| The data isn’t published yet (presumably this was shown during the ASCO conference this weekend) but this is a bit misleading and the results seem underwhelming. The DFS numbers are not new and known, the reason this is in the news as it’s the first report of OS numbers from the initial ADAURA trial. I can’t be definitive without seeing the data yet but: 1. Osimertinib is not that new, in the context of curative intent disease (i.e. resectable stage II-IIIA) it is currently (ideally) used post adjuvant chemotherapy (the only treatment to date with an overall survival benefit). We need clarification on what the placebo arm is and what the subgroup analysis showed. What it should be (and presumably) is an “active surveillance protocol” where patients underwent short course adjuvant platinum based therapy and then followed with imaging. Recurrences are then treated with systemic or locoregional therapy. As this is an update of the ADAURA trial we know that only 40% of patients received the standard of care platinum adjuvant therapy, this article claims an OS benefit was seen in all groups but we don’t have the numbers for the subset of patients who received appropriate adjuvant treatment. 2. Main criticism of the ADAURA trial thus far has been that the results only report “disease free survival”, while that intuitively makes sense as a metric what we really care about is “overall survival”. There are several reasons but to keep it simple this is now the third generation tyrosine kinase inhibitor, the first 2 also had DFS improvements (albeit not as dramatic) but failed to show OS benefits. 3. Osimertinib is expensive. Following the ADAURA protocol (3 years of adjuvant therapy) would have an incremental cost (ICER) somewhere around ~$3-450,000 per patient. “Willingness to pay” is variable, in most places it’s $50,000/quality adjusted life year. Some in the US are pushing for this to be ~$190,000/QALY (3x GDP). Based on extrapolated DFS and earlier OS data in the last year it was estimated that the OS would be around 5-6%, based on this threshold a system would need a willingness to pay of ~$320,000/QALY. Conversely, to meet the GDP threshold above OS would need to be ~20%. In a recent Canadian economic analysis they modelled 6% OS at 10 years and arrived at a ICER of ~$40,000 suggesting this protocol makes economic sense. To my knowledge this is the first report suggesting cost effectiveness and contradicted the Health Canada modelling. 4. Based on this news article, there was an absolute reduction in OS of 10% at 5 years which seems underwhelming given that we know most of the patients did not receive adjuvant platinum based chemotherapy, the OS in the subgroup that received both treatments is what matters here. From the US analysis, this would still not meet the willingness to pay threshold. The Canadian one would need to be re-run with 5 year numbers to see what the ICER is. Overall, it’s potentially a very positive result but at face value the OS numbers seem less impressive than anticipated. If someone has the $ and an EGFR ex19 mutation there is definitely benefit but it remains unclear whether this is cost-effective for a system vs other treatment options we have. (N.B. These numbers are approximate from my recollection of the literature but I can dig up references if something seems off. For background I’m a radiologist focused on oncologic imaging, this has been a hot topic in rounds/case conferences for a few years now hence my familiarity.) |