Exceptions to "early salvage" radiation treatment for recurrence after prostatectomy

Three major randomized clinical trials and a meta-analysis have proved that for most men waiting for early signs of recurrence after prostatectomy (e.g., 3 consecutive PSA rises or PSA of 0.1 or 0.2 ng/ml) to give radiation gave the same outcome as immediate ("adjuvant") radiation (see this link). But there are exceptions. In some men, adjuvant treatment is better. In some men, early salvage may overtreat them.

Adjuvant Radiation Therapy

Tilki et al. did a retrospective study of 26,118 men given prostatectomies at several hospitals in Germany, UCSF, and Johns Hopkins. 2, 424 of them had "adverse pathology" defined as:

• positive lymph nodes, or
• Gleason score = 8-10, and
• Stage T3 or T4

Patients were treated with adjuvant (within 6 months of prostatectomy) radiation therapy (ART), salvage radiation therapy (SRT) after PSA rose above 0.2 ng/ml (biochemically recurrent - BCR), or no radiation therapy. They matched patients on age, initial PSA, and positive/negative margin status. 10-year all-cause mortality was:

• for men with adverse pathology including positive lymph nodes:
• 14% for ART
• 27% for no RT
• 28% for SRT
• for men without positive lymph nodes:
• 5% for ART
• 25% for no RT
• 22% for SRT
• for men with no adverse pathology:
• 8% for ART
• 9% for no RT
• 8% for SRT

This suggests that for men with adverse pathology, ART improves outcomes over early SRT.

Delela et al. found that a high Decipher score can tip the balance toward adjuvant radiation.

No/Delayed SRT

At the other end of the risk spectrum are men with such low risk for clinical recurrence, that salvage radiation can be delayed, perhaps indefinitely. This is based on the observation that while 40% of post-prostatectomy patients may experience a BCR, only 30% of BCR patients develop a clinically relevant recurrence, and all but 16% die of something else before the recurrent cancer kills them. We saw the results of a retrospective study that suggested that those at lower risk of progression (low PSA, Gleason score, and stage) and receiving a higher dose of SRT may not need adjuvant ADT. 

In a major review for the European Urological Association, Van den Broeck et al. reviewed 77 studies covering 20,406 patients who were biochemically recurrent (conventionally measurable PSA) after prostatectomy. They sought to define the patient and disease characteristics that determined which of the BCR cancers led to distant metastases and death from prostate cancer. They found that the following risk characteristics defined a "low risk" BCR prostate cancer that could be safely watched:

• PSA doubling time > 1 year
• Gleason score < 8
• Interval to biochemical failure > 18 months

Tilki et al. validated the EAU study in a retrospective study of 1,125 patients. Preisser et al. validated the study retrospectively among 2,473 men. Pompe et al. validated the risk group in a retrospective study of 1,821 men. To date, there has been no prospective validation in a randomized clinical trial.

(update 5/4/23) Tilki et al. found that if there was one high-risk feature (stage T3/4 or GS 8-10) and PSA ≤ 0.25, salvage radiation provided no survival benefit. However, the survival benefit was significant if PSA>0.25. This suggests that post-prostatectomy patients with a high-risk feature must not wait for the results of a PSMA PET/CT (which usually is uninformative below PSA of 0.5), but should have salvage radiation as soon as they are biochemically recurrent.

Zaorsky et al. point out some additional characteristics of recurrent patients who may be safely watched:

• PSA < 0.5 ng/ml at time of recurrence
• Age > 80 years of age
• Significant comorbidities
• No distant metastases detected with PET/CT imaging (Ferdinandus et al)

It is undoubtedly better to have a low Decipher score as well.

Lacking prospective validation, this is a decision that should be carefully discussed between the patient and the radiation oncologist.

