Christopher Wallis: Hello, and thank you for joining us for this UroToday Journal Club discussion. Today, we're talking about a recent publication entitled, High-dose Radiotherapy or Antigen Deprivation Therapy (HEAT) as Treatment Intensification for Localized Prostate Cancer: An Individual Patient-data Network Meta-analysis from the MARCAP cap Consortium. I'm Chris Wallis, an assistant professor in the division of urology at the University of Toronto. With me today is Zach Klaassen, assistant professor in the Division of Urology at the Medical College of Georgia. You can see here, the citation for this recent publication led by Dr. Kisham and colleagues, and published in European Urology.

As most will know, radiotherapy is a standard of care treatment approach for patients with localized prostate cancer, and treatment intensification has proven beneficial for many patients. This may be accomplished by the use of radiotherapy dose escalation or the use of short or long-term androgen deprivation therapy. These treatment intensification approaches have demonstrated consistent improvements in biochemical control, although benefits in terms of metastasis-free survival have been somewhat variable between studies. It's notable that only metastasis-free survival has been shown to be a valid surrogate for overall survival, and thus, improvements in biochemical control of questionable clinical value.

There are now, based on randomized data, six radiation-based treatment approaches, including low-dose radiotherapy alone, low-dose radiotherapy with short-term androgen deprivation, low-dose radiation with long-term androgen deprivation, or high-dose radiation therapy, again, either alone or with short or long-term androgen deprivation. Due to the nature of parallel randomized trial designs, nearly all studies look at only two of these approaches, and thus, we have a whole variety of treatment paradigms that have never been compared.

The MARCAP Consortium also known as the Meta-analysis of Randomized Trials in Cancer of the Prostate, seeks to address issues related to the variety of treatment approaches by using data repository approach to provide aggregation of international trials. HEAT, or high-dose radiotherapy or androgen deprivation therapy, provides an individual patient-level data comparison of these six treatment approaches.

The authors performed a prespecified systematic review, and they registered their systematic review approach as well as meta-analytic approach with PROSPERO prior to beginning. They evaluated randomized control trials assessing radiotherapy dose escalation or the use of prolongation of engine deprivation in the treatment of patients with localized prostate cancer. They search MEDLINE, Embase, Cochrane Register of Controlled Trials, clinicaltrials.gov, Web of Science, Scopus, as well as major relevant abstracts of conferences and societies. They searched from 1966 to 2020, and excluded studies which did not provide evidence of a metastasis survival data, used nonsteroidal anti-androgens without androgen deprivation therapy, single center in nature, or used life-long durations of androgen deprivation.

As described above, there are six treatment regimes. For the purpose of this analysis, low-dose radiation therapy was defined as less than 74 gray and high-dose defined as greater than equal to 74 gray or the use of a brachy therapy boost to external beam radiation. Short-term ADT was defined as 3 to 6 months of use, whereas long-term was defined as 18 to 36 months. Androgen deprivation could be administered with the use of an LHRH agonist, with or without a first-generation nonsteroidal antiandrogen. The primary outcome was metastasis-free survival, and in secondary analyses the authors considered other related prostate cancer outcomes.

The authors used a pre-specified statistical analysis plan. They assess between trial heterogeneity and publication bias in standard manner. For metastasis-free survival, overall survival, and biochemical recurrence-free survival, the authors use patient-level data to calculate trial specific hazard ratios to use these in pair comparisons. A Frequentist network meta-analysis was used for both direct and indirect comparisons using random effects models, and the authors ranked their six treatment approaches using the P score. Cox proportional-hazard models and Kaplan-Meier derived estimates of the survival function were used to estimate 8-year absolute risk differences and restricted means survival time differences.

In addition, given variable sample sizes and event rates across each comparison, the author's performed a pre-specified Bayesian network meta-analysis. They used a two-step design, the first step, which included standard regression models to obtain aggregate data, and then a second stage using Bayesian modeling to synthesize the results of these aggregate data and compare treatment approaches. Again, as with their Frequentist models, random effects models were used and a Markov Chain Monte Carlo simulation was used. The SUCRA was calculated based on the ranking of treatment strategies to allow the assessment of most likely beneficial treatment approaches.

At this point in time, I'm going to hand it over to Zach to walk us through the results of this HEAT analysis from the MARCAP consortium.

Zachary Klaassen: Thanks so much, Chris, for that great introduction. This is the search strategy used for this analysis. As you can see here, with regards to identification, there was 1,488 records identified through database, and 26 additional records from other sources. After duplicates are removed, 1,148 records were screened, excluding 1,889. Subsequently, 59 full-text articles were assessed for eligibility, and subsequently, 13 studies were included in the meta-analysis.

This table looks at the 13 trials included in this analysis. As you can see, this is a mix of RTOG, EORTC trials, as well as several others, with a very wide range in trial population. Recruitment, generally, was from the late 80s through early 2000s, with several different arms and treatment regimens included in this meta-analysis. As you can see here, these are all relatively large studies, with the largest being 1,974 patients in the RTOG 9408 trial, as well as 1,532 in the RTOG 0126 trial.

