Dan O’Shannessy, A C-level executive. Advisory board member at Cellsway. Expert in diagnostics, translational medicine, and pharma.
I had the pleasure of speaking with Dan O’Shannessy, an Experienced biotech executive, with vast knowledge of the liquid Biopsy space, who has taken multiple products from the “Bench” through regulatory approval. In the interview, we reviewed some of the most crucial considerations in developing and setting the ground for approval of a liquid Biopsy assay, with the focus on circulating tumor cells.
There are many CTC enrichment techniques, each based on different CTC and WBC properties, with their own advantages and disadvantages. In your view, as these technologies undergo evolution in the past years, how important is the enrichment technique element in the grand scheme of all the parts of a clinical assay?
This depends on the clinical assay, both the technology behind it and the clinical indication. If a requirement of the clinical assay is to have an enriched CTC population, for whatever reason, then that aspect of the assay is critically important.
” ..The requirements for standardization of CTC-based assays begin with a clear objective, delineated in an Intended Use Statement, understanding that enrichment of CTCs is but the generation of the sample used for analysis…”
For example, CTC enumeration or expression of a specific marker such as AR-v7, by immunofluorescence, does not a priori require enrichment, although it may benefit from fewer ‘background’ cells resulting from enrichment. This is evidenced by the availability of clinical assays performed for AR-v7 in exactly this way.
On the other hand, mutation analyses or gene expression analyses benefit greatly from CTC enrichment by diminishing the background signal from contaminating non-CTC cells. In fact, techniques such as NGS and RNAseq require enrichment. The degree to which background cells, primarily neutrophils, need to be reduced, i.e., the degree of enrichment required, remains an open question since, as with all such essays, it’s a matter of sensitivity and specificity. Unfortunately, very few, and most often very small studies evaluating the potential of enriched CTCs for mutation/expression analyses have been published so we’re still a long way from a clinical application.
It needs to be stressed that very few head-to-head comparisons of the various CTC enrichment technologies have been performed, be it for enumeration or mutation/expression analyses. As such, comparisons are most often based on ‘not so similar’ literature reports involving limited sampling and disparate techniques. At the end of the day, an assay needs to be fit-for-purpose and for CTCs it is possible if not probable that no single enrichment technology will be fit for all purposes.
” .. mutation analyses or gene expression analyses benefit greatly from CTC enrichment by diminishing the background signal from contaminating non-CTC cells .. “
If we were to break the CTC-based assay to its basic elements requiring standardization, could you rank all elements in descending order from the most difficult to the “least” difficult to standardize? Could you explain the reason you chose the highest-ranked element?
We first have to define what constitutes a CTC-based assay – is it fluorescence-based enumeration, sequencing-based, expression analysis etc. since each assay will have its own specific requirements from a standardization perspective.
Be that as it may, CTC-based assays, or technologies, do come with a unique set of standardization challenges, driven by the fact that you’re dealing with live cells, for the most part.
Probably the most difficult aspect is sample stability – the sample is whole blood and live cells. But again, dependent upon the clinical assay requirements, samples could be fixed by a number of approaches and/or collected directly into commercial blood tubes that stabilize and/or fix cells by various means, aiding stability. In general, sample stability dictates assay accessibility in many respects. If samples need to be processed and analyzed within 24hrs of collection, there are logistic limitations to accessibility. If, however, samples are stable for at least 72hrs such restrictions are by and large negated. The biggest issue here is how to assess stability on real-world samples, not contrived samples where cultured cells are added/assessed. Real-world samples, i.e., patient samples, are notoriously variable, and not just in the number of CTCs.
” ..one does not know the CTC content of a patient’s sample (number, size, EMT status etc.) and therefore cannot determine recoveries..”
This leads to the second most critical standardization element and that is recovery. Recovery is most routinely performed using cultured cells spiked into normal blood draws. While this is useful and informative, it bears no resemblance to real-world patient samples and the inherent variability in phenotypes, including size and shape, of CTCs. But therein lies the problem, one does not know the CTC content of a patient’s sample (number, size, EMT status etc.) and therefore cannot determine recoveries. The best one can aim for is repeatability, i.e., how repeatable is the technique/technology for enriching CTCs from the same patient. This is not a frequent assessment of CTC enrichment technologies, not just because it’s not easy to do. And recovery again needs to be taken in context to the clinical indication. How important is recovery to the actual assay readout? The requirements for standardization of CTC-based assays begin with a clear objective, delineated in an Intended Use Statement, understanding that enrichment of CTCs is but the generation of the sample used for analysis.
We witness a rapid development of next-generation sequencing (NGS) technologies that led to a cost reduction and integration with several Liquid Biopsy tests. That said, we have also seen an increase in companies that are focusing entirely on the isolation of rare and circulating tumor cells and their characterization via immunological methodologies. In your opinion, what would be the ultimate solution to become an actionable clinical endpoint?
Very difficult to answer and I would not presume to know exactly what the ultimate solution is. I will say that I firmly believe that knowledge is power, and this is particularly true in medicine and especially personalized medicine. I also believe that the information obtained from CTCs directly – including number, phenotype, mutation, and expression analyses – is additive to information received from cfDNA analysis for example. Where exactly these two techniques overlap or add to one another I think is the question – and again this can only be answered by framing the clinical question first and then performing the analyses on clinical samples. We all know that we can isolate CTCs, demonstrate antigen expression, phenotype, and sequence their genomes, even down to single-cell analyses. The real questions are (a) cost and (b) clinical relevance.
In addition to CTC enumeration, downstream analysis of CTCs provides valuable insight into mechanisms of cancer metastasis, in the attempt to uncover targets for effective pharmaceutical treatments. In your opinion, what is the potential of biomarker development and drug sensitivity testing on CTCs to provide “real-world” therapeutic prediction response conditions?
Drug sensitivity testing is a real option for isolated CTCs, but only if one can expand the isolated CTCs since there are simply not enough isolated from a single draw to perform such analyses. In so doing, you have to assume that the CTCs have not changed or morphed during said expansion. Secondly, only certain drug classes are amenable to drug screening techniques. For example, many of the new bifunctional strategies engage the immune system, or components thereof, for effect. So having CTCs per se is insufficient to assess such drugs and adding in NK cells or the like simply complicates the assay. Further, short term effect does not address the development of resistance. Finally, the underlying assumption is that CTCs are an accurate representation of the heterogeneity of the solid tumor – this may or may not be correct for many, many reasons.
