Twenty-Five Years of Active Surveillance

Twenty Five Years of Active Surveillance

Approximately 25 years ago this seminal paper, which I don’t think has gotten nearly enough recognition by Nick George in Manchester was published. The natural history of untreated localized prostate cancer, about 150 men with localized disease, most of them were diagnosed after TURP or open prostatectomy, more than 90%. There was no grading. There was a 4% prostate cancer mortality at seven years with no treatment.

This paper at the time felt like a bomb because the mortality was perceived to be a lot higher. He says I think quite pressingly any theory of prostate cancer to embrace a concept of constant biological tumor behavior implies a long quiescent period of growth. This should precede clinical manifestation of active metastatic disease. The quest for clinicians whether younger patients or less likely to die of other causes will always enjoy a similar period of latent grace. It was really I think one of the first papers to really promote conservative management when surgery was an option.

That was pure watchful waiting. There was no intervention. This concept of expected management with selective delayed radical intervention, to the best of my knowledge this came out of a meeting we had of my interdisciplinary group of radiation oncologists and myself at Sunnybrook around 1996. We were heavily influenced by that George paper and by the prior acceptance of non-intervention for T1A prostate cancer after a TURP.

That was another experience that had gotten forgotten. When I trained, it was widely accepted that not only did you not treat microfocal Gleason 6 prostate cancer when it was found on a TURP. You didn’t even tell the patient about it. The word cancer was not mentioned. Somehow we forgot that and got into this tremendous enthusiasm for treating every single patient. We were uncomfortable with that.

The basic reasoning was most patients with advanced disease have a high PSA. If you have patients with a low PSA and they have indolent disease, maybe the PSA will give you a signal that they have more biologically aggressive disease and treat on the basis of that. We explicitly used PSA doubling time as a criteria for intervention. After about five years of this, we had our first publication in 2002. Richard Choo, a radiation oncologist now at the Mayo was the first author. That really led to an explosion of interest around the world, 74 publications from our group. A couple of weeks ago I Googled active surveillance in prostate cancer, and 1,764 publications, 26 prospective cohorts have been reported on around the world.

What has changed since then? Even 2002 is awhile ago in our field, which changes so rapidly. Obviously, greater recognition of the over-treatment problem, the Preventive Services Task Force identified this and wider acceptance of the principle. A lot of information about how Gleason 3 differs from Gleason 4 and the key messages that from a genetic perspective Gleason 3 in most cases looks likes normal prostatic epithelium and Gleason 4 looks like cancer.

Insight into the bad actors, the wolf in sheep’s clothing, called high-grade disease. We know a lot more about this than we did. I should just add that French paper that Neil quoted that 72% of patients with prostate caner have significant disease is purely a function of how you define clinically significant disease. It’s all based on the Stamey criteria 0.5 cc of low grade cancer. We have moved way beyond that. That statement, Neil, I think stands up given what we know today about Gleason 6 prostate cancer. A much too stringent definition. This has been redefined for Gleason 6 and is above about 1.5 cc. Probably almost all Gleason 6, if that’s all the patient has, don’t pose a threat.

The limitations of PSA kinetics, we relied on this for the first decade. It was really the Hopkins group that raised the flag about PSA kinetics and fundamentally sensitive, but not specific. The bad patients tend to have worse PSA kinetics, but so do a huge proportion of patients with very indolent disease. It’s subclinical prostatisis or a healthy bit of sexual acrobatics or something can cause a transient rise in PSA and give a signal for intervention if you’re not careful.

This is now a flat for further analysis of the patient, but not a trigger for intervention. The game changer multiparametric MRI, we’re going to hear more about that. Modeling studies that show that even if the expected prostate cancer mortality with initial conservative management it much higher than anyone has reported it, the advantages of conservative management are so compelling, at least given the alternatives of surgery and radiation, that active surveillance is a robust approach.

Long-term mortality data. The criticism really from the get-go has been you are naïve. You follow those patients long enough as per Johanssen [phonetic], who’s now reported 30-year outcome data with watchful waiting, they’re going to start dying like flies and you’ll be sorry. It took a long time to reach maturity to be able to say we now have a fair bit of confidence. That’s not going to happen. A number of studies including ours are now reporting patients with roughly 15-year followup. There’s no cliff that these patients are falling off. The difference obviously is offering radical treatment to the ones who get reclassified as high risk.

The REDEEM study that showed a benefit of dutasteride in men compared to placebo who are on surveillance of 44% reduction in the signal for radical intervention and the guys on the dutasteride. That’s a lot. The field has come a long way and it continues to evolve.

