This contradicts Nabid's finding that 18 months are as good as 36 months. The reasons for the discrepancy are discussed.
There is a trade-off: BBT can come with severe late-term urinary side effects (among 19% in the ASCENDE-RT trial), while the late-term urinary side effects are milder for EBRT-only (only 2.5% in the DART trial). Only the patient can decide if he is willing to take on 12 months of ADT with BBT vs over twice as long for EBRT-only, given the higher expected radiation toxicity with BBT.
Written by
Tall_Allen
To view profiles and participate in discussions please or .
Interesting, would like to see the data analyzed for sub-groups by Gleason and stage. My gut feeling is "nadir + 12 months ADT" is sufficient, after having done HDR-BT/IMRT plus 18 months ADT for Gl-7/t3bN1-- the last 12 months PSA was u/d.
I read that article with interest a few days ago. I read the original paper.. and questioned how valuable it was since it seemed at times to question its own results and presented one result which left me scratching my head as to how that might happen.
I don't believe the multivariate analysis included health differences between who gets EBRT vs Brachi or EBRT and Brachi, which can certainly have an effect on overall survival and even on the effectiveness of any cancer treatment. Since I believe it was a retrospective paper, that data may not have been available in the original databases they were working with.
The original paper (which this link doesn't take us to, it's Allen's summary of the paper) has a number of curves, one of which I found really puzzling (my background in science sez if it's this odd - the result has to be questioned and repeated by another researcher) that's the curve that shows a strange dip to about zero at 26.3 months (from memory) followed by a rise to above the level where the dip occurred... I believe it was ADT time vs chemical recurrence and the dip was the distance metastasis I believe. It appeared as if the 26.3 months was a magic thing. To me it looked like a data error somehow.
I've looked for the paper (I thought I saved it) but can't find it. I might look some more tomorrow.
Allan - might you have a link to the original paper. I'd like to refresh my memory on it.. Thanks!
The plots I have problems with are the "Natural Cubic Spline" plots. First I had to try to understand the Natural Cubic Spline plots, what they are: towardsdatascience.com/nume...
I'll admit to coming away from that discussion still puzzled.. at this point "Figure 1" and plot [B] I have no idea what it's actually trying to tell us.
More understandable (at least to me) is Figure 2. That appears to present a clear advantage to a longer ADT duration for men undergoing EBRT and that 12 months is probably optimal for men doing EBRT/Brachy.
In reading further into the study, I found more details that I had questions about. Each of the studies they took data from had different maximum EBRT radiation doses for their study candidates. None were what is now considered somewhat standard practice (in excess of 80Gy). It isn't detailed what sort of EBRT (IGRT/IMRT/ARC/3D) was given to the study participants, which presents a lot of variables that could change the results they get.
I suspect a lot of the things I have problems with are the result of this being a retrospective study, not an A vs B sort of study.
I did think about the Nabid study showing 18 months being non-inferior to 36 months of ADT for men doing EBRT. In thinking about why men may have stopped the treatment before their assigned 36-month prescription - could it be by the time they stopped, prior results from men who had received treatment earlier (and completed their entire 36 month treatment) may have indicated that 36 months wasn't necessary? And if 18 months wasn't inferior to 36 months, continuing the ADT treatment for these men was only going to be damaging to them, with no survival advantage? That might be the answer, but only the people doing the study or the participants can answer that.
Interesting stuff, especially for anyone doing radiation/ADT.
Sorry- I should have read this before I replied to your last post.
It is not exactly a retrospective study and neither is it exactly a prospective study - it is a hybrid. In fact, a very novel way of using data. The fact that it was published in JAMA should tell you what high regard this is being given. Their MARCAP initiative draws on data from trials all over the world.
Radiation dose was given at standard levels in all those studies. That's why the authors chose the studies they chose. No one gives over 80 Gy by IMRT - it has been proven to be too toxic. In fact, the various kinds of linacs do not improve oncological results (the subject of the study), they only improve toxicity.
The Nabid study was a superiority study. And no, the reason for dropping out could not have been due to the revelation of results of the lower doses. Clinical trials are not allowed to reveal that.