The definition of SECOND biochemical recurrence (after prostatectomy AND salvage radiation)

There is no standard definition of SECOND biochemical (PSA-detected) recurrence (BCR); that is, after both prostatectomy and salvage radiation (SRT). There are two reasons to have a standard definition of second BCR:

1. Time for next treatment: BCR (after any treatment) is the first indicator of treatment failure, and a signal that it may be time to consider additional treatment. It is not at all clear that immediate additional treatments are beneficial. If a treatment becomes the standard of care after biochemical failure, then it will be necessary to define the PSA or PSA doubling time (PSADT) at which that treatment should begin.
2. Comparison among radiation protocols: Lacking randomized clinical trials among all the variables of when and how salvage radiation are given (pathological characteristics, PSA, PSADT, radiation dose, adjuvant ADT, prostate bed radiation, radiation of pelvic lymph nodes), we can only look at effectiveness across studies to help us hypothesize that one strategy might be better than another. It helps if we have a consistent definition of success.

Miyake et al. looked at three definitions of second BCR:

1. NARA definition: PSA never falls below 0.2 ng/ml; or, it falls below 0.2 ng/ml but later rises over it in two consecutive readings.
2. RTOG 9601 definition: any post-SRT PSA over 0.5 ng/ml; or, nadir + 0.3 ng/ml; or the start of hormone therapy.
3. GETUG definition: nadir + 0.5

They evaluated 118 patients using the 3 definitions. With 49 months of median follow-up after salvage radiation:

• The Nara definition had the highest rate of second BCR; 53%, 45% and 40% for Nara,  RTOG and GETUG respectively.
• Gleason score and Pre-SRT PSA independently predicted Nara BCR, while negative margins and PSADT also predicted RTOG and GETUG BCR.
• There were no discrete cut-offs of the patient characteristics that reliably predicted BCR by any definition

It's worth noting that the definitions may differ for study entry and endpoint (it is usually called "biochemical progression" when used as an endpoint). Many clinical trials use the 0.2 ng/ml definition for the second BCR too.  This trial used PSA≥ 0.2 ng/ml or 3 consecutive rises after RP or SRT. We recently saw that another RTOG trial, the SPPORT trial, used a BCR endpoint definition of nadir + 2 because it correlated well with clinical recurrence. This is the definition advocated by PCWG2.

It is sometimes necessary to define a THIRD BCR as an endpoint to determine whether a therapy that began after a second BCR was successful. For example, an ongoing trial of hormonal therapies for SRT-recurrent men uses a second BCR definition of PSA > 0.5 ng/ml and PSADT ≤ 9 months, and a third BCR definition of a confirmed 25% rise in PSA and nadir + 2 during therapy, and a fourth BCR definition of a confirmed PSA > 0.2 after hormonal therapy.

The definition for FIRST BCR of a confirmed PSA after prostatectomy of 0.2 ng/ml was an artifact of the current lowest discernible PSA before the 21st century, which was 0.1 ng/ml at the time. The American Urological Association decided that anything higher than that would be deemed a BCR. The "ARTISTIC" meta-analysis established that for most patients, it was unnecessary to establish a PSA for first BCR below 0.1 or 0.2.

BCR is just one of a number of elements to be evaluated after SRT. A BCR with a high Decipher score may suggest that immediate salvage ADT is appropriate. With the new generation of PET scans, which can detect metastases at low PSAs, it may sometimes be beneficial to treat pelvic lymph node metastases and possibly distant metastases if SRT had only included the prostate bed.

This small, retrospective study will not establish a new definition, but it does raise the interesting question of whether we need a standard definition, or whether the definition ought to depend upon the purpose for which it is used. If we have definitive evidence that early treatment after failed SRT is beneficial, that will force researchers to investigate the optimum PSA (or PSADT) cutpoint. Until then, the PCWG2 definition (PSA≥ nadir + 2.0), PCWG3 definition (PSA≥nadir+ 1.0), or the combination of PSA and PSA doubling time used in the EMBARK protocol (PSA≥1 and PSADT≤ 9 months) are all reasonable.

New Study: Adjuvant Radiation Saves Lives vs. Salvage

A major new study adds to several other studies that show that, for men with adverse pathology, adjuvant radiation (ART) within 3-6 months of prostatectomy saves more lives compared to waiting until the PSA rises into the range of 0.1-0.5 ng/ml - salvage radiation (SRT).

