High Risk Prostate Cancer: A Complex Landscape, but with Strong Ongoing Trials
Today, a look at the FLAME and PRIME trials as we push for SBRT to clearly demonstrate excellent results high risk patients..
www.protons101.com
Home to the musings of a radiation oncologist - with a slant on protons and dose and optimizing cancer outcomes.
In last weeks discussion, we started the dive into higher risk disease and covered a lot of issues - touching upon dose, field size, ADT, and PSA kinetics and how one needs to consider these in a different light in high risk disease.
Today, we’ll look at two trials (FLAME and PRIME) that are well designed and have a great opportunity to help shape current standards. In the end, I think each of these trials will continue to push SBRT utilization for prostate cancer further into the high risk spectrum of disease. We’ll walk through why I think that is case and why you should watch for these pivotal trials.
Goal on this Substack is to point out both strengths and weaknesses I see. Nothing is perfect and I think it is important to point out places where I believe it could be better. These trials take tremendous work, effort, and funding and are critical to our field. Kudos to the authors and all involved for moving our field forward!
FLAME TRIAL SERIES
FLAME 2.0 acute toxicity data is out - an intermediate and high risk cohort treated with SBRT. It once again shows promising early results shrinking the number of fractions and this time - days of treatment. Below is a graphic of the broader progression of the FLAME trials - they are doing outstanding work out of Belgium and the Netherlands. Kudos to all involved!
As shown, FLAME looked at SIB approach to treat the lesion within the prostate to a really high dose - BED>100. This is still one of my favorite approaches. They next pushed for even shorter courses in hypo-FLAME reporting limited toxicity in a Phase II single arm 100 patient series. And now hypo-FLAME 2.0 acute toxicity has been released, where not only have they pushed the number of fractions down to five, but they are delivering the treatment twice a week instead of just once a week to complete the treatment faster (on the calendar).
I really like the original FLAME randomized trial. Great design demonstrating clear improvement in curing the cancer. Honestly though, FLAME SIB hasn’t caught on despite clear data of improved outcomes. It simply landed at the wrong time in January of 2021 - within one year of what likely is the largest / fastest shift ever recorded in the practice patterns of radiation oncology.
Below is graph showing a MASSIVE shift in UK breast cancer practice patterns in early 2020. FLAME was just 9 months after this cataclysmic shift.
After that, it simply wasn’t “cool” to treat for that many sessions. Today uniform SBRT whole gland treatment, likely to lower dose, is far more common in the US than a longer, higher SIB dose approach - despite this proven benefit. Yes, some are using SIB in both traditional and SBRT settings, but I think this continues to be the minority.
But as the illustration shows, within the FLAME trial structure, they have continued to iterate.
FLAME 2.0 Acute toxicity paper
FLAME 2.0 Acute toxicity paper was published in May. A one arm trial moving from 1 treatment a week to now 2 fractions per week - shortening the course from 29 days under treatment to 15. Dose remained at 35 Gy / 5 fractions to the whole gland while allowing an “iso-toxic” boost (their term for prioritization of normal tissue constraints over tumor dose increases) up to 50 Gy. Both hypo-FLAME and hypo-FLAME 2.0 are each single arm trials with 100 and 124 patients, respectively. Sequential trials not randomized.
Together they show a few things - either way, there was no grade 3 toxicity. Grade 2 toxicity jumped from 47.5% from 7.4% and those treated once a week reported improved QoL for bowel bowel and urinary toxicity even with more experience using the technique - I don’t think that fact should be overlooked.
In simple terms, it shows a real increase in mild/moderate symptoms that is clearly measurable on both CTCAE scales and on QoL patient forms. That said, in this study, the acute toxicity didn’t translate to Grade 3 or 4 toxicity.
Remember though, both PACE data and RTOG 0232 demonstrate that higher acute toxicity translates into higher rates of late toxicity. Below the PACE analysis:
I’ve filed this under, if it shows up early, it resurfaces late - pretty easy to remember. And oh yea and it is based on multiple prospective randomized datasets so file in a prominent location.
In summary, I think one can look at the trial in either light - positive for the lack of clear higher grade toxicity or negative for the clear and significant changes in grade 2 acute toxicity. Kudos to the authors who essentially state that in the conclusion:
Conclusion:
Semi-weekly prostate SBRT with iso-toxic focal boosting is associated with acceptable acute GU and GI toxicity. Based on the comparison between the QW and BIW schedule, patients should be counselled regarding the short-term advantages of a more protracted schedule.
I think the final issue to touch upon is that many/most? high risk trials today include treatment of the pelvic lymph nodes. Primarily based on the POP RT Whole pelvis data. As we discussed in the later part of last weeks discussion, that can still be controversial but I think it is the way that many are leaning.