Of course we have the Task Force recommendation against screening. We may not like it, but it’s there and it was over-diagnosing and over-treatment as really targeted as the real problem. Arguably, this was a wake-up call to stop over-treating the low-grade patients and take our heads out of the sand and start doing things differently.

Just a reminder this was another seminal paper. The Sakr data from Detroit. Autopsy study. The likelihood of harboring microfocal prostate cancer is your age as a percentage: 30% in the 30s, 40%, 50%, 60%, 80%. Part of the aging process. Not a real disease Gleason 6 prostate cancer. I think those are very plausible statements given this kind of information.

The hallmarks of cancer. Does Gleason 6 have it? You all know there’s been this debate. Should we call it cancer? Is it really a cancer? I’ve come to conclusion although I’ve actually debated that it’s not cancer a number of times. Like basal cell carcinoma and like some brain tumors that never metastasize, but are still unequivocally cancer largely because of their ability to invade, clearly Gleason 6 can invade locally perineural invasion, capsular invasion. By those reasonable pathology criteria, it qualifies as what we call a cancer, but it doesn’t metastasize. That is because if you look at the hallmarks of cancer, proliferation genes, invasion genes, apoptosis, resistance, angiogenesis, you name it, Gleason 6 either the gene is expressed normally or if it’s a tumor suppressor gene, it’s present. If it’s an oncogene, it’s not over-expressed versus Gleason 4 is the opposite. There are exceptions. TMPRSS2-ERG present in lots of Gleason 6, but doesn’t confirm a metastatic phenotype.

Just two quick papers. Someone showed this earlier. The Eggener paper, 12,000 Gleason 6 twenty years after their radical prostatectomy. Essentially none of them metastasized. The figure in the paper is 0.2%. The ones we reviewed in that paper from Memorial all have higher-grade cancer on re-review. It really I think has become very clear. Metastasis from Gleason 6 is not the issue. It doesn’t do that.

Then the Ross paper, 14,000 pathologically confirmed Gleason 6, 22 cases with positive lymph nodes around 0.5%. That’s still a few, but they all were upgraded on re-review, so not a single case out of 14,000. True Gleason 6. That is not Gleason 66 on the biopsy, but no higher grade cancer harbored in the prostate and nonmetastasized, so how much more evidence do you need?

Of course biology is rarely so black and white. I mean how perfect can histologic evaluation be? PTEN is one of those things that creates a little concern. PTEN’s important in the progression of prostate cancer, a PTEN deletion that is. In one study about a third of patients who were Gleason 6 had PTEN deletion.

There are clearly, and we’ve learned from the GPS and Prolaris data that there are patients with histologic low-grade cancer that harbor worse genetic characteristics and may go on to behave in a more adverse fashion. But at least these isolated genetic abnormalities don’t seem to confirm a metastatic potential. Part of that is the heterogeneity. This is PTEN. You have one gland here. Half the gland is positive; half is negative. You have areas that are positive and negative varying through the gland. This profound heterogeneity, of course it’s not going to be black and white.

Then this N of 1 case that the Hopkins folks published in JCO last year of a guide. Of course they did annual biopsies and they’re men on surveillance. This guy was 64, Gleason 5, 12 biopsies over 12 years. They were all negative or only Gleason 6. They finally said you’ve reached age 75; we’re going to leave you alone for awhile. Then his PSA jumped up. They rebiopsied and Gleason 9. He died. Metastatic disease.

Here’s the thing. The PTEN present in all. The 12 biopsies then absent. There’s a complete switch, ERG switch, P53 switch, Ki-67. This was a brand new tumor. This might’ve developed from normal prostatic epithelium for all we know.

There are these bizarre cases. They shouldn’t drive our management. The idea is that you’ve got this kind of bifurcated pattern of development, a proliferative phase just like bladder cancer that goes to Gleason 3, but doesn’t metastasize. A dysplastic growth pattern that goes to high-grade cancer.
Rarely –this is the Haffner case–you may get progression from 6 to higher grade cancer. That is not the norm. It means you’ve got to follow these patients, but you don’t need to treat them.

We think the rate of true biologic grade progression based on our data that relates the time to upgrading, time from the original biopsy to upgrading’s about 1% per year. Others have estimated around the same proportion. That means you have to follow them after 10 years, but 10% of Gleason 6 may progress to higher grade cancer. If they are now 80 years old, I don’t think it matters. If they’re 65, maybe it does. Long-term followup, but not necessarily treatment.