Three previous randomized clinical trials have shown an advantage to adjuvant radiation over a "wait-and-see" approach. However, only one of them (SWOG  S8794) showed that there was an improvement in freedom from metastases and overall survival attributable to earlier treatment. That study was limited in its generalizability because only a third of the "wait-and-see" cohort ever received salvage radiation. ARO-96-02 was designed to detect differences in progression-free survival (which were significant), but it was underpowered to detect overall survival differences. EORTC 22911 was designed to detect differences in progression-free survival (which were significant), but although it had a larger sample size, overall survival did not improve. Sub-group analysis showed the survival benefit was limited to men under the age of 70. A recent meta-analysis of the three trials showed that freedom from biochemical failure, freedom from life-long ADT,  and freedom from distant metastases were significantly improved by adjuvant treatment. But less than half of the men in the wait-and-see arms ever received salvage radiation, and 20-40% of  them never suffered a recurrence. All three trials used salvage radiation doses that would now be deemed too low. ART utilization rates are at an all-time low of 17% in men with adverse pathology.

What we really want to know is: what is the downside of waiting until the PSA rises to some arbitrary level, say 0.2 ng/ml? That is the subject of three randomized clinical trials, but we will not have the findings for several years. Meanwhile, some researchers looked at historical data in a new way to determine whether there is any evidence that might aid in decision-making.

Hwang et al. have pooled the databases from ten top institutions: Massachusetts General, Cleveland Clinic, University of Michigan, Duke University, Washington University, Mayo Clinic, University of Chicago, University of Miami, Virginia Commonwealth University, and Thomas Jefferson University. There were 1,566 patients who were treated between 1987-2013. Patients either had fully contained prostate cancer (T2) with a positive margin or extraprostatic extension (T3a)/ seminal vesicle invasion (T3b) with or without a positive margin.

They used a statistical technique called "propensity score matching" that in some respects resembles what would have resulted from a prospective randomized trial. Every patient who had ART was matched, in terms of patient characteristics, to a patient who had SRT. Patients are chosen randomly from among those with matched characteristics.  Patients were matched on age at surgery, year of surgery, Gleason score, T stage, margin status, postoperative ADT, and pelvic nodal RT. In this way, they were able to generate 366 matched pairs of patients. This technique works quite well in predicting outcomes of prospective randomized trials as long as there is a large enough sample size, considerable overlap in patient characteristics (which there was) and there aren't any prognostic patient characteristics that were missed.

The researchers found that all measured outcomes were significantly better among those who received ART:

• 12-year freedom from biochemical failure: 69% for ART vs. 43% for SRT
• 12-year freedom from distant metastases: 95% for ART vs. 85% for SRT
• 12-year overall survival: 91% for ART vs. 79% for SRT
• Patients who suffered biochemical failure were more likely to have had SRT, have been stage T3b, have had higher Gleason score, had not been treated with lymph node radiation, and had not had postoperative ADT.
• The advantage of ART was only lost if more than 56% of them would have been overtreated, but based on nomograms, no more than 46% would have been overtreated (using the assumption that 2/3 were GS 3+4 and 1/3 was GS 4+3).

Pending confirmation by the randomized clinical trials, this study is our best evidence to date that ART is preferable to SRT. However, there are a few very important caveats:

• They defined SRT as treatment when the PSA is in the range of 0.1 - 0.5 ng/ml. (They actually call this "early" salvage -- a term I would prefer to reserve for radiation when the ultrasensitive PSA (uPSA) is below 0.10 ng/ml.) For uniformity reasons in this 10-institution study, any PSA below 0.10 ng/ml on an uPSA test was deemed "undetectable," and those treated at very low PSAs were considered to have had ART. They had to use those definitions in their analysis because of the heterogeneous data set with PSAs recorded as early as 1987 (before there were any ultrasensitive PSAs). Because the risk of overtreatment with ART is high (they estimate 33%-52%), it behooves patients to track their post-prostatectomy PSA with an ultrasensitive test. We have seen that for men with adverse pathology,  any uPSA over 0.03 ng/ml reliably predicts that it will keep going up to 0.2 ng/ml (see this link). In men without adverse pathology, only a convincing pattern of PSA rises is prognostic.
• Adverse pathology in this study included anyone with positive margins, but others advocate that the length of the positive margin and the Gleason score at the margin are important considerations. A patient with focal positive margins and GS 6 at the margin may never need additional ART or SRT.
• They lumped together men whose PSA was undetectable but then climbed higher and men whose PSA was persistently elevated after prostatectomy. Persistent PSA with adverse pathology is a clearer indicator that gross amounts of cancer were left behind and calls for some quick action.
• The Decipher genomic test was not available throughout most of the study period. For those sitting on the fence, it may be a decisive factor.
• The newer PET scans (Axumin and PSMA-based) can find metastases if PSA is greater than 0.2 ng/ml. Multiparametric MRI may be able to find sites in the prostate bed or among the pelvic lymph nodes where tumor size is longer than 4 mm. Because of the advantage of earlier treatment, most men will require treatment before metastases become detectable. Some will be overtreated if the cancer is already systemic.
• Among very high risk patients (i.e., GS 8-10, seminal vesicle invasion (T3b) or invasion of nearby organs (T4), and very high persistent PSA) the probability that ART or SRT will be curative may be very low. Patients should understand what the population-based risk is from a nomogram.
• The radiation doses delivered were at a median dose of 66 Gy. More recent evidence suggests that higher doses may be necessary to achieve a cure. The value of adjuvant ADT and whole pelvic radiation suggested here has also been suggested by a number of other studies.
• This study excluded patients with detected positive lymph nodes. That is a clear indication for ART.

