Laboratories - the forgotten stakeholder
"Pathologists and laboratories are often overlooked as key stakeholders by pharma and this factor can have a considerable impact on the uptake of targ...
Jordan Clark and Peter Keeling of Diaceutics critique a journal article exploring the effectiveness of parallel FISH and IHC methods for the detection of ALK. They go on to discuss the financial implications for pharma companies if they do not engage with laboratories before they launch their targeted therapy.
We note with great interest a recent paper in Journal of Thoracic Oncology Vol. 9 Number 3 March 2014[i], exploring the effectiveness of parallel FISH and IHC methods for the detection of ALK status in 3,244 NSCLC patients treated in Bordeaux and Rennes in France.
The paper provides an excellent analysis of the complexities imposed upon pathology labs around molecular testing for new targeted therapies. For example, as was found in a recent BCR-ABL, EU laboratory validation study undertaken by the laboratory network division of this company[ii], it is often the pre-analytical steps rather than the testing method chosen which is at issue. The authors cite, “Many pre-analytical factors may account for the apparent discrepancies between both methods, suggesting that hierarchical screening may underscore ALK-positive cases”. Overall, the paper highlights the need for pharmaceutical companies to engage with leading laboratories, preferably well in advance of targeted therapy launches, to determine the most efficient testing methods and combinations.
In order to illustrate the financial benefit of such often-questioned pre-launch engagement with leading laboratories, we have applied our proprietary financial model (Diaceutics Rx/Dx Financial Model) to understand the REAL WORLD microeconomic implications for pharmaceutical companies considering their companion testing technology strategies. The results of this analysis enable us to put a dollar figure against technology and communication strategies deployed by pharmaceutical companies and are described below.
Diaceutics Rx/Dx Financial Model enables the integration of test performance metrics into therapy drivers and ultimately a return on investment analysis for targeted therapies. The model is described elsewhere in more detail[iii].
Using the Diaceutics model and the extensive data supplied from the French study, we determined the financial impact of a single versus a reflex testing strategy with this study cohort. We focused on only two primary financial drivers:
Three scenarios were developed to illustrate the patient and revenue impact of:
Table 1 shows key metrics used across the three scenarios and Diagram 1 plots the financial impact to pharma (in this instance Pfizer) of each of the scenarios assuming an impact in 2012 (1 year) only.
Table 1. Comparison of the microeconomics of Rx/Dx dependency in personalized medicine.
Diagram 1. Scenarios revenue comparison.
This simple analysis suggests that in the French patient cohort a single test strategy would create a loss of treatment opportunity in 28 per cent of ALK positive patients. Assuming the same P2P ratio, ten fewer patients would have received crizotinib, representing an approximate loss of €500,000 in revenue to Pfizer in one year alone.
Scaling this study on a pro rata basis to illustrate the impact a single test strategy would have across France is helpful. There are approximately 17,480 NSCLC patients in France per annum[v] and this study represents only 18.6 per cent of those patients. Assuming similar real world testing metrics were achieved in other parts of France, the additional Crizotinib revenue accruing to a reflex versus a single test strategy would be €2.7 million in one year alone.
We note, however, that the P2P ratio in this study was extremely low at approximately 30.1 per cent. Of 150 ALK positive patients determined by reflex testing in this study, 46 patients went on to receive crizotinib in the study, 44 patients received other chemo therapies or erlotonib or gefitinib and the remaining patients were either lost to follow-up, had surgery or died. It is assumed that only patients who received other chemotherapy are potential additional candidates for crizotinib treatment based on analogues of efficient companion diagnostic markets. It is also assumed that higher levels of testing confidence along with favourable reimbursement or better physician education could convert and additional 20 per cent of ALK positive patients onto crizotinib. Thus, in our third scenario, reflex testing combined with an optimized P2P strategy would potentially double the number of patients on crizotinib. In this study group alone, this would represent an additional €2.3 million of crizotinib revenue over the single test strategy and, on a national French basis scale, up to an approximate additional €12.4 million in revenue opportunity.
Despite the 10+ years of data now describing the financial interdependency of targeted therapies and their companion diagnostics in oncology, there remains an ongoing dialogue with executives tasked with launching novel personalized medicine strategies as to the benefit of deeper involvement in creating efficient testing outside the contractual relationships with a diagnostic (kit) provider.
From a technical standpoint, this French study provides a real world glimpse of the impact of pharma partnering and the technology decisions (e.g., one test or two, hands-on or hands-off laboratory engagement) potentially to be made in clinical care. Our observation is that optimization of testing is too often left as a post launch focus, triggered perhaps when the therapy revenues are not matching expectations. This is not peculiar to ALK testing – our tracking of companion diagnostics illustrates these same real world implementation issues have existed with all the headline markers and their respective targeted therapies including HER2, BCR-ABL, KRAS and EGFRm.
The goal of our additional lens on this study is to provide some ‘back of the envelope’ understanding that when the clinical goals of efficient testing (defined as the optimum combination of test methods and education) is undermined by reactive or late pharma involvement, there is a direct shareholder impact. Our analysis supports our overall contention that a timely dollar spent on creating efficient testing markets in advance of therapy launches delivers a potentially higher return on investment than other more traditional pharmaceutical marketing approaches.
Our thanks to the extensive research team involved in this French study for furthering the dialogue on this urgent and important topic.
[iv] P2P is a phrase coined by Diaceutics to define the complex relationship between prescribing behaviour and education levels and confidence with companion testing. Analogues indicate that even though a patient may test positive for a given biomarker, targeted treatment does not automatically follow even though the relationship between the biomarker and therapy may be well described in therapy labels and clinical guidelines. There is in short a multifactorial relationship between timeliness of education on the biomarker, confidence in testing quality, turnaround times and competitive marketing directly connected with P2P rates. Our data show that P2P needs to be earned and cannot be automatically assumed to be an 80-90% ratio.