Monday, August 13, 2018

Salvage Radiation Dose: Decision-Making Under Uncertainty

A large, well-done, confirmed randomized clinical trial (RCT) is the only evidence that proves that one therapy is better than another. According to current consensus, this is deemed "Level 1a" evidence. But this high level of evidence is seldom available. This is especially true of prostate cancer because it takes so long to achieve acceptable endpoints like overall survival, prostate cancer-specific survival, and metastasis-free survival. Such studies are very expensive and difficult to carry out.

Alexidis et al. analyzed the National Cancer Database for men treated with adjuvant or salvage radiation therapy (SRT) after prostatectomy failure from 2003 to 2012. SRT with doses above 66.6 Gy were labeled "high dose," and SRT with doses above 70.2 Gy were labeled "very high dose." Between 2003 and 2012:

  • High dose SRT utilization increased from 30% to 64%
  • Very high dose SRT utilization increased from 5% to 11%
  • Utilization of high and very high dose rates was greatest at academic centers, lowest at community centers.

The authors decry the fact that this doubling of high dose SRT took place in the absence of RCTs that would definitively establish proof. They point out that the evidence for it is based on observational studies (see, for example, King and Kapp and Ohri et al.), which are fraught with confounding due to stage migration,  selection bias and ascertainment bias. Stage migration was the result of better imaging becoming increasingly available to rule out SRT from patients already harboring occult distant metastases. Also, three randomized clinical trials published in the middle of the observational period convinced many radiation oncologists that earlier SRT led to better tumor control than waiting. Selection bias occurred because the patients selected to get higher doses of radiation were healthier and those whose cancer was less progressed -- they would have done better regardless of the dose. Ascertainment bias resulted from the longer observational period for patients treated in 2003 vs. 2012 - the opportunity for treatment failure increases with the amount of time that has passed. The authors also doubt that biochemical recurrence-free survival (which is what was used in observational studies) is a good enough surrogate endpoint for overall survival. They are right that all these factors may be confounding the previous retrospective analyses, and the only way to know with certainty is to conduct a trial where patients are randomized to receive high or low SRT doses,  and follow patients long enough so that median survival or at least metastasis-free survival is reached in the low dose group.

There has been one randomized clinical trial of SRT dose escalation in the modern era. The SAKK 09/10 trial found little difference in acute toxicity symptoms at 70 Gy compared to 64 Gy, but patient-reported urinary symptoms worsened. Unfortunately, many patients were treated with three-dimensional conformal radiation therapy (3D-CRT), which had higher toxicity than the IMRT in widespread use now. Also, it uses freedom from biochemical failure (not yet reported) as its surrogate endpoint.

So, what is a patient to do in the absence of Level 1a evidence? Should he accept the higher doses with possibly added toxicity and better tumor control, or should he go for a lower dose with possibly less toxicity and less tumor control?

As a compromise, Mantini et al. recently reported 5-year biochemical disease-free survival (bDFS) and other outcomes for patients who received higher dose SRT (70.2 Gy vs. 64.8 Gy) depending on their post-operative pathology. They also may have received (depending on pathology) whole pelvic radiation and adjuvant hormone therapy. Those patients who received the higher dose had equivalent 5-yr bDFS in spite of their worse disease characteristics. Those who received only 64.8 Gy still had a 5-year bDFS as high as 92%. We do not know how many of those recurrent men with favorable disease characteristics actually needed any SRT. They were all treated with 3D-CRT and toxicity was not reported.

The other thing we can do when our information is imperfect is go through the Bradford Hill checklist. It can give us more confidence if we have to make a decision based on less than Level 1 evidence. The factors that ought to be considered are:

  • Strength of Association (larger associations are more likely (but not necessarily) causal)
  • Consistency of Data (independent studies all lead to the same conclusion)
  • Specificity (a very specific population is differentially affected)
  • Temporality (the effect has to occur after the cause)
  • Biological gradient (too some extent, more drug/radiation dose leads to more effect) 
  • Plausibility (one can come up with a plausible explanation)
  • Coherence (lab studies demonstrate a plausible mechanism for the observed effect)
  • Experiment (has the effect been prevented by modifying the cause)
  • Analogy (similar factors may be considered)


Unfortunately, the authors did not refer to Dr. King's more recent analysis of SRT dose/response, which we discussed in depth here. He looked at 71 studies, demonstrating consistency. While it is not Level 1 evidence, it is Level 2a evidence. In it, he observes that the salvage radiation dose response conforms exactly to the primary radiation dose response.  In other words, the prostate tumor is equally radio-resistant whether it is in the prostate or the prostate bed. This increases the plausibility of a dose effect of SRT. What's more, dose escalation was proven to be beneficial for biochemical recurrence-free survival, metastasis-free survival, and freedom from lifelong ADT use, for primary radiation in intermediate risk men by a RCT (RTOG 0126). So, we also have greater confidence in SRT dose escalation by analogy.

RTOG 0126 did not find an increase with higher dose in 8-year overall survival or cancer-specific survival. This calls into question whether these longer-term effects are really useful endpoints if we are to be able to obtain and use the results of any clinical trial in a reasonable time frame.

Dr. King proposed a randomized clinical trial of 76 Gy vs. 66 Gy for SRT. Meanwhile, he is routinely giving his SRT patients at UCLA 72 Gy. Dr. Zelefsky at Memorial Sloan Kettering Cancer Center and other eminent radiation oncologists have also upped the radiation dose to 72 Gy. Such doses seem to be safe and effective, but it is one of many factors in the SRT treatment decision that must be carefully considered by patients and their doctors.


Thursday, July 26, 2018

F18-PSMA-1007 - the latest PSMA-based PET indicator

The development of new PET indicators for prostate cancer continues. As we've seen, the Ga-68-PSMA-11 indicator is already making a difference in clinical practice. Many of the new PET indicators have been developed in Germany, although the best one so far before this, F18-DCFPyL was developed at Johns Hopkins.

Researchers in Germany have developed a new PSMA-based PET indicator, F18-PSMA-1007, that seems to be even better. They tested it on 251 biochemically recurrent (after prostatectomy) patients at 3 academic centers.

  • 81% had a recurrence detected
  • 44% had a local (prostate bed) recurrence
  • 41% had a pelvic lymph node recurrence
  • 20% had a retroperitoneal lymph node recurrence
  • 12% in lymph nodes above the diaphagm
  • 40% had bone metastases
  • 4% had visceral organ metastases


Detection rates varied by PSA:

  • 62% in those with PSAs from 0.2-<0.5
  • 75%  in those with PSAs from 0.5-<1.0
  • 90%  in those with PSAs from 1.0-<2.0
  • 94%  in those with PSAs >2.0


Interestingly, those who had ADT in the last 6 months had higher detection rates (92%) compared to those who'd had no ADT recently (78%). This may be because those who had ADT recently had more advanced tumors. There was some early evidence in mice and lab studies (like this one and this one) that ADT upregulated PSMA. One clinical study indicated that ADT improved detection of PSMA. Two studies  (this one and this one) showed no effect of ADT on PSMA detection. More recent evidence indicates use of ADT negatively impacts detection rates. The patient should avoid ADT before getting a PSMA-based PET scan, if possible.

