Angle — Update 17 November 2015

The Parsortix cell separation system is able to capture circulating tumour cells (CTCs) from a blood test, constituting a liquid biopsy. Endorsement of its medical utility by key opinion leaders will be crucial to securing its widespread use in the diagnosis and treatment of cancer. H215 saw positive evaluations in multiple cancer types, adding to the H115 positive evaluation in ovarian cancer.

CRUK Manchester Institute published the results of a study comparing Parsortix to CellSearch for CTC isolation in small cell lung cancer (SCLC). CellSearch is the only FDA-approved CTC device (approved for enumeration of CTCs for prognostic purposes in metastatic breast, colorectal and prostate cancer); it isolates CTCs using magnetic particles coated with anti-epithelial cell adhesion molecule (EpCAM) antibodies. Once extracted, CTCs are enumerated on the basis of their morphology, expression of cytokeratins and a lack of the CD45 leucocyte marker. However, CTCs are heterogeneous, and CTCs can change phenotype in a process called epithelial-to-mesenchymal transition (EMT) that results in reduced expression of the target markers. Therefore, antibody-based systems limited to epithelial phenotypes could result in false negatives. In addition, mesenchymal CTCs have been implicated in the process of cancer metastasis,1 and so have significant clinical implications. Parsortix uses a patented step-based microfluidic technology to capture CTCs on the basis of their size and compressibility, and therefore is not limited by cell marker bias. The main study findings were as follows:

Lung cancer is the leading cause of cancer death globally (cancerresearch.org); its location makes initial biopsy difficult, and repeat biopsies even more so. SCLC accounts for 15% of all lung cancers. In advanced SCLC large numbers of EpCAM expressing CTCs are seen, but the CTC heterogeneity demonstrated in the study underlines that one cannot solely rely on marker-dependent CTC isolation methods.

Further, as Parsortix enables the harvesting of the CTCs for analysis, it can be used to provide additional information for treatment stratification, particularly as many cancer therapies are now targeted to specific mutations of the cancer. In addition, biomarkers are increasingly used to guide therapy. The recent FDA approval of the immunotherapy Keytruda (Merck) was limited to PD-L1 positive non-small cell lung cancer, as high tumour expression of PD-L1 appears to predict those most likely to benefit. This highlights the significant value-add that Parsortix could provide to cancer care, across multiple cancer types.

cfDNA consists of small DNA fragments from the dead cells. In cancer patients, a proportion of the cfDNA will be circulating tumour ctDNA (ctDNA), which is released as fragments from necrotic and apoptotic tumour cells. The potential of liquid biopsies utilising ctNDA has recently been highlighted in a case study that followed a single patient with metastatic breast cancer over three years, extensively comparing biopsy and plasma samples. The study found that mutation levels in the plasma reflected those inferred from sequencing tumour biopsies and tracked different treatment responses across metastases. While these results would need to be confirmed in a larger cohort of patients, the findings lend further support to the potential utility of liquid biopsies in precision cancer medicine.

CTCs offer advantages over ctDNA as the information provided is not limited to DNA only, so also allowing for analysis at the RNA and protein levels. Further, having come from dead cancer cells, information from ctDNA may not be as clinically relevant. Previously, one of the great appeals of ctDNA is the ease with which it can be isolated from blood plasma, with CTCs being comparatively more difficult to capture. Now, Parsortix offers the capability to capture CTCs. When designing Parsortix, Angle recognised the utility of ctDNA and designed the machine so that the same blood sample can be used to harvest both CTCs and ctDNA, as demonstrated by the CRUK study. This provides flexibility to those using the machine and also enables the direct comparison of molecular readouts from both cfDNA and CTCs.

The BCI published the results of a study in which it evaluated the Parsortix system for CTC harvest in prostate cancer. The study compared Parsortix to CellSearch and to IsoFlux, another antibody-coated magnetic bead-based CTC isolation system that relies on cell surface markers to select CTCs. The findings were as follows:

The study concluded that, if the clinical value of Parsortix-captured CTCs can be confirmed, Parsortix will make CTC analysis much easier for clinical diagnostic application owing to its simple approach, ability to capture varying phenotypes of CTCs, with shorter isolation time, higher purity and lack of contaminating beads. Incidentally, the study used an earlier version of software with Parsortix, with the newer software enabling it to process larger blood volumes, at greater speed, while maintaining optimal capture.