These two trials included a pretty strong high risk cohort - about 67% high risk. Including 15% Grade group 4 and 5% Grade group 5 disease. About 10% have PSAs between 20 and their upper cutoff of 30. Pretty consistent with the cohort of high risk disease I see.
So while this is good data, I think there is stronger data on the horizon for high risk - really high risk disease.
PRIME Trial:
The PRIME Trial is out of Tata Memorial in India. I think it has a great structure and likely will answer a bunch of questions. It was recently presented at ESTRO 2023.
The patient mix there (India) vs. here (US) is quite different, but I think we’ll be able to broadly apply many of the answers from this study to our patients with pretty good approximation of results. Just for context, I agree with Dr. Murthy who described it as “the first report of high risk, very high risk, and node positive prostate cancer treated with SBRT”. Widmark (amongst a few others) includes some high risk (11%), but this trial clearly addresses a more advanced cohort.
Couple of comments from the peanut gallery (me).
I don’t love non-inferiority trials as I’ve discussed previously - here (Non-inferiority Trial Design). But on the plus side, we have a publication of the statistical assumptions published in Feb 2020:
This is a non-inferiority trial and the non-inferiority margin between the standard and the test arms is 9% (delta). To detect this delta difference in the primary end point of a 4-year BFFS (80% in standard arm and 71% in the test arm; HR 1.53), with a power of 80% and a one-sided 5% alpha value, total of 135 events are required, with a minimum number of 422 patients.
It’s a bit interesting to just compare that calc to POP-RT trial. The whole pelvis arm achieved a 5 yr BFFS rate of 95%. Patient subsets are different, but both are in high risk and very high risk, although the PRIME trial obviously looks to include lymph node positive cases. 80% appears quite low and 71%? Well that is the reason I don’t like non-inferiority designs - as a patient I’d pick 95%.
But here, the design is used to produce quite a de-escalation in number of treatments. It isn’t a slow push from say 25 to 15, but we move from low 20-25 to 5. And while I generally don’t love these types of designs, I don’t think it unreasonable in this particular instance. Not my favorite, and with hindsight, I’d pick a higher base and a tighter delta, but that does grow numbers so admittedly a tough balance. If the standard arm produces 90% or above, I think that “fixes” some of the low initial assumptions - we'll see.
The second comment is that we also use a primary metric of BFFS rather than disease free survival. I don’t quite understand that approach. I understand they have clear rules for progression to PSMA scanning but personally, I’d still like to see the primary outcome be an all-inclusive approach to counting all failures - even if the absolute difference is small.
To me, these large critical trials carry massive importance (say 10,000,000x this post) and help define precedence and create standards in reporting - not just for themselves, but across our industry and to some extent, across other approaches in other fields of medicine looking to address prostate cancer - think hifu, cryo etc. I believe it serves us better to set consistently high bars that others struggle to meet - I have that much faith in our approach using radiation as the primary modality for this disease.
What do I like for an outcome measure? Look to PACE:
8.4 PACE-B Primary Endpoint 8.4.1 Freedom from biochemical (Phoenix definition) or clinical (commencement of androgen deprivation therapy, local recurrence, nodal recurrence or distant metastases) failure. The primary timepoint of interest is 5 years from randomisation.
Simple, straightforward and includes all failures.
Moving along, a second slide:
Prostate SBRT is here to stay
To me, this speaks to two points - 1) SBRT is coming and will grow in utilization based on these current trials. 2) In general, there really is much more limited data in true high risk disease.
A quick refresher on the trials in the slide above:
NRG-GU 005: Intermediate risk - big margins - 8mm and 5mm in the “standard arm” vs. 5mm/3mm in the SBRT arm - 70/28 or 60/20 vs 36.25/5. A margin / dose trial with a SUPER SHORT - 2 yr biochemical disease free survival primary outcome metric - essentially designed to guarantee equivalence at the early 2 yr outcome mark.
HYPO RT-PC - Published: Widmark - 84% DFS - 11% High risk - mainly Intermediate risk - 42.7 in 7 fractions - 3 days per week vs. 78 Gy - so unique on its “ultrahypofractionated” approach.
PACE B 78 Gy / 39 or 62 Gy / 20 vs. 36.25 / 5 - Low Risk and Intermediate Risk patients. Powered for 85% goal for standard and enrolling to have equivalence at 79% or higher. (Note: I’m on record that my estimate for the SBRT arm is ~88%).
PACE C Intermediate Risk and (more limited) High Risk disease (no GG5 for example). 60 Gy / 20 fractions vs. 36.25 / 5 fractions with 6 months ADT powered again for 85% - non-inferiority a little tighter at 5% so 80% and up. It includes a blend of IR and lower HR risk without pelvic treatment but with ADT. Will be interesting to see how many would have qualified for pelvic treatment via POP-RT Roach formula approach (estimated risk 20% or greater for pelvic nodal treatment).