The detection rate among those with PSAs between 0.2-2.0 was 78%, which is comparable to the 88% detection rate reported for men with PSAs between 0.2-3.5 for F18-DCFPyL and much better than the detection rate of 66% reported for Ga-68-PSMA-11 in that PSA range. F18 has an advantage over Ga-68 in having a longer half-life (118 minutes vs 68 minutes) and is more tightly bound to the ligand. Because it is not appreciably excreted through the urinary tract, it can be seen more easily around the prostate - important when the recurrence is near the site of the anastomosis, as most recurrences are. In a mouse study, it was superior to F18-DCFPyL. In a clinical pilot study, they both detected the same tumors.

As of now, the F18 PSMA-based PET indicators seem to be superior, but others are working on ligands that detect other prostate cancer proteins more sensitively and more specifically. Leading candidates are hK2, FMAU, Citrate, Prostate-Stem-Cell-Antigen, , DHT/androgen receptor, uPAR receptor, VPAC receptor, or multiple ligands.

Also see:




Wednesday, July 25, 2018

The Danger of Complementary and Alternative Medicine

Researchers at Yale did two database analyses. They looked at the National Cancer Database and found 1,901,805 patients who were treated for either nonmetastatic prostate, breast, lung or colorectal cancer between 2004 and 2013. In one analysis they looked at use of complementary medicine; in the other, they looked at use of alternative medicine.

Complementary medicine was defined as use of “other-unproven: cancer treatments administered by nonmedical personnel” in addition to at least one conventional cancer treatment modality, defined as surgery, radiotherapy, chemotherapy, and/or hormone therapy. 258 patients who chose a complementary therapy were matched to 1032 patients who did not use any complementary medicine on age, clinical group stage, Charlson-Deyo comorbidity score (CDCS), insurance type, race/ethnicity, year of diagnosis, and cancer type using the propensity score matching technique.

After 5 years of follow-up, comparing users of complementary medicine to matched non-users:
  • There was no difference in delay of treatment, but there was a greater probability of refusal of surgery (7% vs 0.1%), chemo (34% vs 3%), radiotherapy (53% vs 2%), and hormone therapy (34% vs 3%).
  • 82% survived for 5 years vs 87% among non-users, and were 2.1 times more likely to die after adjustment.
  • The differences in survival were attributable to refusal of conventional treatment.
  • Differences in 5-year survival were significant for breast cancer (85% vs 90%), and colorectal cancer  (82% vs 84%), but not for lung cancer  or prostate cancer. 
Alternative medicine was defined as “other-unproven: cancer treatments administered by nonmedical personnel” and who also did not receive conventional cancer therapy, defined as chemotherapy, radiotherapy, surgery, and/or hormone therapy. 281 patients who used alternative medicine were matched to 560 patients with similar characteristics (cancer type, age, clinical group stage, CDCS, insurance type, race, and year of diagnosis) who did not use alternative therapies using propensity score matching.

After 66 months median follow-up, comparing users of alternative medicine to matched non-users:
  • 55% survived for 5 years vs 78% among non-users, and were 2.5 times more likely to die after adjustment.
  • Differences in 5-year survival were significant for breast cancer (58% vs 87%), lung cancer (20% vs 41%), colorectal cancer  (33% vs 88%), but not prostate cancer  (86% vs 95%)
  • The survival curves for prostate cancer had just begun to diverge at 5 years (75% were low or intermediate risk).
Complementary and alternative medicines consisted of herbs and botanicals, vitamins and minerals, probiotics, Ayurvedic medicine, traditional Chinese medicine, homeopathy and naturopathy, deep breathing, yoga, Tai Chi, Qi Gong, acupuncture, chiropractic or osteopathic manipulation, meditation, massage, prayer, special diets, progressive relaxation, and/or guided imagery.

Although these observational studies did not follow prostate cancer patients long enough to detect differences in survival, we see the damage that use of both complementary and alternative medicines had on patients with more virulent cancers. Patients who get complementary medicine are more likely to refuse conventional treatments (even though they received at least one conventional treatment) and are about twice as likely to die because of that decision.

(update 5/2019) The CAPSURE database shows that the use of complementary and alternative medicines among men with prostate cancer is increasing. Comparing the period of 2006-2010 to 2011-2016. They report that:

  • Use of complementary medicines increase +128% (from 24% to 54%)
  • Vitamin D use has more than doubled in spite of Level 1 evidence that supplementing confers no benefit.
  • Happily, Vitamin E use has decreased based on Level 1 evidence from the SELECT trial.
  • Almost a quarter of men with prostate cancer take omega-3 fatty acids. A secondary analysis of omega-3 use in the SELECT trial, confirming an earlier study, found an association between high omega-3 fatty acid serum levels and increased risk of prostate cancer. Level 1 evidence showed no association with prostate cancer incidence or prostate cancer death.

note: Level 1 evidence means that a cause/effect relationship was proven by a large randomized clinical trial (RCT). Interested readers may consult the Oxford definition, which is widely accepted. Many patients rely on mouse and lab studies which are almost always disproven when tried in clinical trials. They constitute the lowest quality of evidence (Level 5), and should only be used for hypothesis generation for clinical trials or to demonstrate plausibility for a cause/effect relationship found in an RCT. Additionally, the Bradford Hill checklist is used.

Sunday, July 22, 2018

Vitamin D has no effect on prostate cancer, heart disease, or bone mineral density

(updated)

Observational studies have reported conflicting effects of Vitamin D on prostate cancer - some reported no association (as in this study), some reported a positive association (as in this study), and some reported an inverse association (as in this study). We now have three large randomized clinical trials (Level 1a evidence) that show that large doses of Vitamin D have no effect on cancer incidence or deaths. This trumps all previous observational and epidemiological studies.

Manson et al.  reported the results of the VITAL randomized clinical trial (RCT) on 25,871 (including 5,000 African-Americans) men over 50 and women over 55 who were given either:
  1. Vitamin D3 at 2,000 IU per day and marine omega-3 fatty acids (1000 mg/day containing 840 mg EPA and DHA)
  2. Vitamin D3 placebo and fish oil placebo
  3. Fish oil and Vitamin D3 placebo
  4. Vitamin D3 and fish oil placebo
Blood samples were taken and questionnaires were administered to assure compliance. The group assignment was double-blind. Participants agreed to keep vitamin D from all other sources (like sunlight, milk, multi-vitamins, etc.) to under 800 IU per day.