Other than skin cancer, prostate cancer is the most common cancer in American and European men, and the second most frequent cause of cancer-related death.2,3 As CTCs hold valuable information about the tumour, including the genetic mutations that drive the tumour’s growth and resistance mechanisms, liquid biopsies could provide an alternative diagnostic and monitoring tool for real-time personalised treatment of prostate cancer. Further work is being done in the study, with further announcements expected in H116.

In September, the University of Southern California Norris Comprehensive Cancer Centre presented the results from the first phase of its work with the Parsortix system in breast cancer. The work compared RNA molecular analysis of a Parsortix liquid biopsy with that of conventional solid biopsies of metastatic sites in metastatic breast cancer patients. Parsortix provided highly enriched CTCs from all of the patients initially tested (n=4); the CTCs were suitable for rare cell amplification and RNA sequence analysis to detect biomarkers. Comparable gene expression between the CTCs and traditional solid biopsies was found, suggesting that a simple blood test may provide the same information as that provided by a solid biopsy on a repeatable and ‘real-time’ basis. Biopsies of metastases are often difficult due to the common locations of metastases so require invasive surgery that carries risks, are not easily repeatable and are costly and time-consuming.

Not only was the information gleaned from the CTCs comparable to that from the solid biopsies, the study found evidence of heterogeneity in the CTCs captured with Parsortix. This is particularly significant as Parsortix-harvested CTCs may provide a more complete genomic landscape of the tumour than that provided by a single-site biopsy. This is highly relevant for breast cancer, where HER2 status guides therapy and overt distant metastases and CTCs have been found to have discordant HER2 statuses compared with the primary tumour in up to 30% of cases.4 The second phase of the work will include analysis of the remaining patient cohort and the profiling of CTCs sequentially over time to support treatment decisions. Similar findings in a larger patient sample would help substantiate these initial findings.

In H115, the Medical University of Vienna reported the “unprecedented sensitivity and specificity” of an approach using Parsortix alongside their RNA markers in the diagnosis of ovarian cancer, with 100% sensitivity and specificity with analysis of 30 RNA markers.

At the European Cancer Congress European Society for Medical Oncology 2015, the university reported further data from its work with Parsortix in the detection of ovarian cancer. The new data relate to further work comparing the use of Parsortix and qPCR (quantitative polymerase chain reaction), with conventional immune-fluorescent (IF) staining for the molecular analysis of CTCs in patients with gynaecological cancers. Compared to conventional IF, qPCR-based analysis resulted in a significantly higher CTC detection rate in blood samples taken from patients with gynaecological malignancies, across all cancer types the mean detection rate was 70% vs 26%. Using the 30 gene markers (threshold set for 100% specificity) 100% sensitivity was achieved in ovarian cancer patients (both primary and relapse; n=7) and all cancer patients were correctly classified by the Parsortix/qPCR-based analysis, whereas just one of the sample was IF-positive.

This provides further support for an approach combining qPCR and the Parsortix system for the diagnosis of ovarian cancer and other gynaecological malignancies. A clinical trial evaluating this approach for the pre-surgery triaging of ovarian cancer is due to commence in Q415. Success of the study could provide validation for the Parsortix system and lead to clinical sales for use in this indication in Europe as early as H2 FY17.

We value Angle at $145m or $24.6/ADS, based on a three-phase DCF model, assuming a discount rate of 10%, terminal growth of 2% and a long-term tax rate of 20% (reflecting the expected benefit of the Patent Box incentives). The breakdown of our valuation is shown in Exhibit 1. We have not changed the assumptions on our modelling, and continue to include only the sales of Parsortix for use in research and clinical sales in ovarian mass triaging, with no value assigned to its use in other indications, or the inherent value of the technology platform. The potential application in multiple cancer types, as suggested by the latest studies, offers significant upside to our valuation.

Our valuation is based on the following assumptions:

Following the $12.4m net equity raise in February and March 2015, Angle finished FY15 with $12.8m cash; Angle has no debt. Our model continues to suggest that this should be sufficient to fund operations through to 2017. We forecast $2m of illustrative financing included nominally as long-term debt on the balance sheet in 2018 (as per our policy). We anticipate profitability in 2019. The increase in net loss to $5.9m in 2015 from $3.3m in 2014 reflects the planned increase in operating expenditure on the Parsortix system.