There are many factors to consider including comorbidities, continence and potency recovery. This will seldom be a straightforward decision. Patients with adverse pathology and uPSA over 0.03 ng/ml should be talking to a radiation oncologist and not a urologist.

Most of the recurrences after primary radiation failure are salvageable

Salvage therapy is curative in about half of men who have a biochemical failure after primary therapy. That's true whether the primary therapy was surgery or radiation. It's true when the salvage therapy was radiation after surgery. And it's true whether the salvage therapy was surgery, cryotherapy, or brachytherapy after radiation. Salvage success rates can be as high as 3 in 4, in certain well-selected patients treated with appropriate therapies (see this link, for example), but it can be a lot lower too. Salvage therapy always increases the complications over what they were for the primary therapy, so we would avoid it if we knew it was likely to be futile. Thanks to the new generation of PET scans, we are beginning to understand why, and what we may be able to do to improve those odds.

For any salvage therapy to be effective, two conditions must be met:

1.  The recurrence must be local. Local means in the prostate, seminal vesicles, the prostate bed, nearby organs (e.g., bladder, rectum, etc.), and/or in the pelvic lymph nodes.
2.  The recurrence must not be distant. Distant means metastases in the bones; remote organs like the lungs, liver, or remote lymph nodes; or in systemic circulation in the bloodstream.

In the past, it has been difficult to ascertain that both conditions were met. Bone scans are not very reliable when the PSA is below 20 ng/ml, and they are not specific for metastases. Moreover, by the time the PSA increases that much, the cancer is almost certainly distant and incurable. The NaF18 PET/CT scan can detect metastases sometimes at a PSA as low as 4 ng/ml, but it only detects bone metastases, and it is not specific for metastases. An Ultra-Small Superparamagnetic Iron Oxide (USPIO) MRI may sometimes detect metastases, but only in lymph nodes.  A multiparametric MRI may sometimes detect local recurrences, and may be used to target areas for biopsy in the prostate and prostate bed. It may be reliable after primary radiation (see this link). However, it tells us nothing about distant metastases.  CT scans only detect the larger lesions that may be suspect. A transperineal template mapping biopsy may detect prostate cancer in the prostate, but tells us nothing about distant metastases. It should be noted that biopsied prostate tissue looks very different after radiation, and it should be analyzed by highly experienced pathologists.

Clinical trials have proved that adjuvant radiation after prostatectomy has better outcomes than waiting, and recent studies suggest that overtreatment may be avoided by using early salvage radiation rather than adjuvant radiation therapy. Perhaps early salvage therapy after primary radiation therapy may have improved outcomes too. That is, it may be more successful if started before the patient's PSA reaches the nadir+2 level, which is the official definition of biochemical recurrence after primary radiation therapy.