After a year, serum 25-hydroxyVitamin D increased from 30 ng/ml to 42 ng/ml among those supplementing Vitamin D and didn't change in the placebo groups.

After 5.3 years of follow-up, there was:
  • No difference in incidence any kind of cancer (including prostate, breast and colorectal cancers) between Vitamin D3 and Placebo.
  • No difference in deaths from any kind of cancer
  • Low BMI  (<25) may potentiate the effect of Vitamin D on cancer.
  • No difference in any kind of cardiovascular disease, including myocardial infarction, stroke or death from myocardial causes or of interventions like PCI or coronary bypass.
  • There was no synergism with omega-3 fatty acids.
  • There were no statistically significant differences in any subgroup.

Manson et al. also did a separate analysis of omega-3 fatty acids. After 5.3 years of follow-up, there was:
  • No difference in incidence any kind of cancer (including prostate, breast and colorectal cancers) between omega-3 and Placebo.
  • No difference in deaths from any kind of cancer.
  • There was no synergism with Vitamin D.
  • Those with low fish consumption (<1.5 servings per week) may gain some cardiovascular benefit from omega-3 supplementation.
  • No difference in cardiovascular disease overall, stroke or death or of interventions like PCI or coronary bypass.
  • There was significant improvement in the rate of myocardial infarctions and total coronary heart disease among those taking omega-3s.
  • African-Americans, especially those with multiple CV risk factors, taking omega-3s had lower incidence of myocardial infarctions.
  • Myocardial infarctions (MI) were also better for those taking omega-3s among younger people (<67), men, smokers, diabetics, people with hypertension, people taking cholesterol medications, no parental history of MI, 3 or more risk factors, baseline aspirin use, and baseline statin use.
(update 11/18/20) Chandler et al. reported on an updated analysis of the VITAL RCT. They looked at whether Vitamin D supplementation affected the risk of developing metastatic or fatal cancer among people who were cancer-free at baseline. With a median intervention period of 5.3 years, there was almost no chance of finding metastatic or fatal prostate cancer in men who were prostate cancer-free at baseline (In the ProtecT trial, 10-year prostate cancer survival among men initially diagnosed with localized prostate cancer was 99%, and metastasis-free survival was 96%.) Because the metastasis-free and cause-specific survival with prostate cancer are so long when starting from a "no cancer" diagnosis, the authors looked for the effect on other cancers, excluding prostate cancer. They found:
  • there were no significant differences due to Vitamin D on the incidence of any cancer
  • there were no significant differences due to Vitamin D on the metastatic spread across all cancers
  • there were no significant differences due to Vitamin D on all-cancer mortality
  • Adding together metastases and fatalities due to all cancers, the difference (2.1% vs 1.7%) was statistically significant, especially after the first two years
  • The reduction was only statistically significant among those with a normal body-mass index (<25)
  • For prostate cancer patients, there were only 6 such cases among those who got Vitamin D and 14 such cases among those who got the placebo - not significantly different. Presumably, they were missed at diagnosis or had a rare virulent type of PCa.
(Update 3/16/2021) in an ancillary analysis of the VITAL trial data, Albert et al. reported that neither Vitamin D nor Omega-3 supplementation had any effect on the incidence of atrial fibrillation.

(update 2/19/2022) Neale et al. reported the results of the D-Health RCT. This was a randomized double-blind prospective trial among 21,315 Australians aged 60 or over.
  • 10,662 were given Vitamin D; 10,653 were given a placebo
  • Vitamin D3 was given as 60,000 IU/month for 5 years
  • A random sample of both groups was checked for compliance via a blood test for serum 25-hydroxyVitamin D; it was 115 nmol/L in the treatment group vs 77 nmol/L in the placebo group.

After 5 years of follow-up:
  • There was no statistically significant difference in the number of deaths (5% in each group)
  • There was no statistically significant difference in cardiovascular disease mortality
  • There was no statistically significant difference in cancer mortality
  • There was no statistically significant difference in all other causes of mortality
  • Excluding those who died during the first 2 years of follow-up, cancer mortality was 24% higher among those taking Vitamin D

Scragg et al. in a post-hoc analysis of the ViDA trial reported on 5,110 50-84 year-old people seen in community practice in NZ between 2011 and 2012. The study was originally set up to investigate cardiovascular benefit to Vitamin D supplementation (it found none), but they also asked about incidence of cancer and tracked deaths from cancer.
  • 2558 were given Vitamin D; 2552 were given a placebo
  • Vitamin D3 was initially given as one 200,000 IU pill, followed by 100,000 IU monthly pills
  • Serum 25-hydroxyVitamin D was 26.5 ng/ml (seasonally adjusted) at baseline
  • Serum 25-hydroxyVitamin D consistently increased by 20 ng/ml among a sample of treated patients
  • Compliance was excellent
  • There was no difference in the percent who took calcium or Vitamin D supplements or in sun exposure
After a median of 3.3 years of follow-up, there were:
  • 375 new cancer cases; 60 died of new cancers. 
  • 24% had a pre-existing cancer diagnosis; 29 died
  • no significant difference between the Vitamin D cohort and the placebo cohort in the number of new cancers or cancer deaths.
  • 6% had a pre-existing prostate cancer diagnosis; 7 died
  • 64 new cases of prostate cancer (1 died)
  • no significant difference between the Vitamin D cohort and the placebo cohort in the number of new prostate cancers or in prostate cancer deaths.
One can argue that a consistent daily Vitamin D3 intake might have had an effect, or that it takes more than 3 years for an effect (whether beneficial or increased risk) to be observed. There is, at present, only observational studies for either assertion. Sample size prevents consideration of the hypothesis that Vitamin D may prevent early growth of prostate cancer but may accelerate metastases (as in this mouse study).

Jiang et al. report the results of a Mendelian randomization study of the causal connection between serum Vitamin D levels and prostate cancer. They identified 6 genetic mutations associated with low serum levels and looked for them in 79,148 men who were diagnosed with prostate cancer. They found no greater incidence of those genetic mutations in men with prostate cancer or advanced prostate cancer. Nor was there any association in women with breast cancer. The genetic mutations were also not statistically different in 73,699 people who did not have breast or prostate cancer. This proves there is no causal connection between low Vitamin D and prostate cancer.

No Effect or Negative Effect on Bone Mineral Density

Some men on hormone therapy take Vitamin D and calcium for the purpose of maintaining bone mineral density (BMD). 