PRIME TRIAL DESIGN:
Above is the trial structure - both simple and yet quite elegant in evaluating several important questions within a single trial framework. I’ll explain what I see in the approach.
First, let’s look at the “standard” arm. Note it integrates two important and necessary concepts to be relevant today and into the future: modest hypofractionation and SIB which we know from the FLAME-SIB trial is a superior approach.
Further, it looks at 25 vs 20 fractions with good modern dosing. Quick EQD2 calcs put pelvic doses in a 46 - 50 Gy range and prostate doses in an 80 EQD2 range. I guess one (like me), might argue there is some potential to push dose even further but in high risk, I think this is a strong standard arm by any metric.
Compare this design to the current US COMPARRE trial comparing IMRT to protons - 78 Gy in 39 fractions and 60 Gy in 20 fractions. We know 60 / 20 is a lower dose - PACE-B escalated it (as ADT is not allowed) and it doesn’t include SIB and offers a very traditional 39 fraction approach. From my standpoint and in contrast to the PRIME Trial design, I see little of interest and little that has the ability to move the needle so to speak - wish I had a different view.
But back to PRIME - then they include a second elegant arm using SBRT. The dose is once again appropriate - perhaps lower than some today, but with real potential to demonstrate that with ADT, there is no need to go higher. It uses a lower EQD2 dose to the pelvis - but the one nearly EVERYONE uses - 25 / 5. Yes it is a little lower than traditional pelvic doses if you convert towards traditional fractionation, but they did great in not dropping the standard dose arm to 45 / 25. Instead, IF the trial shows equivalence, you can’t really question the lower end dosing of the SBRT arms because they kept the standard arm quite high.
And then, just as with the standard arm, they threw in an extra wrinkle - asking a question that many have today - every other day vs. weekly for control and toxicity in the SBRT arm.
Even from the peanut galley, I really can only throw applause at the design. A great trial asking great questions. It has modern staging. It has EOD or weekly SBRT options. Frankly it will address so many questions for high risk and very high risk disease.
My Prime Guess:
If I looked back at data and guessed, I would guess they are equivalent in outcomes for cancer control for at least 5 years. Its likely to be THE trial that moves large segments to SBRT within the high risk space. If you strongly believe in pelvic treatment (which I’m less convinced for my patient cohort) I think one can argue the dose differences favor the standard arm. But ultimately I don’t think the dose differences are enough to matter and show up in the results. And on the initial data release, even the toxicity appears to be comparable between the two arms which I think is the larger concern for a difference in favor of the standard approach. Finally, I’ll be interested to see the breakdown of EOD vs. weekly in the SBRT arm.
As the slide says, Prostate SBRT is here to stay. (kind of like protons :))
In Closing:
Today, I think that is a good stopping point. There is a bit more to cover in high risk disease but today, we’ll break. I think the PRIME data (amongst other series) will likely build strong support for SBRT in even high risk disease. Am I convinced it is better for cancer outcomes? Far less clear. But I don’t really define “better” as shorter. At the end of the day, these trials significantly build out our SBRT data within high risk and for that, they should be applauded and followed closely.
In the upcoming week, we’ll look at randomized trials that are ongoing in Europe comparing protons to photons and then discuss why I believe that randomized controlled trials are absolutely critical - it is a walk through history that is quite interesting.
To support the work, please share the site or subscribe. Thank you for reading and keep pushing for better!
IMPORTANT TRIAL REFERENCES:
From once-weekly to semi-weekly whole prostate gland stereotactic radiotherapy with focal boosting: Primary endpoint analysis of the multicenter phase II hypo-FLAME 2.0 trial
https://pubmed.ncbi.nlm.nih.gov/37178932/
Change in the Use of Fractionation in Radiotherapy Used for Early Breast Cancer at the Start of the COVID-19 Pandemic: A Population-Based Cohort Study of Older Women in England and Wales
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9151525/
Prostate-Only Versus Whole-Pelvic Radiation Therapy in High-Risk and Very High-Risk Prostate Cancer (POP-RT): Outcomes From Phase III Randomized Controlled Trial
https://pubmed.ncbi.nlm.nih.gov/33497252
Study protocol of a randomised controlled trial of prostate radiotherapy in high-risk and node-positive disease comparing moderate and extreme hypofractionation (PRIME TRIAL)
https://pubmed.ncbi.nlm.nih.gov/32114475/
The Association between Acute and Late Genitourinary and Gastrointestinal Toxicities: An Analysis of the PACE B Study
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9954684/
The PACE Trial (Design Document for entire PACE structure)
https://www.icr.ac.uk/media/docs/default-source/default-document-library/pace_protocol_v12_clean.pdf?sfvrsn=130f3069_0