The FDA-approved C-11 Choline PET/CT (or the similar C-11 Acetate PET/CT) fills some of the critical information gaps. It can detect prostate cancer in the radiation-treated prostate, the local area, and throughout the entire body at a PSA as low as 2 ng/ml, especially if the PSA has been rapidly rising. However, its sensitivity is not very good for small sites of cancer (they must be larger than 5mm), or cancer in lymph nodes. And when used to detect cancer within the prostate, prostatitis and BPH may generate false positives. Some of the new experimental PET scans (e.g., DCFPyL) may be more sensitive. Now that we have an adequate tool for detecting both of the above-mentioned conditions (local and not distant), we are beginning to be able to select which recurrences can be cured with salvage therapy, and which can only be managed with lifelong hormone therapy.

Parker et al. report on the Mayo Clinic experience with 184 patients with rising PSAs after primary radiation therapy on whom the C-11 Choline PET/CT was used to detect local and/or distant prostate cancer progression.

• 87% of patients were PET-positive.
• The C-11 Choline PET/CT correctly identified 98% of patients who were later found to have residual prostate cancer on subsequent histological analysis. 
• However, 42% of patients that were identified as negative by the C-11 Choline PET scan later suffered from cancer progression - they were false negatives.
• Patients were especially likely to be PET-positive if they had higher pretreatment PSA, were high risk, had higher PSA level at the time of the PET scan, had a greater increase from nadir PSA, had a shorter PSA doubling time, and had a higher PSA velocity. All of those with PSA≥ 10 ng/ml were PET-positive.
• Risk category, PSA increase from nadir, and time since primary radiation therapy were independently associated with PET-positivity, and can help predict when recurrences are salvageable.
• 59% of PET-positive patients were confirmed by histological analysis (either biopsy or salvage prostatectomy). 76% were confirmed by a multiparametric MRI.
• 46% of those who were PET-positive had cancer only in the prostate and seminal vesicles. These patients were potentially salvageable with any of the salvage therapies mentioned above.
• An additional 16% (62% in total) had cancer in the soft tissue pelvic region. These may be salvageable with extended pelvic lymph node dissection (ePLND) or radiation in select areas of the pelvis that were not treated originally.
• While only a few patients (21) had a PET scan before their PSA reached nadir+2, half of them had a local recurrence only, and are potentially salvageable. This suggests that the  patient does not have to wait for nadir+2. However on this small sample, the salvageability does not seem to be very different for those who detect it earlier.

This study confirms the findings of the larger study at Memorial Sloan Kettering (MSKCC) (reviewed at this link).  In that study, 55% had a recurrence in the prostate and/or seminal vesicles only, compared to 46% at Mayo. At MSKCC, an additional 8% had recurrences in the pelvic lymph nodes only, compared to 16% at Mayo. There were important differences between the studies. At Mayo, unlike MSKCC, patients may have had brachytherapy as all or part of their primary therapy, they may have had enlarged lymph nodes from the start, they had significantly lower doses of radiation (76 Gy vs ~80 Gy), they were younger (65 vs 69), fewer had adjuvant hormone therapy (30% vs 54%), they all had rising PSA but not necessarily nadir+2, and they all received a C-11 Choline PET/CT, there was less histological confirmation (59% vs 71%), and the median follow-up time was shorter (68 months vs 83 months).

As noted in the commentary of the MSKCC study, these findings may not apply when the primary therapy used a very high biologically effective radiation dose, such as with brachy boost therapy, SBRT, or high dose rate brachytherapy.

It makes sense to rule out the possibility of distant metastases using an advanced PET scan. Even at a cost of $2,500 or so, it may save the patient much more than that for the cost of salvage therapy. However, unless the PET scan is done at Mayo using C-11 Choline, is done as part of the clinical trial using the newly FDA-approved PET indicator fluciclovine, or is one of the free ones at NIH, the out-of-pocket cost may be formidable. Hopefully, the FDA will approve more of them, and availability will expand. Unfortunately for those considering early salvage after a prostatectomy failure, none of them are accurate for PSAs that low (≤0.2 ng/ml).

The authors constructed a nomogram to help the prospective patient predict whether his recurrence, detected with a C-11 Choline PET/CT, is likely to be a salvageable recurrence or unsalvageable recurrence. In the first table, fill in the number of points that comes closest to your situation, and add them up. In the second table, look up the probability of a distant recurrence (unsalvageable) that comes closest to your total number of points.