(update 7/27/22) An update of the VITAL randomized clinical trial reported no difference in fractures among people supplementing Vitamin D vs. people not supplementing it. There were no differences by age, sex, race, BMI, baseline use of calcium or Vitamin D supplements, or serum Vitamin D levels.

Reid et al. reported that Vitamin D supplementation had no effect on bone mineral density. They further noted that lower doses had more effect than higher doses, probably because Vitamin D has been found to pull calcium out of bones at high doses. However, Datta and Schwartz reported that at 200-500 IU/day Vitamin D and 400 mg-1,000 mg calcium supplementation had no effect on men's bone mineral density. Calcium supplementation has been associated with increased risk of prostate cancer (see this link or this link). Trajanoska et al. found that mutations in the genes responsible for regulating serum Vitamin D levels had no effect on fracture risk, nor did the genes regulating the tolerance for dairy (which is correlated with calcium intake). They also question the routine use of Vitamin D and calcium supplements in men who are taking Xgeva or a bisphosphonate like Zometa to preserve bone mineral density. (Estrogen patches may also prevent loss of bone mineral density.) They wrote:

Studies seeking to show whether these supplements do increase the efficacy of osteoporotic treatment or decrease adverse events (that is, hypocalcaemia) are lacking. In either case, screening for vitamin D deficiency and seeking its correction should be warranted before the initiation of anti-resorptive treatment [e.g., Zometa or Xgeva].  Moreover, in a recent mendelian randomisation study investigating the role of vitamin D in maintaining bone mineral density, increased levels of vitamin D had no effect on bone mineral density measured by [DEXA scan]. However, increased 25-hydroxy-vitamin D was associated with a slight reduction in heel bone mineral density estimated by ultrasonography. These results are consistent with our mendelian randomisation findings of no causal effect of vitamin D levels on fracture.
(Update 9/2/2019) Burt et al. published the results of a randomized clinical trial (RCT) of three different doses of Vitamin D on the bone mineral density of 311 men with 3 years of follow up. The experiment was set up as follows:

  • 1/3 received 400 IU/day; 1/3 received 4000 IU/day; 1/3 received 10,000 IU/day
  • none had osteoporosis at baseline
  • all had baseline serum Vitamin D between 30-125 nmol/L (12-50 ng/ml)
  • those whose total intake of calcium (dietary+supplements) was < 1200 mg/day received calcium supplements
  • baseline serum calcium was 8.4-10.2 mg/dl
  • 53% were men
  • average age was 62 (range:55-70)

After 3 years on their vitamin D regimen:
  • Serum Vitamin D was 77 nmol/L, 132 nmol/L, and 144 nmol/L in those taking 400 IU/day, 4000 IU/day, and 10,000 IU/day, respectively
  • BMD in the radius bone of the arm decreased in all groups in a dose-dependent manner:
    • -1.2% in the 400 IU/day group
    • -2.4% in the 4,000 IU/day group
    • -3.5% in the 10,000 IU/day group
  • Hypercalcemia (too much calcium in the blood) and hypercalciuria (too much calcium in the urine) increased with increased Vitamin D dose
  • Kidney/liver dysfunction, falls, fractures and cancer did not vary with dose.

In a Medpage interview, the authors point out that this is plausible because of two effects of Vitamin D:
  • It increased bone resorption (more than bone formation), as measured by an increase in CTx.
  • It increased parathyroid hormone, either directly or by increasing calcium absorption from the gut

The study did not include people taking bisphosphonates, Zometa or Xgeva.

Also, the dose-dependent effect (higher doses were more damaging than lower doses) increases the plausibility of this being a real effect.

 It’s worth noting that Vitamin D, unlike other vitamins, is a steroid. Steroids tend to interact and to have wide-ranging effects in humans. Overwhelming our steroid-control systems with massive doses of any one steroid is bound to have unintended consequences.


Possible increase in testosterone 

It should be remembered that Vitamin D is a steroidal hormone (like testosterone, estrogen, progesterone, and cortisol) and there are receptors for it on virtually all cells, healthy and cancerous. It has far-ranging effects. It also is part of the human biochemical factory that inter-converts many different kinds of steroids. In fact, Anic et al. showed there was a positive association between serum Vitamin D level and the amount of serum testosterone - not an effect that a man who is taking androgen deprivation wants.

Given that Vitamin D has no effect on incidence of cancer or cancer mortality, that it has no cardiovascular benefit, and no effect on bone mineral density, there is no reason to take supplemental Vitamin D unless serum levels are too low (below 20 ng/ml). 

Sunday, July 8, 2018

The Best Therapy for Gleason 10s

We recently saw (see this link) that men diagnosed with Gleason score (GS) of 9 or 10 had lower rates of metastases and better prostate-cancer survival if they were treated with a combination of external beam radiation (EBRT) plus a brachytherapy boost to the prostate ("brachy boost therapy" - BBT) than if they were initially treated with EBRT, or if they were initially treated with surgery (RP). The same researchers looked at a subset of patients who were initially diagnosed as GS 10.

There were only 112 patients who were biopsy-determined as GS 10. Of those,

  • 26 were initially treated with RP (median age 61)
  • 48 were initially treated with EBRT (median age 68)
  • 38 were initially treated with BBT (median age 67)

The median follow-up was relatively short:

  • 3.9 years for RP
  • 4.8 years for EBRT
  • 5.7 years for BBT


  • Upfront androgen deprivation was given to 98% of EBRT patients vs. 79% of BBT patients
  • Post RP radiation was given to 34%
  • Pre-RP systemic therapy was given to 35%


By 5 years of follow-up:

  • Only 3% of the BBT cohort received systemic salvage therapy vs. 23% of the RP group and 21% of the EBRT group
  • Distant-metastasis-free survival (adjusted) was 64% for RP,  62% for EBRT, and 87% for BBT
  • Prostate cancer-specific survival (adjusted) was 87% for RP. 75% for EBRT, and 94% for BBT
  • Overall survival was not significantly different in the 5-year time frame

While GS 10 is often more aggressive, it is noteworthy that 87% of those receiving BBT had no distant metastases detected within 5 years. Among men who received RP,  57% were upstaged to T3/4 and 41% were downgraded to GS 7-10 by post-prostatectomy pathology. We have no reason to believe those percentages would differ markedly among those who received radiation.

Although the numbers here are small, this is the largest analysis of Gleason 10s broken down by the therapy that they received that we have ever seen. Only a randomized clinical trial can provide a definitive answer. Given the aggressive course of GS 10, patients with this diagnosis are advised to talk to a radiation oncologist who specializes in this therapy.


Monday, April 30, 2018

First randomized clinical trial of SBRT

In the first trial ever to randomly assign patients to extreme hypofractionation, primary radiation therapy delivered in just 7 treatments had the same effectiveness and safety as 39 treatments.