The real problem is to identify the wolf in sheep’s clothing. Then 30% of these Gleason 6’s harbor higher grade cancer. How do you find those guys better? I’m just going to show this. It was just published actually last month in JCO or updated. About 1,000 patients. Third of these patients have had intervention. The shape of this curve is a clue to what the problem is. What we really want is this curve. We want to find that 30% within the first year and treat those guys. Then we should see a 1% slope of this curve out here.

Of course this original data was not done with MRI or biomarkers. This was just done with repeat biopsy, PSA kinetics. This window here is how we can improve where MRI and biomarkers are going to come into it. About a third have died. There’s a grand total of 15 deaths, 1.5%, and another 12 metastases, so 5% actuarial mortality at 15 years.

Then this is the really new data that I think really tells us where we can improve. This is the 27 men who have had metastasis in that group. Fifteen of them have died of prostate cancer and some of them are still alive. You can see these early guys who metastasize within the first five years were probably metastatic from the get go. These are the patients for whom surveillance has failed. They died of metastatic disease despite having more than a five-year period before they developed the metastasis when maybe curative therapy would’ve made a difference. That is about 1.5 to 2% of the whole cohort. Each one of those is a tragedy, but that’s where we can improve.

Prospective series. Most of them don’t yet have the maturity to report on the death rate, but it’s coming.

Biomarkers. We’ve heard about that. The GPS. I think they probably will have a role, but it’s going to be integrating them with MRI, which is the real challenge. Finally, where the research opportunities are, we can better stratify the low-risk patients. We did a pretty good job over the last 15 years. We can identify the intermediate-risk patients who are surveillance candidates because there are clearly lots of those with small volume Gleason 4. Earlier identification of wolves. The 30% with the higher grade cancers is the unmet need.

I think too many TRUS-guided biopsies is also an unmet need. No one likes those and they have lots of limitation. Whether it’s going to be MRI or biomarker, going forward I think we’ll be doing less of these biopsies.

Some psychosocial studies. It’s not clear which patients are not suitable. Optimizing the message. There’s room for marketers here to get patients off that knee-jerk response that they have cancer and must be treated. Finally, I think focal therapy is going to play a role for some, although not all of these patients.

Bratt O, Folkvaljon Y, Loeb S, et al. Upper limit of cancer extent on biopsy defining very low-risk prostate cancer. BJU Int. 2014 Jul 23. [Epub ahead of print]
http://www.ncbi.nlm.nih.gov/pubmed/25053197

Cuzick J, Berney DM, Fisher G, et al. Prognostic value of a cell cycle progression signature for prostate cancer death in a conservatively managed needle biopsy cohort. Br J Cancer. 2012 Mar 13;106(6):1095-9.
http://www.ncbi.nlm.nih.gov/pubmed/22361632

Eggener SE, Scardino PT, Walsh PC, et al. Predicting 15-year prostate cancer specific mortality after radical prostatectomy. J Urol. 2011 Mar;185(3):869-75.
http://www.ncbi.nlm.nih.gov/pubmed/21239008

George NJ. Natural history of localised prostatic cancer managed by conservative therapy alone. Lancet. 1988 Mar 5;1(8584):494-7.
http://www.ncbi.nlm.nih.gov/pubmed/2893917

Haffner MC, De Marzo AM, Yegnasubramanian S, et al. Diagnostic challenges of clonal heterogeneity in prostate cancer. J Clin Oncol. 2015 Mar 1;33(7):e38-40.
http://www.ncbi.nlm.nih.gov/pubmed/24638011

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011 Mar 4;144(5):646-74.
http://www.ncbi.nlm.nih.gov/pubmed/21376230

Ross HM, Kryvenko ON, Cowan JE, et al. Do adenocarcinomas of the prostate with Gleason score (GS) ≤6 have the potential to metastasize to lymph nodes? Am J Surg Pathol. 2012 Sep;36(9):1346-52.
http://www.ncbi.nlm.nih.gov/pubmed/22531173

Sakr WA, Grignon DJ, Haas GP, et al. Age and racial distribution of prostatic intraepithelial neoplasia. Eur Urol. 1996;30(2):138-44.
http://www.ncbi.nlm.nih.gov/pubmed/8875194

Wilt TJ, Brawer MK, Jones KM, et al. Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med. 2012 Jul 19;367(3):203-13.
http://www.ncbi.nlm.nih.gov/pubmed/22808955