelvis may be ordered to search for evidence of cancer in the lymph nodes. A bone scan may be ordered to determine if the cancer has spread to the bones. And an endorectal magnetic resonance imaging (MRI) scan may be ordered to determine whether the cancer has spread beyond the prostate capsule. Your doctor will use this information to come up with a clinical stage.

Most doctors today use the tumor, nodes, and metastases, or TNM, system to stage prostate cancer. Although prostate cancer can range from stages T1 (the smallest) to T4 (the largest), early-stage prostate cancers are those that are defined as T1a, T1b, T1c, and T2 (see Figure 2). Most men diagnosed following an abnormal PSA test have tumors classified as T1c — cancers that are too small to be felt through a DRE and that are causing no symptoms whatsoever.

Likewise, numbers are assigned to indicate whether any cancer-containing lymph nodes (N) or metastases (M) are detected. A patient with an early-stage tumor, with no evidence of positive nodes or metastases, would have a clinical stage of cT1c N0 M0 (the first “c” denotes clinical stage). If there is a palpable abnormality in the prostate gland detected during physical examination, two of the nodes tested positive, and a bone scan found evidence of metastasis, the clinical stage would be cT2 N1 M1.

It’s important to remember that clinical staging is the best estimate of how advanced a cancer is, one made using the available tools. As technology advances, clinical staging may become more precise.

Figure 2: Early-stage prostate cancer

T1: This refers to a cancer that is so small that it cannot be detected through a DRE; instead, it may be found by accident, while operating on the prostate for some other reason (as for treatment of benign prostatic hyperplasia), or via a PSA test.

T2: Cancerous tumor that can be detected through DRE, but which is still confined to the prostate gland.

T1a: A small amount of microscopic cancer detected during treatment for BPH

T1b: Microscopic cancer of a higher grade detected during treatment for BPH

T1c: Cancer detected by PSA elevation, with a normal physical exam

T2a: Palpable, but less than one half of lobe

T2b: Palpable and found in more than one half of lobe (pictured above)

T2c: Involves both lobes (not pictured)

Pathological staging

Pathological staging is more accurate than clinical staging. If the CT or MRI scans reveal anything unusual, your doctor may perform a needle biopsy to sample lymph nodes and check for any evidence of cancer. But if these imaging tests reveal no abnormalities, pathological staging may not take place until your doctor performs surgery to remove the prostate. At that time, a doctor removes not only the prostate gland but also tissue from surrounding structures, such as the seminal vesicles and lymph nodes. Following surgery, pathologists examine not only the prostate but also these other excised tissues surrounding the prostate gland — known as the “margin” — for evidence of any cancer cells. A positive margin or evidence of cancer in any lymph nodes indicates that the cancer has spread beyond the prostate.

In many cases, the pathological stage will be the same as the clinical stage. But in other cases, the pathological staging will reveal that a cancer was clinically understaged. In effect, this means there was more cancer than the doctor realized — and the treatment originally recommended may not be enough to eradicate or contain the disease.

Understaging occurs with alarming regularity. By some estimates, as many as 40%–50% of men whose clinical staging indicates that they have early-stage cancer confined to the prostate actually have more extensive disease that will not be eradicated with surgery or radiation.

An example may make the issues clearer. Let’s say Bob Hadley undergoes a complete clinical workup after his doctor detects a prostate tumor through a DRE. Because the tumor is palpable, it is at least a stage T2 tumor. A CT scan of the lymph nodes and a bone scan, however, both come back negative, and his doctor assigns a clinical stage as follows: cT2 N0 M0.

Mr. Hadley undergoes a radical prostatectomy to remove the diseased prostate, including the removal of a greater number of lymph nodes than were originally seen on the CT scan. When a pathologist examines the excised tissue, however, he finds not only a positive margin — indicating that the prostate cancer has expanded beyond the confines of the prostate capsule — but also discovers that cancer is present in two lymph nodes. As a result, the pathological stage of the cancer is changed to pT3 N1 M0 (the small “p” denotes pathological stage). Significantly, this means treatment must also change. Mr. Hadley went into the operation believing he had early-stage localized disease, but now it’s clear that his cancer may already have spread — and he needs to readjust treatment.

Of course, it’s also important to note that it’s possible that during Mr. Hadley’s operation, the surgeon will detect palpable lymph nodes, indicating that the cancer has already spread beyond the prostate. In most cases, the surgeon will likely stop the operation, as there is no longer a need to remove Mr. Hadley’s prostate. Instead, systemic therapy will be necessary.

Prostate nomogram online

Memorial Sloan-Kettering Cancer Center has developed an online prostate nomogram you may be able to use if you have been diagnosed with localized prostate cancer and want to better understand your risk and treatment options. The calculator is not for everyone; for example, it is not intended for men whose cancer has probably already spread by the time of diagnosis.

To use it, follow the instructions and plug in your PSA level, Gleason score, and other medical information as prompted. The calculator helps determine which treatment approaches will most benefit you.

The calculator is available at

Do a nomogram check

A nomogram is a mathematical model that predicts the probability that someone with your clinical profile — PSA level, Gleason score, and tumor size — has cancer that is confined to the prostate or is the type likely to become active. Nomograms are therefore important in helping to determine what your prognosis is and which type of treatment will provide you with the best chances of recovery.

A number of nomograms exist, but two in particular are useful in providing a personal risk assessment so that you can make an informed treatment decision. Some are available on the Internet, so if you know your PSA level, Gleason score, and clinical stage, you can plug in the numbers and get your own risk assessment of the likelihood of being cancer-free after certain treatments (see “Prostate nomogram online”).

Partin tables. These tables use PSA level, Gleason score, and estimated clinical stage to help predict pathological stage and probability that the cancer has spread. The tables are based on information collected from thousands of actual patients, but are continuously updated to reflect new data.

Kattan nomograms. These nomograms come in various forms, but use data collected from thousands of patients to predict probability of disease stage. The Kattan nomogram most often used predicts the chances of biochemical recurrence (as measured by PSA) after a radical prostatectomy for someone with your clinical profile.

Even if you never actually see a nomogram, there are some general parameters you can use in determining whether it is likely that your cancer has been accurately staged or whether there is a chance it has been understaged (and therefore requires different treatment). In general, as tumor size, PSA level, and Gleason score increase, so, too, do the chances that your cancer may have been understaged initially and is likely to become active in the future.

Once again, though, it’s important for you as a decision-maker to understand that there are no absolute predictors of your cancer’s stage and likelihood of its becoming active. Even the most sophisticated nomograms available today offer only an estimate of probability. And when it comes to making a treatment decision, probability of success is the best we can offer at this point.


  1. Chottu

    Hello. I am a Prostate Cancer Survivor that was diagnosed with Stage 1 prtsoate Cancer at age 58. My PSA before treatment was 7. My Gleason Score was 3+4 (7). I had 25 treatments of External Beam Radiation by Calypso. At the end of EBR treatment I had high dose Brachy Therapy directly to the prtsoate. I did not have seed implants. I finished treatment January 25th 2012. My PSA at my 6 month follow up was undetectable. Is there someone that will insure me?

Commenting has been closed for this post.