The results of the HYPO-RT-PC randomized clinical trial were published in The Lancet. There was an earlier report on toxicity. Details of the trial specs are available here. Between 2005 and 2015, they enrolled 1200 intermediate- and high-risk patients at 12 centers in Sweden and Denmark to receive either:
  1. Conventional fractionation: 78 Gy in 39 fractions
  2. SBRT (stereotactic body radiation therapy): 42.7 Gy in 7 fractions
The biologically effective dose is 19% higher for SBRT in terms of cancer control. The biologically effective doses are equivalent in terms of toxicity.

The patients were all intermediate (89%) to high risk (11%), defined as:
  1. Stage T1c-T3a
  2. PSA> 10 ng/ml 
  3. Gleason score ≥7
80% of the men were treated with a technology called three-dimensional conformal radiation therapy (3D-CRT), which is seldom used for prostate cancer external beam therapy anymore at major tertiary care centers. It is never used for SBRT in the US because it is considered not precise enough, and too toxic. SBRT is usually delivered in 4 or 5 fractions in the US. CyberKnife and VMAT are the most common technologies in use, and use of sophisticated image guidance throughout each treatment is a common practice.

With follow-up of 1,180 patients for 5 years, they report biochemical recurrence-free survival of 84% in both arms of the study.

They also reported updated late-toxicity results. By 5 years after treatment:
  • Grade 2+ urinary toxicity was 5% for conventional fractionation, 5% for SBRT - no significant difference.
  • Grade 2+ rectal toxicity was 4% for conventional fractionation, 1% for SBRT - no significant difference.
Up until now, we've only had reports from clinical trials using SBRT (like this one) or conventional fractionation (like this one), and it could have been reasonably argued that SBRT results looked good because of selection bias. With this study, we now have Level 1 evidence of non-inferiority. This will not be surprising to those of us who have followed the randomized clinical trials of moderately hypofractionation vs. conventional fractionation (see this link). This will be hailed as a victory for patients who no longer have to endure and pay the high cost of 8 weeks of treatments. radiation oncologists, who are reimbursed by the number of treatments they deliver, probably will not be as thrilled.

Tuesday, March 27, 2018

Should perineural invasion influence active surveillance and radiation treatment options?

Perineural invasion (PNI) is a risk factor detected on a biopsy in 15%-38% of men with a prostate cancer diagnosis. It means that the pathologist saw nerves infiltrated with cancer cells. As they grow, tumors cause nerves to innervate them. The cancer infiltrates in and around small nerves that connect to nerve bundles (ganglia) outside the prostate, becoming a route of metastatic spread (see this link).  The data on whether it is independently prognostic for T3 stage after surgery are equivocal, although PNI is often the mechanism for extracapsular extension.  After considering Gleason score, PSA, stage, and tumor volume, PNI does not seem to add much to the risk of recurrence after surgery. PNI is not associated with higher surgical margin rates, and it is not considered sufficient to preclude nerve-sparing surgery. An open question is whether it raises risk enough to warrant more aggressive radiation options, like brachy-boost therapy, whole-pelvic radiation and long-term adjuvant ADT.

Peng et al. retrospectively examined the records of 888 men who were treated with external beam radiation at Johns Hopkins from 1993 to 2007. 21% of them had biopsy-detected PNI. Compared to men with no PNI, those with PNI had:

  • lower 10-year biochemical failure-free survival (40% vs 58%)
  • lower 10-year metastasis-free survival (80% vs 89%)
  • lower 10-year prostate cancer-specific survival (91% vs 96%)
  • similar 10-year overall survival (68% vs 78%)

It isn't surprising that PNI is associated with higher risk, but does it add any new information not already captured by Gleason score, stage, and PSA (i.e., the NCCN criteria for risk stratification)? After correcting for those other risk factors, PNI was still found to be associated with lower rates of biochemical failure-free survival, but not of metastasis-free survival, prostate cancer specific survival or overall survival.

PNI independently predicted for lower biochemical failure-free survival in low-risk and high-risk patients, but not for intermediate-risk patients.  Although it is a relatively rare finding among low-risk patients, when found, PNI also predicted for lower prostate cancer-specific survival. Biochemical failure in low-risk men with PNI differed according to whether they received adjuvant ADT or not:

  • 33% in men not treated with ADT
  • 8% in men treated with ADT

An earlier analysis of 651 men treated at the University of Michigan similarly found an association between PNI and biochemical failure-free survival, freedom from metastases, prostate cancer-specific survival, but not overall survival at 7 years after radiation treatment. They also found a more marked effect among high-risk patients. A meta-analysis of 5 studies among men who received EBRT found that PNI increased the risk of biochemical recurrence by 70%.

Although PNI may increase the risk associated with an unfavorable intermediate-risk or high-risk diagnosis markedly, brachy boost therapy is the best treatment for any such patient regardless of PNI, according to our best retrospective study and prospective studies like ASCENDE-RT. This study suggests that adding ADT may be beneficial for these patients. Low and intermediate-risk patients with PNI who opt for conventional IMRT may also benefit from the addition of short-term ADT.

(update 4/2020) In a ten-year follow-up of the TROG 03.04 RADAR  randomized trial, Delahunt et al. found that PNI detected at biopsy was independently associated (after adjusting for other risk factors) with later appearance of bone metastases.

Biopsy-detected PNI may have implications for active surveillance. Cohn et al. detected PNI in only 8.5% of 165 men selected for active surveillance. Within 6 months, they were given a confirmatory biopsy. AS was excluded at the confirmatory biopsy due to higher Gleason grade in 57% of men with PNI vs. 13% of men without PNI. PNI should not automatically exclude active surveillance, but it should be recognized as a risk factor in the decision. It would be interesting to know if there is an association between PNI and genomic risk (based on Oncotype Dx, Prolaris, or Decipher tests). It has yet to be determined whether PNI is still a significant risk factor after NCCN risk category, % core involvement, and genomic risk have been accounted for.

It is worth noting that PNI is not always reported on biopsy cores by pathologists, and there is no uniform method for quantifying it. Whether nerve infiltration is small or large, or outside or inside the nerve sheath, it is just reported as PNI, if it is reported at all. It will be difficult to include PNI as part of any risk stratification system until its reporting has been standardized.

Note: Thanks to Daniel Song for allowing me to see the full text of the study.

Monday, March 19, 2018

Escalated radiation dose doesn't improve 8-year overall survival in intermediate risk men (but does it matter?)

Last week, we saw that escalated dose did not improve 10-year overall survival in high-risk men (see this link). The latest published findings of the randomized clinical trial (RTOG 0126) prove that 8-year overall survival was not improved in intermediate risk men who received a higher radiation dose. In both studies, we are left wondering whether that matters.