These are the patient characteristics, as you can see here, stratified by the six treatment regimens, as Chris described. This includes low-dose RT alone, as well as with short-term and long-term ADT, as well as high-dose radiotherapy alone, in combination as well as with short-term and long-term ADT. Looking here at the column of all patients, we'll quickly summarize this. The follow-up for these trials was quite long, at just under 9 years. The age of treatment was 70. Majority of these patients were high at intermediate NCCN risk group patients. In terms of the Gleason score, most commonly was Gleason score 7 at 42%. Initial PSA at the time of study inclusion was 11.4. The most common T stage was T1/T2 at 68% of the patients included in this study.

The next several slides we'll look at the forest plot derived from the Frequentist NMA of treatment strategy. This first one looks at metastasis free survival, and the setup of this forest plot is that all of these combination treatments are compared to low-dose radiotherapy. I've highlighted the significant treatment options that improve metastasis-free survival compared to low-dose radiotherapy. And this includes high dose radiotherapy with long-term ADT, as well as low-dose radiotherapy with both short and-long term ADT, significantly improving MFS compared to low-dose radiotherapy. A similar forest plot, but looking at overall survival this time, again, compared to low-dose radiotherapy. We see significant improvement with the combination of high and low-dose radiotherapy with long-term ADT, as well as patients with low-dose radiotherapy and short-term ADT.

Finally, biochemical recurrence-free survival. Again, compared to low-dose radiotherapy, we see that all of these treatment options, including low and high-dose radiotherapy combined with long-term and short-term ADT, as well as high-dose radiotherapy alone, improve biochemical-free survival.

The next several panels, we'll look at selected pairwise comparison from the NMA evaluating the impact of escalating radiation dose, adding ADT or prolonging ADT. This first panel here looks at metastasis-free survival, and on the left, you can see several different combinations and comparisons between these options. What I've highlighted at the bottom here is the metastasis-free survival looking at several of these combinations of interest. So we see no difference in MFS for high-dose radiotherapy plus long-term ADT versus high-dose radiotherapy alone. Sorry, we do see a difference between those two. We do not see a difference with regard to the bottom three combinations. So this includes high-dose radiotherapy versus low-dose radiotherapy, high-dose radiotherapy versus low-dose radiotherapy among all patients receiving short-term ADT, as well as high versus low-dose radiotherapy among all patients receiving long-term ADT.

When we look at biochemical-free survival, we see that these combinations, generally, when they're compared against each other, all have significant improvement in biochemical-free survival, except for, at the bottom, not included in the box here, high-dose radiotherapy versus low-dose radiotherapy among all patients receiving long-term ADT.

Finally, overall survival looks very comparable in terms of the comparisons with regard to metastasis-free survival. Again, we see a difference between high-dose radiotherapy significantly improving survival with the addition of long-term ADT, but we see no difference in overall survival for high-dose versus low-dose radiotherapy, as well as high-dose versus low-dose radiotherapy among patients receiving short-term ADT, and no difference for high-dose versus low-dose radiotherapy for those receiving long-term ADT.

Next several studies will look at the predicted treatment rankings with similar figures. Here we can see that the six treatment options are color coded on the right, and this is for metastasis-free survival. The interpretation of this figure is that there's a 50% probability that high-dose radiotherapy plus long-term ADT would be the most likely to be used, and there's a 49% probability that low-dose radiotherapy plus long-term ADT would be the second most likely combination to be used with regards to metastasis-free survival. Similarly, a similar figure for overall survival, 81.2% likelihood that high-dose radiotherapy plus long-term ADT would be the most likely to be used for overall survival, followed by 17.2% for low-dose radiotherapy plus long-term ADT.

With regards to biochemical recurrence-free survival, again, we see that there's a 68.1% probability that high-dose radiotherapy plus long-term ADT would be the most likely to be used for biochemical recurrence-free survival.

Several discussion points from this network meta-analysis. This is the first NMA using IPD across treatment strategies in localized prostate cancer. Although ideally every treatment strategy would be randomized against each other, this would require 64 randomized trials to assess every treatment strategy in this network meta-analysis. This study showed that high-dose radiotherapy plus long-term ADT appears to be the optimal strategy for all endpoints, including biochemical recurrence-free survival, metastasis-free survival, and overall survival. These results strengthen previously underpowered observations regarding the importance of ADT, even in the context of dose-escalated radiotherapy. It's also demonstrated that further radiotherapy dose intensification, even in the context of long-term, ADT, has a high probability of improving biochemical recurrence-free survival.

The key limitation of this study is that network that meta-analyses rely on homogeneity between trials of similar interventions and transitivity between trials of different interventions. And given the time spent of these trials from 1987 to 2010, there are likely inherent limitations to the methodology.

In conclusion, conventionally escalated radiotherapy up to 79.2 gray alone or in presence of ADT does not improve MFS, while the addition of short-term ADT or long-term ADT to radiotherapy alone, regardless of radiation dose, consistently improves MFS. Secondly, radiotherapy dose escalation provides a high probability of improving biochemical recurrence-free survival, and provided it can be delivered without compromising quality of life, this may represent the optimal treatment strategy when used in conjunction with ADT. Thank you very much for your attention, and we hope you enjoyed this UroToday Journal Club discussion.