Risk Factor	Points to assign	My Points
PSA increase from nadir	2 ng/ml: 13 5 ng/ml: 32 10 ng/ml: 63 15 ng/ml: 95
Years since RT	1 yr: 100 2 yrs: 95 3 yrs: 90 5 yrs: 80 10 yrs: 52 20 yrs: 0
Risk Group	Low: 0 Intermediate: 8 High: 45
	TOTAL

My Total Points	Probability of recurrence outside of the pelvic area
66	5%
88	10%
120	25%
153	50%
185	75%
216	90%
240	95%

This nomogram outperformed using a PSA threshold alone in its predictive power, and may help the patient decide whether potentially-curative salvage therapy or lifelong hormone therapy is the better course of action.

I'm not sure why radiation dose was not significantly correlated with the site of recurrence at Mayo (p = 0.1) as it was in the MSKCC study. In fact at MSKCC, those who received doses of at least 79.2 Gy had half the rate of recurrence compared to those who only received 75.6 Gy (which seemed to be the norm at Mayo). It may be that those who were treated at Mayo only received higher doses when their cancer was already systemic. We know that this is on the steep part of the dose/response curve where even a small increase in dose can increase its effectiveness greatly. Whatever the reason for the data discrepancy, higher doses do prevent local recurrences.

(update 11/18/2018) Hayman et al. reported on 49 men who had a biochemical recurrence after whole pelvic primary radiation therapy and long-term ADT who were clinically staged as node positive (N1) via MRI. Using imaging (probably a PET/CT scan) they found the site(s) of recurrence in 46 of the men:

• 25 (54%) had a recurrence in the prostate only
• None had a recurrence in lymph nodes only
• 21 (46%) had a recurrence that included a distant metastasis

This is very similar to Mayo and MSK.

note: Thanks to Dr. Will Parker for letting me review the full text of his published study.

Risk Stratification for Radiation Therapy

Risk stratification involves assigning patients to categories based on diagnostic risk factors. The goal is to identify those patients who are more or less likely to respond to specific therapies (or active surveillance).  It is an aid to judgment for the patient and doctor, and helps assess prognosis and define the standards of care. It also provides for consistency between research studies so that they are more comparable. Because we depend on those studies for treatment guidelines, we don’t want to change the risk categories frequently.

In 1998, Anthony D’Amico introduced the most widely accepted risk stratification system. It has since been tweaked somewhat by consensus of the National Comprehensive Cancer Network (NCCN). It mainly relies on 3 risk factors – PSA (in 3 groups), Stage (in 3 groups), and Gleason score (in 4 groups) to create 3 risk categories (low, intermediate and high) with 2 sub-categories in each of the three. The “very low risk” sub-category also includes number of positive cores, highest % cancer in those cores, and PSA density. The “very high risk” sub-category also includes number of cores with Gleason score 8-10.

There are competing risk stratification systems. UCSF, for example, uses a system called the CAPRA score that includes age and % positive biopsy cores. Each risk factor is assigned points, and the points are summed to determine the risk category. It is also possible to use nomograms based on historical statistics to help with prognosis. While nomograms will always produce a risk probability as a%, those probabilities may, in some cases, be projected off of a very small dataset and their accuracy is questionable.

A risk stratification system is created through a multistep process. The risk factors are assessed to find the ones that independently predict recurrence after treatment.  For example, stage, Gleason score and PSA, although they are somewhat correlated, independently predict recurrence. Those risk factors are then grouped such that the risk is about the same within the group, but is different between the groups. For example in the NCCN system, Gleason scores of 8, 9 and 10 are all roughly the same at predicting recurrence, but carry much greater risk than lower Gleason scores. Then the risk factors are combined (either by selection or by adding points) such that the risk is about the same within the risk category, but significantly different between risk categories.

D’Amico developed his risk stratification system based on data from patients treated from1989 to 1997. His dataset comprised 888 surgery patients treated at the University of Pennsylvania, as well as 766 patients treated with external beam radiation, 66 patients treated with LDR brachytherapy monotherapy and 152 patients treated with LDR brachytherapy plus ADT at the Joint Center for Radiation Therapy in Boston. He only looked at biochemical progression, which was defined as PSA≥0.2 ng/ml for surgery patients and 3 consecutive rises for radiation patients. External beam radiation was only 67 Gy – far below what is now considered curative. Biochemical recurrence after radiation has since been redefined because the previous definition over-predicted clinical recurrence. Radiation therapies did not include combined modalities, HDR brachytherapy, SBRT or proton therapy.