Michalski et al. reported the results of RTOG 0126, a randomized clinical trial (RCT) designed to prove that escalated dose improves survival in intermediate risk men. It was a very large trial:
  • 1499 men
  • 104 sites in the US and Canada
  • Patients treated from 2002 to 2008
  • Median age was 71
The patients were all intermediate risk, defined as:
  • Stage T1b-T2b, and
  • Gleason score ≤ 6 and PSA ≥10 and <20 (16%), or
  • Gleason score = 7 and PSA < 15 (84%)
  • 71% were Gleason score 3+4
The treatment consisted of:
  • either low dose 70.2 Gy/ 39 treatments 
  • or high dose 79.2 Gy/ 44 treatments
  • delivered using 3D-CRT (66%) or IMRT (34%)
  • none had adjuvant ADT, but they may have had salvage ADT or other salvage therapies if RT failed
After a median follow-up of 8.4 years:
  • 8-year overall survival was 75% for the low-dose group vs. 76% for the high-dose group (not significantly different)
  • 8-year prostate cancer mortality was 4% for the low-dose group vs. 2% for the high-dose group (not significantly different)
  • 8-year biochemical failure was 35% for the low-dose group vs. 20% for the high-dose group (significantly different)
  • 8-year local progression (felt with DRE) was 6% for the low-dose group vs. 3% for the high-dose group (significantly different)
  • 8-year distant metastases (bone scan/CT detected) was 6% for the low-dose group vs. 4% for the high-dose group (significantly different)
  • 8-year salvage therapy was 22% for the low-dose group vs. 14% for the high-dose group (significantly different)
Toxicity outcomes were as follows;
  • Acute grade 2+ urinary toxicity was 17% for the low-dose group vs. 17% for the high-dose group (not significantly different)
  • Late-term grade 2+ urinary toxicity was 7% for the low-dose group vs. 12% for the high-dose group (significantly different)
  • Acute grade 2+ rectal toxicity was 5% for the low-dose group vs. 7% for the high-dose group (not significantly different)
  • Late-term grade 2+ rectal toxicity was 15% for the low-dose group vs. 21% for the high-dose group (significantly different)
In a separate analysis of the high-dose group:
  • Acute grade 2+ urinary and rectal toxicity was 15% among those treated with 3D-CRT vs. 10% among those treated with IMRT (a significant difference)
  • Late-term grade 2+ urinary toxicity was not significantly different among those treated with 3D-CRT vs. IMRT
  • Late-term grade 2+ rectal toxicity was 22% among those treated with 3D-CRT vs. 15% among those treated with IMRT (a significant difference)
This RCT raises many important questions about the design of clinical trials and the validity of conclusions drawn from them. Dr. Michalski addressed some of these concerns in an audio interview presented with the published study. This was an enormous undertaking, running almost two decades from design to reporting, and coordinating the treatments and reporting of 1,500 men in over 100 sites spread throughout Canada and the US.

The results show that dose escalation was not needed to increase 8-year survival in these intermediate risk patients. But this probably won't change practice for a number of reasons.

The intervening endpoints are of considerable importance to patients: the anxiety associated with rising PSA, the toxicity of all the salvage therapies, and the pain and possible crippling due to metastases all impact quality of life.

The median age of the men at treatment was 71, and they were screened for good performance status. The actuarial life expectancy in the US for a 71 year-old men is 14 years. This implies that they ought not make a decision based on expected survival for only 8 years. Also, as radiation-treated men get treated at a younger age, the gap will become more pronounced. According to the Memorial Sloan Kettering Life Expectancy Nomogram, a 71 year-old intermediate-risk man in good health has only a 8% probability of succumbing to prostate cancer in 10 years (vs 3% in 8 years in this study), and 12% at 15 years if he had no treatment whatever. At the same time, his probability of dying from other causes is 30% in 10 years, and 51% in 15 years. The overall survival improvement may not become apparent until median survival is reached in 15 years. And differences in prostate cancer survival are difficult to discern when numbers are this low. But it is difficult and costly to track patients for 15-20 years. We have to look to surrogate endpoints.

While 8-year overall survival and prostate cancer-specific survival did not improve, all the intervening endpoints did. Biochemical failure, local progression, distant metastases, and use of salvage therapies were all worse in the low-dose group.  It is very costly and difficult to run an RCT long enough to see a survival difference in men with localized prostate cancer. As we've seen, the few RCTs that have run the longest for each type of therapy have been single institution studies with much smaller sample sizes. Distant metastasis-free survival is probably a better surrogate endpoint if the study can't run for 15-20 years. There were enough metastatic events to see a difference. A recent analysis by the ICECaP Working Group of 12,712 patients in 19 clinical trials of radiation in localized prostate cancer showed that 5-year metastasis-free survival was almost perfectly correlated with overall survival. By reducing the time needed to accumulate data, this might increase the relevance of such trials while reducing their costs.

As Dr. Michalski points out, survival in both groups was much better than expected when the study was designed in 2001. This is largely because life-extending salvage therapies (e.g., docetaxel, GnRH agonists, Zytiga, Xtandi, Xofigo, and Provenge) have become prevalent in the interim.

Toxicity was markedly reduced by the introduction of IGRT/IMRT technology that became increasingly available, especially in the US, in the last 20 years. With the improvement in beam accuracy and the knowledge of the dose/toxicity relation of organs at risk, tighter dose constraints for organs at risk have been utilized. Because of the technology changes, a high-dose regimen today is probably no more toxic than a low-dose regimen. So, if there is little toxicity cost to the more effective treatment, why not use it? Rapidly adopted changes in radiation technology in the last 20 years, especially the shift from 3D-CRT to IMRT, render many of the findings irrelevant to today's standard practice.

Another RCT reported by Nabid et al. at the 2015 Genitourinary Conference had similar findings. They found that 10 year overall survival was no different for higher dose (76 Gy vs 70 Gy) or the addition of short-term ADT. Biochemical failures were actually worse in the higher dose group, but only if short-term ADT was not used with it. Zaorsky et al. conducted a meta-analysis of dose escalation trials in intermediate risk men and arrived at a similar conclusion. A contrary finding was noted by Kalbasi et al. in their analysis of the National Cancer Database. They found that there was a significant survival increase associated with higher dose (hazard ratio = 0.84). In fact, for every 2 Gy increase in dose, there was an 8% reduction in the hazard of death in intermediate-risk patients. Being retrospective, their analysis suffers from selection bias - it may be that the frailest patients got lower doses. However, they did include more unfavorable intermediate risk patients, including those treated with adjuvant ADT.

