All right. Good afternoon, and welcome to the Novalis Circle Symposium. We're gonna be talking about clinical and technical innovations in spine radiosurgery treatment. And I'm very honored to be part of the program. And I've been asked to get this going. I think all of you have the agenda. We'll follow the agenda as much as possible and I think we really have some great topics here and some, I think, really interesting speakers. So it's a real honor to be part of this. My name is Josh Yamada. I'm a radiation oncologist at Memorial Sloan Kettering Cancer Center and I've been asked to talk about the value of stereotactic radiosurgery over conventional fractionation for the treatment of bone metastases.
Here are my disclosures. So, I think we're all familiar with the benefit that conventional fractionated radiation provides in terms of palliating pain for bone metastasis. This is probably one of the most common indications that every clinic in the world routinely uses radiation therapy for very effective palliation of metastases in bone, including the spine. And a number of years ago, Dr. Peter Gerszten at the University of Pittsburgh and I did a literature review of the impact of conventional fractionated radiation and palliation specifically of spine metastasis since that's my particular interest and today we're really talking about the spine. We found from our literature review that there are two randomized control trials and they show about a 50% median survival after treatment. There are three prospective randomized trials showing about a 70% to 80% improvement in pain, and when you look over the literature, that's essentially what you get. And one thing that comes through with conventional fractionated radiation is that the tumor phenotype or the histology will have an impact on your outcomes with conventional fractionation.
So what we did is we looked through the literature. We looked over 100 published peer review reports, some of it prospective, most of it retrospective. And investigators, without fail, will notice that the histology has an impact on the likelihood of treatment success in palliation, where the majority of histologies, except for breast cancer and prostate cancer, the rest of that histologic types that we commonly see as metastatic spine are unfavorable responders to conventional radiation.
And you would expect things like renal cell carcinoma, or melanoma and sarcoma of course, and many GI primaries to be less responsive. But interestingly enough, across the board, I'm just gonna point out that non-small cell lung cancer in these published reports were all considered unfavorable responders to conventional fractionated palliative radiation in the spine. And some people, I think, find that a little surprising that non-small cell lung cancer is that way. But in our paradigm, we certainly feel that non-small cell lung cancer is not a favorable responder to conventional fractionated radiation therapy.
When you look at the favorable responders, I'm just gonna quote two reports. One is the Maranzano study. These patients did not undergo surgery. They had kind of a split course conventional fractionated schedule which was a prospective study or a large Japanese series using conventional fractionation showing that favorable responders have a very high likelihood of response and that it's relatively durable. On the other hand, when you look at these large studies and prospective studies, unfavorable responders have a very short median duration of response, and a two-year local control rate in the Japanese series which is over 600 cases only had a 30% probability of local control at 2 years.
This is another relatively old data set. So, just to show you that this idea that histology is an important factor in response to conventional fractionation for metastases in the bone is not a new idea. We've known this for a long time and we know in clinical practice that certain histologies are not gonna respond as well as others when we use conventional fractionation.
As you may be aware, there are a number of randomized trials that were done looking at the optimal dose fractionation schedule for bone metastases. And although we're talking... These are not specific to the spine. The majority of these patients had spine metastases or vertebral body metastases. So this one is the UK bone Mets Trial. Over 760 patients that tested a number of fractionation schedules, 800 centigrade times 1, 400 centigrade times 5, or 300 centigrade times 10, and what they found is that there was actually really no difference in terms of palliative benefit, whether you used a single 800 times 1 or 300 times 10. However, the time point that they reported was less than three months.
Now, back in the day when overall survival was not high, a three-month endpoint was probably appropriate, and most of the randomized trials, and there are a number of them, the R2OG did a couple of them. Most of these randomized trials that were seeking the optimal fractionation schedule came...and there's ASTRO, as you know, an ASTRO white paper that suggests that the optimal fractionation schedule for bone metastasis is 800 times 1. But that's based on data presuming an endpoint of three months. So, if your patient is only gonna live three months, then perhaps 8 Gy times 1 is an appropriate fractionation schedule. But if you're gonna live longer than that, this data shows that these patients had a doubling of their failure of palliation at 12 months compared to 3 months. So, if your patient is lucky enough to live 12 months, but you gave them 8 times 1, then they have more than double the chance of failing treatment simply because they were good enough to live that long.
So, that makes you really wonder, is this kind of data appropriate for us as we manage patients in the modern era? So here's our experience with stereotactic radiosurgery for spine metastasis. This is our institutional experience from 2003 to 2015, over 800 lesions treated with stereotactic radiosurgery. The median dose was 24 Gy in a single fraction. We define local failures as any radiographic progression and we had 28 of these failures. The mean time to failure was 26 months. And what we found, interestingly enough, is that, when you give a high enough dose, the tumor size, for example, has no bearing on the probability of durable tumor control, and very different than conventional fractionation, tumor histology, the phenotype of the tumor, has no impact on the likelihood of local failure. The only thing that matters with stereotactic radiosurgery is that you give a high enough dose. If you can do this, then you have a very high likelihood of long-term durable tumor control.
So this is a intent...this is a time to failure analysis looking at the dose given to the PTV. And so, this red line here indicates that the likelihood of local failure at 48 months is 2%. If you give 24 Gy in a single fraction, this is equivalent to our 24 Gy prescription point. If you give 24 Gy in a single fraction, the likelihood of durable tumor control of local failure is only 2%. And so, this suggests that at this dose point, you're really giving on a very ablative course of radiation because the outcomes are quite durable. As opposed to that, giving less than 24 Gy, the median does here was 18 Gy in a single fraction, with longer follow-up, we continue to see local failures as opposed to the 24 Gy cohort where, with longer follow-up, we are no longer seeing tumors fail. And you'll see that we have follow-up in some of our patients that now exceeds 10 years.