In 2007, Johns Hopkins validated D’Amico’s risk groups among 6,652 prostatectomy patients. In 2008, the Mayo Clinic validated D’Amico’s data among 7.591 patients treated with radical prostatectomy only. They also broadened outcome data to include clinical recurrence, evidence of systemic progression, overall and cancer-specific survival. I am not aware of any validation studies for external beam radiation or brachytherapy.

Because treatments and outcomes have changed so much for radiation therapies, it may be time to take another look at the risk stratification used for radiation therapy. An Italian group looked at data on 2,493 patients treated at 10 centers between 1997 and 2012. Patients were treated with a median dose of 76 Gy of EBRT and 62% also received ADT (almost half were high risk as defined by NCCN.) They call their risk stratification system the Candiolo Classifier. Like the CAPRA score system, it assigns points to each risk factor. Classification is based on the sum of those points.

They found that age and% positive cores at biopsy significantly added to their model’s ability to stratify the risk of patients. The following table shows the breaks that discriminated best, and the number of Candiolo points assigned to each risk factor.

Risk Factor	Candiolo  (points)	NCCN
Age	<70 (0) ≥70 (22)	NA
% Positive Cores	1-20% (0) 21-50% (29) 51-80% (50) 81-100% (81)	<3 positive cores, ≤50% cancer in a core, and PSA density <0.15 ng/ml/g used in “very low risk” definition. <50% positive cores used in “favorable intermediate risk” definition. >4 cores with GS8-10 used in “very high risk” definition.
PSA (ng/ml)	<7 (0) 7-15 (42) >15 (96)	<10 10-20 >20
Gleason scores	3+3 (0) 3+4 (35) 4+3 (48) 8 (76) 9-10 (106)	3+3 3+4 4+3 8-10 5+(5,4,or 3)
Stage	T1 (0) T2 (17) T3-T4 (58)	T1-T2a T2b-T2c T3a T3b-T4

They defined 5 risk classes that discriminated well with risk of biochemical recurrence. The following table shows the biochemical progression-free survival (bPFS) for each risk class at 5 and 10 years. The relationship is similar for clinical progression-free survival, systemic (metastatic) progression-free survival, and prostate cancer specific survival.

Risk Class	Points	5-yr bPFS	10-yr bPFS
Very Low	0-56	94%	90%
Low	57-116	85%	74%
Intermediate	117-193	80%	60%
High	194-262	67%	43%
Very High	263-363	43%	14%

The Candiolo system beat the 3-tiered (low, intermediate and high risk) NCCN system in predicting all measures of progression after external beam radiation. For bPFS, its concordance index (a measure of how accurate its predictions are) was 72% vs. 63% for the NCCN system. It predicted metastases and prostate cancer survival with an accuracy of 80% vs. 69% for the NCCN system.

The Candiolo Classifier certainly seems to be an improvement, but should be validated by another group of researchers before it gains wider acceptance. Ideally, we would also have data on risk categories suitable for other kinds of radiation therapy, boost therapies, use of adjuvant ADT, and whole-pelvic radiation.

This “new, improved” system raises some interesting questions:

• The D’Amico/NCCN risk stratification system is based on antiquated data and a small dataset for radiation. Is it time for a make-over?

• Do we have to have a single risk stratification system against which all therapies should be assessed? It certainly facilitates comparisons between therapies if we have a single system. However, different risk factors (e.g., age and % positive cores) may be important in determining the risk associated with one therapy but not another.

• At what point has our ability to measure risk factors changed enough that the entire stratification system should be altered? The ability of multiparametric MRIs and advanced PET scans to more accurately assess stage and to target biopsy cores to more suspicious areas may increase the detected risk beyond what it was when the system was first set. Also, the Gleason scoring system and the AJCC staging system has changed over the years.

• How do we maintain comparability with older clinical trials and with our databases if we change our risk stratification? Many trials were established a decade or more ago with pre-set risk groups. When the data mature, will they be hard to analyze? A similar effect occurred when biochemical recurrence after radiation was redefined by the Phoenix consensus in 2005. In many studies, both definitions were presented for a while.