We are now recognizing that unfavorable intermediate risk patients may benefit from adjuvant ADT and higher doses, whereas the favorable intermediate risk patients may not. EORTC 2291 and the Nabid et al. trial established that short term (6 month) ADT markedly improved progression-free survival. Several retrospective studies (like this one and this and this) suggest that the benefit is limited to those with less favorable disease characteristics. It may well be that higher doses are necessary to overcome the radioresistance of high volumes of Gleason pattern 4.

The degree to which RTOG 0126 is irrelevant to contemporary decision-making is heightened by the success of hypofractionated IMRT and SBRT in intermediate risk patients. Both provide much higher biologically effective doses, equal efficacy to conventional IMRT, and about the same toxicity. Also, their cost is lower and patient convenience is higher. Unless a patient has an anatomical abnormality such that dose constraints cannot be met, it is hard to come up with a reason why higher biologically effective doses should not be used.

Note: Thanks to Dr. Howard Sandler for allowing me to see the full text of the study.


Thursday, March 15, 2018

Bounces after Primary Radiation Therapy

Perhaps the single most annoying "side effect" of radiation is not a side effect at all; it's the periodic fluctuations in PSA, called "bounces," that can occur for years after therapy. A new analysis from Memorial Sloan Kettering assures us that our anxiety is misplaced -- PSA bounces predict better cancer control.

Romesser et al. reported on a retrospective study among 776 patients treated from 1990 to 2010 with dose-escalated external beam radiation therapy. The median radiation dose was 81Gy. None received adjuvant ADT. They defined a bounce as a PSA rise ≥ 0.2 ng/ml but less than 2.0 ng/ml above the lowest level (nadir) it had reached thus far, followed by a decrease to as low or lower than the previous nadir. After a median follow-up of 9.2 years, they found:

  • 16% of patients had a bounce
  • The bounce occurred after a median of 24.6 months
  • It was a median of 0.37 ng/ml over the previous nadir

Bounces were more likely to occur in patients who:

  • were younger (see this link)
  • had a lower Gleason score
  • were lower  T stage
  • received a higher radiation dose

At 8-years follow-up, they reported that bounces were associated with:

  • Greater time to reach ultimate PSA nadir (43 months vs 26 months) 
  • Lower risk of PSA relapse (9% vs 29%)
  • Decreased risk of metastases (1% vs 10%)
  • Decreased prostate-specific mortality (0% vs 3%)
  • Decreased overall mortality (6% vs 11%)

Very similar findings have been reported for other forms of radiation: SBRT, Low Dose Rate Brachytherapy (seeds), High Dose Rate Brachytherapy, and Brachy Boost Therapy. A 2004 study of EBRT and bounces found an inverse correlation between bounces and PSA relapse-free survival. The difference is probably attributable to much lower radiation dose (only 1% received > 78 Gy) and because the higher risk men were treated between 1986 to 1995, mostly before PSA testing became prevalent.

The percent of men who experience a bounce ranges from 15%-50%, and depends on how the researchers define a bounce. It ranges from a minimum of  0.1- 0.8 ng/ml above previous nadir in most studies. Bounces are often above +1 ng/ml, may last for more than a year, and are usually noted between 1 year and 4 years after therapy.

The reason that bounces occur is a bit of a mystery. There are various theories:

  • Prostatitis - either pre-existing, arising after invasive procedures (e.g., biopsy, fiducial placement, or brachytherapy), or induced by radiation.
  • Immune infiltration: after radiation releases cancer antigens, T cells are activated to eventually attack the remaining cancer in the prostate (see this link).
  • Cancer cells that have been dormant, eventually emerge and undergo "mitotic catastrophe."
  • Delayed apoptosis (cell death) among late-responding healthy cells
  • PSA drops most sharply and consistently in more aggressive cancers because radiation kills the most rapidly dividing cells first.
  • PSA measurement variation (e.g., different test kits, different labs, natural fluctuations)

Whatever the reason, bounces are a good thing. For patients that were diagnosed with low or intermediate risk prostate cancer, a slow, bouncy PSA decline should engender a feeling of relief rather than anxiety. But what of the unfavorable risk patient with bounces so high that they approach or exceed +2.0 ng/ml and stay elevated? While recurrences usually occur later than bounces, is there a method available for early detection of a local recurrence? Biopsies are invasive and non-informative while the cancer is still in the "slow death" phase. However, there is a kind of MRI called MR Spectroscopy (MRS) that may be able to non-invasively distinguish between bounces and PSA recurrence. In a pilot study (and this one), the researchers found that the MRS-detected choline/citrate ratio might be markedly elevated and focal if the cancer is metabolically active, but low and diffuse if there is only benign inflammatory activity.

Thursday, March 8, 2018

Brachy boost therapy and surgery extend survival about the same in high risk patients, but brachy boost does more

Two retrospective studies were published in the last week, and they had some similar findings, but some dissimilar things to say about which treatment is best for high risk prostate cancer. The three therapies they looked at were the combination of brachytherapy and external beam radiation (brachy boost therapy - BBT), external beam therapy alone (EBRT), and surgery (RP).

Kishan et al. reported on 1,809 men with Gleason score of 9 or 10 who were treated between 2000 and 2013 at 12 tertiary cancer care institutions (UCLA, Los Angeles VA, California Endocurie Therapy Center, Fox Chase, Mt. Sinai, Cleveland Clinic, Wheeling Jesuit University, University of Michigan, Johns Hopkins, Oslo University, William Beaumont Hospital, and Dana-Farber).

Patient characteristics:
  • 639 were treated with radical prostatectomy (RP).
  • 734 were treated with EBRT only.
  • 436 were treated with BBT (BT was either low dose rate in 62% or high dose rate in 38%).
  • All patients were Gleason 9 or 10 on biopsy.
  • Pelvic LN involvement was discovered in 17% of RP patients ; 40% had positive surgical margins.
  • RP patients were younger (61 years of age) compared to EBRT or BBT patients (68 years of age)
  • RP patients were lower stage ( 87% clinical stage T1/T2) compared to EBRT (70% clinical stage T1/T2 ) or BBT patients (79% clinical stage T1/T2)
  • RP patients had lower pre-therapy PSA (7 ng/ml) compared to EBRT or BBT patients (10 ng/ml)
  • RP patients had lower percentage of Gleason score 10 (4%) compared to EBRT (6%) or BBT patients (9%)
Treatment specs
  • Among the RP patients, 43% had adjuvant or salvage radiation therapy (68 Gy).
  • Among radiation patients, about 90% had adjuvant ADT
  • Median dose of EBRT was 74 Gy.
    • adjuvant ADT continued for 22 months, median.
  • Median equivalent dose of EBRT+BT was 92 Gy
    • adjuvant ADT continued for 12 months.
Oncological outcomes

After a median follow-up of 4.2, 5.1 and 6.3 years for RP, EBRT, and BBT, respectively, the oncological outcomes (adjusted for age and disease characteristics) were as follows:
  • The 10-year rates of distant metastases were
    • 46% for RP 
    • 44% for EBRT
    • 13% for BBT
    • Differences between BBT and the two others were statistically significant.