This is a report that we presented at Astro by my colleague, Dr. Zaleski, where we did a randomized trial comparing 900 Gy times 3 versus 24 Gy times 1 for predominantly spine metastases, but these were just oligometastatic patients. About two-thirds of these are vertebral body metastases. And what we found again is that, with 24 Gy in a single fraction, we had, in this randomized trial, a much higher probability of local failure if we hypofractionated. And so, in this case, what we found is the single fraction, a 6% cumulative incidence of local failure in this prospective study as opposed to fractionation, 9 Gy times 3, there's a significantly higher rate of local failure at 23% cumulative incidents of local failure with hypofractionated treatment.
This is the Toronto experience that was published last year in the Red Journal. This is 12 Gy times 2. They have it 2 years and 18% cumulative incidence of local failure as opposed to our 24 Gy cohort. Of course, these are not the same patients. I'm just showing you this line just for hypothesis generation only, but it suggests that the single fraction paradigm has a significantly higher likelihood in all probability of local...have reduced local failure at less than 2% compared to 17.5% local failure.
This is also a paper from our institution looking at conventional fractionation for very radioresistant histology of sarcoma. This is conventional fractionation probability of local control compared to hypofractionated. This is 9 Gy times 3 or 6 Gy times 5 compared to single fraction radiosurgery. Again, significant differences in the probability of local control in favor of single session, high dose radiosurgery.
In our randomized trial, one of the most interesting things, I thought was interesting, because I really wasn't surprised that 24 Gy times 1 gives better local control than 9 Gy times 3. I don't think that's a surprise to anybody. But what we did find as a surprise is that the likelihood of developing new metastases was significantly lower in the 24 Gy times 1 group. I was actually kind of surprised to see that. But this is the data suggesting that, at three years, the cumulative incidence of distant metastases for the single fraction group was less than 5% compared to 18% for the hypofractionated patients.
Again, this is a prospective randomized trial and not something that we were expecting to see, but nonetheless, there may be a connection between more aggressive local therapy and the development of new metastases. All these patients that were enrolled in this trial were oligometastatic. They had five or less metastases at the time of randomization. So that these patients all had very good prognosis, and the kind of patient you would want to select for SBRT.
Now, we're all very familiar with this trial. I think this is a just fascinating small randomized trial led by the group in Canada with one to five oligometastatic metastases. These patients were randomized either conventional treatment or to SBRT, getting 16 to 24 Gy in a single fraction for a spine metastasis. And what's interesting about this is that in many instances, the patients who got more aggressive local therapy actually had better outcomes and there was actually a doubling in overall survival and benefit of more aggressive SBRT as compared to conventional radiation. Again, suggesting that local outcomes may have an impact upon the systemic status of patients who have oligometastatic relatively favorable prognosis disease. There was a few more cases of toxicity and they even reported in the Lancet paper just recently that there were three treatment related deaths for SBRT. So just a reminder that we have to be careful about how we do this, but if you do it well, you're gonna get as... You may potentially be giving your patients a real boost that they won't get with conventional management.
One of the last things I wanna touch on before I close is the idea of cost effectiveness. Is it more cost effective to give more expensive but more efficacious treatment upfront or should you go for the cheaper option and then see what happens? Because you're saving...with conventional palliation, of course that's not as expensive as SBRT. And so, is there gonna be a cost effectiveness?
Well, if you think in the spine, how much does it cost to have local failure in the spine after radiation of any kind? Well, these patients are often highly symptomatic. They often require surgery for salvage, hospitalization. It's not difficult to see that, in the case of the spine where the consequences of local failure is so high that it may be cost effective to give more...effective even if it's more costly treatment upfront in the long run, especially as this analysis shows. This shows that if you... Whoops, excuse me. If you have a patient who's gonna survive more than 11 months, spine radiosurgery actually meets the criteria for cost effective. And so, it may not be in every single patient that it's gonna be cost effective. But if you have a patient with relatively good survival, and we know that each week there's a new targeted therapy and survival is improving incrementally every month almost, it seems like, for patients with stage 4 disease, that it makes sense to be considering radiosurgery as perhaps even more cost effective because it is so much more efficacious and much more durable than conventional fractionated radiation.
So, because of that, the NCCN Guidelines are now including... This is the 2018 version, the NCCN Guidelines for metastatic spine disease that consider radiosurgery if you have a patient who has progression after conventional radiation has significant pain or tumor progression, particularly, they say, for radio resistant tumors, such as renal cell, melanoma, sarcoma, hepatocellular, colorectal, and non-small cell lung cancer, or in the setting of recurrence after prior radiation.
So stereotactic radiosurgery has now come of age. It's now in the NCCN Guidelines. And I think that with the current level of knowledge that we have that this is a fairly appropriate summary of who is going to benefit, who of our patients are gonna benefit from stereotactic radiosurgery. It's our good prognosis patients. And many of our stage 4 patients are now going from not so good prognosis into the better prognosis category, and their survivals are improving. So more and more patients are going to benefit from a more aggressive local therapy paradigm such as stereotactic radiosurgery.
So conventional radiation is not expensive. It's quick, it's effective in the short term. So there's still an important place for it, I believe. Its response, however, is dependent upon histology and it may be better suited for regional treatment treating multiple segments of the spine, whereas SBRT paradigms, higher dose per fraction results in more durable responses than conventional radiation. You can have complications, but done well, it's uncommon to have serious adverse complications. There's a high rate of tumor control regardless of histology or tumor size. It's better suited, however, for discrete targets for shorter treatment times. It's more cost effective in patients will live long enough to enjoy the added benefit of high dose per fraction treatment.
All right. So, with that, I'll close, and thank you for your attention.