DART had toxicity data for EBRT- the other 2 did not. I drew toxicity data from ASCENDE-RT for BBT, as I wrote.
Only one of the 3 studies was retrospective. It was a multi-institutional study among 16 of the top hospitals and had nearly full oncological and treatment data on patients.
I think you are talking about the cubic spline curve fitting. You may want to read about the technique before forming an opinion. It is not magic - it is a solid, technique for interpolating data- better, in fact, than normal multiple polynomial fitting. Even if that is above your pay grade, you can perhaps better understand the other way they arrived at the same result. Both techniques arrived at the same conclusion.
Allen, thanks for the reply, and as I noted, I did find the link in your article, I apparently had overlooked it on first read at 2AM in the morning.
You're absolutely correct that cubic spline curve fitting is above my pay grade. The other plots were more understandable.
I did note though that it appears the data plotted for the 28 months of ADT was based on a much smaller population of patients than either of the other (6 and 18 month) plots. To me, that means other factors not considered take a greater weight when there are fewer samples. Please let me know if I understand these plots correctly, it appears some of the 28-month plot data points are based on 1 patient. Did I miss something there?
Just on a side note - I found varying definitions of a "retrospective" study after googling the topic. Some of the definitions claimed (perhaps not correctly) that any study based on data previously gathered or on events that happened previously were "retrospective", and any studies based on future data to be gathered were "prospective".
And a really off-topic but somewhat humorous incident - once at Bell Labs a renowned researcher had submitted a paper for peer-review where the primary conclusion of the paper was based on a chart with a single data point. Needless to say - the plot fit the data perfectly but wasn't judged acceptable for publication. That always makes me look closely at any plot with a curve fitted to a single data point... the researcher mentioned was rumored at one point to be considered for the Nobel Prize in Chemistry, but has never been awarded it, supposedly due to that paper and that plot.
They used data from all the trials for the curve fitting for which there was data. The paper shows the sample in each arm - it was several hundred in each arm, as I recall. DART cut off at 28 months, but the 16-member consortium study could have any duration. RADAR limited patients to 18 months, which would only add to the sample as far as that point.
It is a retrospective analysis, certainly, but both DART and RADAR were prospective trials. I would evaluate the level of evidence as 2a.
I think this really doesn't answer the question that clearly. The conclusions from the study were:
"In conclusion, the findings of this cohort study suggest that the optimal duration of ADT for patients receiving high-dose EBRT may be more than 18 months—implied by the findings in all 3 cohorts. A secondary conclusion, based on the retrospective data set, is that durations of less than 18 months may be sufficient for patients receiving EBRT+BT. "
(bold/underline - mine.) The "may be" to me is significant. I suspect a real answer to the question raised would require more information such as the type of radiation treatment used, when the ADT was started (before treatment, after radiation was complete) and the health of the subjects of the study. Using the term "may be" to me means they simply don't have a good answer to that question given the unknown variables in their study.
Interesting - though I believe for high risk these days the triad of ADT, EBRT, and low or high dose Brachy is the preferred course of treatment - as outline in the ASCENT study. This particular study is only using a two-pronged approach.
The idea of this study is whether they can duplicate the benefit of brachy boost therapy with lower toxicity risk by using EBRT only with longer adjuvant ADT.
Allen: Wasn't the ASCENDE-RT trial low dose rate brachytherapy boost? If I am not mistaken, the high dose rate boost has a much lower toxicity profile.
Kishan believes that prostate boosts, whether with brachy, SBRT, or protons, should get 12 months of ADT. But so far, it has only been shown for brachy.
I am basically in the middle of doing treatment based upon the Ascend RT trial,; 25 sessions of protons including whole pelvic, LDR brachy boost, and ADT (Lupron) for at least one year and the info is very helpful.
Content on HealthUnlocked does not replace the relationship between you and doctors or other healthcare professionals nor the advice you receive from them.
Never delay seeking advice or dialling emergency services because of something that you have read on HealthUnlocked.
Suggestions on Tolerating ADT?
At last I've come to a decision.
\"late toxicity\" issues when dealing with radiation therapies.
