The SOLO1 trial is the first international randomised phase III study to incorporate a PARP inhibitor, in this case olaparib, into the treatment context of front line ovarian cancer chemotherapy. It enrolled women who were newly diagnosed stage 3 or 4 high grade serous or high grade endometrioid ovarian cancer who had a BRCA mutation, either germline or somatic. They had to be of excellent performance status and they had to have an attempt at a cytoreduction. They also had to have either a complete or partial response to chemotherapy following their front line chemotherapy which is common. Then they were randomised in a two to one fashion to receive olaparib 300mg tablets twice daily or placebo and were stratified by complete or partial response. Treatment was continued until disease progression or if no disease progression was noted treatment was stopped at two years. The exception to that was for patients who entered SOLO1 with partial response if they still had evidence of disease but it was stable at that two year mark and the patient’s treating physician felt that they were obtaining clinical benefit they could extend treatment beyond two years with permission from the study.
The primary endpoint was investigator assessed progression free survival and then key secondary endpoints were progression free survival as assessed by the blinded independent central review service, sort of a sensitivity analysis for the primary endpoint, progression free survival to time to first and time to second subsequent therapy and then health related quality of life and certainly we’ll be looking at overall survival but those results were not presented at this meeting.
When we look at the women who came on to study they represented what we see with front line BRCA positive ovarian cancer. They were a very good performance status, the majority of them were stage 3 as opposed to stage 4, about 80% were stage 3, 75% of them were BRCA1 versus BRCA2. Although we allowed both germline and somatic patients to come on study, if you came on study with a somatic test it had to be done with a commercially available test for tumour at that time but in 2013/14 there really weren’t any of those tests so only two patients came on to study with a somatic mutation only, there was one each BRCA1 and BRCA2. About 75% of patients had an attempt at front line cytoreduction and of those 75% were cytoreduced to no gross residual disease. So this was a very good performance status group of patients, 80% of patients had a complete response when they started SOLO1 therapy.
When we look at the primary outcome, the group of women who were randomised to placebo performed exactly as we expected them to do. So their median progression free survival was 13.8 months. You have to remember that that is from the completion of chemotherapy, so most studies measure from diagnosis or from the beginning of chemotherapy so that may seem low to some people but actually if you add in 4-5 months of chemotherapy it’s 18 or 19 months and that’s really what you would expect for this control group. So we felt like they performed exactly as we expected.
Among the women that were randomised to olaparib, however, their median progression free survival has not yet been reached with over 40 months of follow up. So the difference between those two curves gives us a hazard ratio of 0.3 which said in other ways would equate to a 70% reduction in the risk of progression. Although we do not have a median progression free survival based on investigator assessed PFS, we have done several sensitivity analyses to account for any potential bias. Number one, all of the hazard ratios of those analyses were very consistent – 0.27-0.3 - but two of those analyses did just give us a median for that olaparib group that’s somewhere between 47 and 49 months. So we can estimate the improvement in progression free survival for use of olaparib at about three years over that which would be expected with no therapy, so profoundly clinically and statistically significant.
Further, at the time of data cut-off progression free survival 2 was able to be analysed as well and this will be the final analysis for PFS2 which demonstrated that we maintained statistical significance for those patients who were randomised to olaparib with a hazard ratio of 0.5. That is even more significant when you consider that over a third of the patients who were initially randomised to the placebo arm basically crossed over and received a PARP inhibitor, we don’t know which one or ones, but they received a PARP inhibitor as part of their second line therapy. So with that amount of crossover we still see a maintained superiority of front line olaparib use with a hazard ratio that’s highly statistically significant of 0.5. There were really no new safety signals noted, the toxicity profile was very consistent with what has been demonstrated in prior studies of olaparib, just low grade manageable non-hematologic toxicities – fatigue and GI. The only notable hematologic toxicity is a 21.5% incidence of grade 3 or higher anaemia which is manageable with transfusions and dose modifications and is really no different than what we see when we use olaparib in the second line.
The final endpoint that we could report at this meeting was the primary quality of life endpoint which is the FACT-TOI. For reference, a clinically meaningful change in the FACT-TOI from baseline is plus or minus 10 points. So we did see over time a 3 point decrement in the FACT-TOI score from baseline in those women randomised to olaparib but over the 24 months of therapy the TOI scores were broadly overlapping and, although statistically significant, it is not felt to be clinically meaningful at all.
So we feel that these data really support immediate, as soon as it’s available, access to olaparib as front line maintenance therapy for patients with BRCA mutated ovarian cancer. We’re hopeful, given the positivity of PFS2, that this will translate into more women with long-term disease free survival and potentially even cures but that will take years of follow up to determine so we’re not going to be reporting that out for a long time, the overall survival endpoint, but PFS2 is a surrogate for that and so we’re encouraged and hopeful. Really the safety profile is very consistent so there’s not any signal that we should not be incorporating this into front line therapy.
The original design of the study when we were designing it, we really wanted to push it out a little bit further, maybe to three years or more, but there was concern because really this population of patients has never been studied in a prospective clinical trial, BRCA positive advanced ovarian cancer. We have registry data, we know they have better prognostic features but we’ve never done a trial only of BRCA patients in the front line. So there was concern that you did have some patients who would be cured with chemotherapy alone who you are unnecessarily exposing to prolonged therapy that they didn’t need so we didn’t have anything to counter that. The two year mark was picked really because of what I said, you want to cover the expected progression free survival of your control group and a little beyond that, that’s the time period you want to cover so that’s really where the two years came from, and really it’s interesting that the morphology of the survival curve does not change at two years. So it’s very different than what you see with maintenance studies of bevacizumab which is also a great drug for ovarian cancer; it shows nowhere near this magnitude of benefit but it’s a very effective drug for ovary. But when you stop maintenance bevacizumab those curves start to come back together almost immediately which has prompted a whole other generation of studies exploring longer duration of therapy, those should result out soon. We don’t see that here, in fact you see no change in the curve at two years, so what does that mean? It’s hard to say. When you look at the control group survival curve you do see this initial drop off in that first year of people recurring, women recurring, so presumably they had subclinical disease that just grew right back when they stopped therapy. So the hope would be, certainly the olaparib group had the same thing but we may have killed that disease with the first one to two years of olaparib and now they’re just going to ride out for as long as we can. So we see the flattening of the curve there that seems to be indicative of a durable response to two years of therapy.
So that gives us a lot of confidence and we need to be very clear about that with patients. This isn’t like an option to stop at two years, this is how the study was designed for a reason and there’s no loss of benefit for those patients who are in complete response, those that still have disease can continue. But those patients in complete response should stop at two years because we don’t see any reason not to at this point, we’re not seeing a drop off of the curve. No treatment is safer than treatment every day of the week so to keep people disease free and you don’t have to treat them, that’s a win-win.