• Can a stratification system from Europe gain acceptance in the US and particularly by the NCCN? How do we get widespread agreement on which system is the “gold standard.” As far as I know, the CAPRA Score is only used by UCSF, even though it is an NCCN member.

• What is the role of other biochemical measures? PHI, 4KScore, PCA3, Oncotype Dx and Prolaris all measure risk. Should any of them be used in a risk stratification system? Should first-degree relatives who have had prostate cancer be included as a risk factor? What about African-Americans? And how should PSA be counted when the patient is taking 5ARis (Proscar or Avodart) for BPH?

Johns Hopkins: ultrasensitive PSA after surgery predicts biochemical relapse

We’ve looked at several retrospective studies this year that found that early ultrasensitive PSA (uPSA) results following surgery can reliably predict eventual biochemical relapse. Johns Hopkins examined its own database and found the same thing.

The study by Sokoll et al.  looked at the records of 754 men treated with surgery at Johns Hopkins between 1993 and 2008 whose first post-surgery PSA, taken at about 3 months, was “undetectable” (<0.1 ng/ml). They reanalyzed the stored serum samples using an ultrasensitive PSA assay that could detect values of 0.01 ng/ml or higher. Each man was tracked until biochemical recurrence (BCR) – defined as PSA≥0.2 ng/ml – or for at least 5 years if there was no biochemical recurrence (median of 11 years).

• ·      Among men whose first uPSA was ≥ 0.01 ng/ml, about half eventually had BCR.

o   57% were BCR-free at 5 years, 49% at 11 years.

o   Mean BCR-free survival: 10 years

• ·      Among men whose first uPSA was < 0.01 ng/ml, 87% remained BCR-free.

o   92% were BCR-free at 5 years, 86% at 11 years.

o   Mean BCR-free survival: 15 years

• ·      Among men whose first uPSA was ≥ 0.03 ng/ml, 77% eventually had BCR.

o   27% were BCR-free at 5 years, 22% at 11 years.

o   Mean BCR-free survival: 5.5 years

• ·      Among men whose first uPSA was < 0.03 ng/ml, 85% remained BCR-free.

o   91% were BCR-free at 5 years, 84% at 11 years.

o   Mean BCR-free survival: 15 years

Other predictors of recurrence were the usual suspects: initial PSA, pathological stage and Gleason scores, and the presence of positive margins.

They additionally tracked a cohort of 44 men who’d had a cystoprostatectomy for bladder cancer in order to determine whether extra-prostatic sources of PSA might interfere with the uPSA test’s sensitivity to detect recurrent prostate cancer. All but two had uPSA<0.01 ng/ml, and those two had low values, 0.01 and 0.02 ng/ml.

So we see that a uPSA cutoff of 0.01 ng/ml on a first test is no better than a coin toss at predicting eventual BCR, and would lead to a great deal of overtreatment. On the other hand, a uPSA cutoff of 0.03 ng/ml correctly predicted eventual BCR 77% of the time. It missed about 15% of the men who would eventually recur, but was no worse than the lower cutoff in this regard. Clearly, a uPSA cutoff of 0.03 ng/ml is prognostic of BCR and a lower cutoff is not. The authors seem to miss this point in their conclusion.

Their analysis seems congruent with the other studies we’ve seen lately. Koulikov et al. also found that the 0.03 ng/ml cutoff was prognostic, but only when uPSA was increasing steadily. Kang et al. also found that a cutoff of 0.03 ng/ml at any time after surgery optimized BCR predictions with a median 18-month lead-time advantage among men diagnosed with adverse pathology (pT3 and/or positive margins). In a separate analysis among men with more favorable pathology (pT2, irrespective of margin status), Kang et al. found that a cutoff of 0.03 ng/ml on a first uPSA was predictive of later (median of 33 months) BCR.

While we await more definitive results from randomized clinical trials, there seems to be an emerging consensus that 0.03 ng/ml is the optimal uPSA cutoff. Using a lower cutoff for early salvage or adjuvant RT will lead to overtreatment, and there seems to be no risk attached to waiting for it. Provisionally, I believe it should be viewed as a surrogate for the traditional BCR definition.