  • The 10-year rates of prostate cancer-specific mortality (PCSM) were
    • 23% for RP
    • 26% for EBRT
    • 13% for EBRT + BT
    • Differences between BBT and the two others were statistically significant.

  • The 10-year rates of all-cause mortality (ACM) were
    • 32% for RP
    • 39% for EBRT
    • 31% for BBT
    • None of the differences were statistically significant.
    • There was a difference at 7.5 years in favor of BBT that vanished by 10 years.
In additional analyses, the authors looked at outcomes by duration of androgen deprivation for those receiving any kind of radiation. They found that ADT duration made no significant difference in detected metastases or PCSM within EBRT or BBT, and did not account for the difference between them. They also looked at radiation doses. EBRT patients who received <70 Gy had PCSM significantly worse than those who received ≥ 78 Gy. The rates of metastases did not differ. Notably, very few (11%) of the EBRT patients had both ≥ 78 Gy and ≥2 years of ADT, a combination that is now considered standard of care. Those that did had superior outcomes compared to RP. The use of LDR-BT or HDR-BT as part of BBT made no difference.

The authors conclude:
Among patients with Gleason score 9-10 prostate cancer, treatment with EBRT+BT with androgen deprivation therapy was associated with significantly better prostate cancer–specific mortality and longer time to distant metastasis compared with EBRT with androgen deprivation therapy or with RP.

In an analysis of the National Cancer Database, Ennis et al. reported on the overall survival of patients who were treated with RP, EBRT, and BBT for high-risk PC from 2004 to 2013. The database covers about 70% of all new prostate cancer patients treated in the US. The patient profile was:

  • 24,688 patients treated with RP, at least at first
  • 15,435 patients treated with EBRT
  • 2,642 patients treated with BBT.
  • All EBRT patients also had adjuvant ADT
  • BBT patients may or may not have had ADT
  • All were high risk by the NCCN definition: Either Gleason score 8-10, stage T3/4, or PSA≥20 ng/ml
  • RP patients were younger (62 years of age) compared to EBRT (70 years of age) or BBT patients (67 years of age)
  • RP patients were lower stage ( 89% clinical stage T1/T2) compared to EBRT (84% clinical stage T1/T2 ) or BBT patients (85% clinical stage T1/T2)
  • RP patients had lower pre-therapy mean PSA (19 ng/ml) compared to EBRT (23 ng/ml) but the same as BBT patients (19 ng/ml)
  • RP patients had lower percentage of Gleason score 8-10 (70%) compared to EBRT (78%) or BBT patients (73%)
  • Comorbidities were similar among groups.
  • The above risk factors as well as socioeconomic factors and year of diagnosis were used to adjust the raw data.
  • It is unknown what percent of RP patients had adjuvant or salvage radiation.
  • There was no data available on post-reatment metastases or prostate cancer-specific survival
Because surgery is sometimes aborted when pelvic LN cancer is discovered, they estimated the probability that patients had positive nodes, and included it as a risk factor. This would seem to double count those risk factors, but the authors say it had little effect. Based on their model, they estimated that the percent who had positive nodes was 10% of RP patients, 34% of EBRT patients, and 23% of BBT patients.

After a median follow-up of 36 months, the relative oncological outcomes (adjusted for age and other patient and disease characteristics), expressed as hazard ratios were as follows:

  • RP: 1.0
  • EBRT: 1.53 (i.e., 53% worse survival vs. RP)
    • EBRT with < 79.2 Gy: 1.68
    • EBRT with ≥79.2 Gy: 1.33
  • BBT: 1.17 (not significantly different from RP)
    • not different if ADT included
    • no interaction between comorbidities and treatment effects
The authors conclude:
This analysis showed no statistical difference in survival between patients treated with RP versus EBRT plus brachytherapy with or without AD. EBRT plus AD was associated with lower survival. 
In an accompanying editorial, Ronald Chen discusses the problem of drawing conclusions about comparative effectiveness from this kind of registry data in the absence of clinical trial data. He points out that patient selection criteria are not completely reflected in comorbidity data. He believes that those who are selected for EBRT are just less healthy than those who can undergo anesthesia for surgery or brachytherapy. Other unmeasured confounders include burden of disease, and patient and physician preferences.

The two studies had similar conclusions, but tell us different things. They both found no effect of treatment on overall survival. Lest one walk away thinking it then doesn't matter, the experience of living with painful, crippling metastases and the experience of dying from prostate cancer are horrific in themselves. In the Kishan study among top institutions, there is greater confidence than in many studies that deaths due to prostate cancer could be distinguished from death from other causes. Still, overall survival is impaired in patients with cancer, even if the cancer itself isn't the ultimate cause of death.

Although several randomized clinical trials (RCTs) have demonstrated significant improvements in progression-free survival from BBT compared to EBRT, none have yet demonstrated improvements in overall survival. We saw this recently in the 2005 Sathya RCT. But the prostate cancer-specific mortality advantage of BBT has been confirmed in another study. In a recent analysis of the SEER database, PCSM was 40% higher among patients who had EBRT compared to those who had BBT.

Other than the lack of metastasis data and PCSM in the NCDB, there were other important differences between the two studies. In the Ennis study, only 25%-35% were gleason 9 or 10, whereas all were in the Kishan study. Other differences included the lack of comorbidity data in the Kishan study, and the lack of adjuvant/salvage radiation data in the Ennis study.

Prostate cancer-specific mortality rates were cut in half by BBT, and metastases were only a fraction compared to the other treatments. While this does not prove causality (only a randomized clinical trial can do that), it is highly suggestive that escalated dose can provide lasting cures. There may be good reasons why some high risk patients may have to forgo brachy boost therapy in favor of high dose EBRT or RP with adjuvant EBRT, but for most, brachy boost therapy will probably be the best choice. Patients who are treated with EBRT only, should receive a radiation dose of at least 79.2 Gy and two years of adjuvant ADT.

Sadly, a recent analysis of the National Cancer Database showed that utilization of brachy boost therapy for high risk patients has declined precipitously from 28% in 2004 to 11% in 2013. If a patient sees anyone other than the first urologist, he often only sees a single radiation oncologist who only informs him about IMRT. In most parts of the US, there is a dearth of experienced brachytherapists.

- with thanks to Amar Kishan for allowing me to see the full text.