Written by: Suzanne Jenkins and Kenneth Thress
This week, the US Food and Drug Administration approved a cutting edge plasma-based test, known as a ‘liquid biopsy’, that can identify non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor (EGFR) T790M mutations who are eligible for treatment with AstraZeneca’s new T790M tyrosine kinase inhibitor (TKI).
The new test helps break down a significant barrier for lung cancer patients in the US. Up to now, physicians have relied on a fresh biopsy, or tumour tissue sample taken at the time of disease progression, to detect resistance mutations, such as T790M, in the EGFR gene and match patients to suitable treatment. However, such biopsies require a hospital procedure and, in a few cases, may even pose clinical risks. The need for non-invasive testing is particularly acute when detecting resistance mutations, where up to 40% of patients may find it impossible to donate a tissue sample, because tumours are hard to reach or patients are not well enough to undergo biopsy.
The cobas® EGFR Mutation Test v2 companion diagnostic, developed in collaboration with Roche Molecular Systems (RMS), takes a different approach. It looks for minute fragments of circulating tumour DNA (ctDNA) in the patient’s bloodstream, which allows a diagnosis just using a blood sample. This enables patients to avoid an invasive tissue biopsy when their lung cancer progresses after first-line treatment with an EGFR-TKI. Such tests are available in Europe, but this is the first time a test has been approved in the US for resistance mutations.
Behind this important advance lies over a decade of research by scientists at AstraZeneca’s Innovative Medicines and Early Development (IMED) Biotech Unit, who recognised the potential advantages of so-called ‘liquid biopsy’ for patients, clinicians and healthcare providers at a time when the science was emerging and few were willing to commit resources to such challenging research
ctDNA testing on trial
‘Liquid biopsy’ can check for tiny fragments of circulating tumour DNA (ctDNA) in a patient’s plasma that have been shed from any tumour in their body, however inaccessible.
Following an initial ctDNA research collaboration with Cancer Research UK in 20061 and undeterred by early concerns over the reliability of using such small amounts of DNA, IMED scientists committed to collecting blood samples from patients in our NSCLC clinical trials of novel EGFR-TKIs. In this way, they could investigate the effectiveness of ctDNA testing and compare results with those from tumour tissue samples.
Every patient sample is precious, whether it is a piece of tissue or a tube of plasma, and patient samples collected during our studies were analysed responsibly, to maximise correlative information. Research soon showed that plasma, rather than serum, was the best source of ctDNA, with fragments containing as few as 150 base pairs (from approximately 200,000 building blocks that make up the EGFR gene) yielding reliable information about EGFR mutations.
IMED scientists focused on demonstrating that novel ctDNA tests were not only sensitive (i.e. they could detect small amounts of mutated EGFR sequence within a blood samples), but were also specific (i.e. they didn’t detect EGFR mutations where they didn’t exist). Not least, we needed to show that results obtained with ctDNA tests reflected those obtained with tumour tissue samples.
Reliability of ctDNA confirmed
In 2014, data was published showing high concordance between EGFR mutations identified in ctDNA and tumour tissue2 and, a year later, AstraZeneca set a new precedent for EGFR testing in Europe when a ctDNA companion diagnostic test was approved for use with our first EGFR TKI, if a suitable tissue sample was unavailable.
In 2015, we reported good concordance between the results of plasma ctDNA tests designed to detect the EGFR T790M mutation and tumour tissue tests.3 We also showed over 90% concordance between results of the cobas® plasma EGFR T790M test now approved by the FDA and a next-generation sequencing (NGS) reference method – widely seen as state-of-the-art methodology.4
In support of the FDA application for the cobas® plasma T790M test and to ensure a complete scientific and clinical picture, IMED scientists felt it was essential to demonstrate that patients with the T790M mutation detected by plasma ctDNA respond just as well to our new T790M-targeting EGFR TKI as those whose mutation is detected in a tumour tissue sample. In a paper in the Journal of Clinical Oncology published online in June 2016, equivalent treatment outcomes for patients following plasma ctDNA and tissue T790M testing were reported.5
Additional benefits: T790M heterogeneity
When resistance to a first-line EGFR TKIs is identified, it is important that clinicians are able to look for the T790M mutation as soon as possible, in order to offer patients the right treatment options. As our research has progressed we have learned that, despite those initial concerns about reliability, a ctDNA test for T790M can offer additional clinical benefits to a tissue biopsy.
Patients with mutations that are sensitive to EGFR-TKIs have these mutations in all their tumour cells, but this is not the case for the resistance T790M mutation. This ‘T790M heterogeneity’ means that mutations may be missed if a tissue biopsy does not pick up the cells containing the T790M mutation. In contrast, a plasma test collects DNA fragments shed from a wide array of tumour cells throughout the body, so it may detect mutations that are missed in the tissue sample. This advantage confirmed for us the urgent need for a plasma-based test to help patients get a fast and accurate diagnosis.
Ensuring access to ctDNA testing
We endeavour to ensure that plasma ctDNA EGFR T790M mutation testing is available to as many patients with NSCLC as feasible so that they can benefit from the convenience, speed and reliability of the new technology, and healthcare providers can take advantage of the lower costs associated with reduced need for surgical biopsy.
AstraZeneca is implementing educational initiatives to ensure quality and consistency of test results and is working to overcome logistical obstacles to optimal access to ctDNA testing. For example, for clinicians who are unable to process plasma and freeze and ship samples to testing centres on dry ice, we are exploring the use of specialised blood collection tubes that do not require immediate plasma processing and can be transported by post.
Future potential of ctDNA testing
Will plasma ctDNA testing completely replace tissue biopsy for detecting genetic mutations in tumours? We think that’s unlikely, but expect that its use to complement tissue biopsies will grow. Some tumours simply do not shed DNA into the blood, so it cannot be collected to identify genetic mutations. That is why it is recommended that any patient whose plasma ctDNA test is negative for an EGFR T790M mutation should have a tissue biopsy to ensure that opportunities for targeted therapy are not missed 4.
Plasma ctDNA testing has now been integrated into all our oncology research programmes and, as the technology advances, we expect it will be used to monitor response to treatment, based on the changing burden of tumour mutations, as well as to diagnose them. Scientists are now investigating how the utility of ctDNA could be extended in the future, for example to include next generation sequencing or for use in monitoring.
The last decade has seen a remarkable revolution in understanding the ways in which tumours evolve and in the emergence of fast, reliable technology to detect genetic mutations for which novel treatments are being developed. We and our colleagues across the business are very proud to have played our own small part in that revolution, and we look forward to seeing yet more of these possibilities become reality for patients.
- RE Board et al, Br J Cancer. 2009 Nov 17;101(10):1724-30. doi: 10.1038/sj.bjc.6605371. Epub 2009 Oct 27.
- Douillard JY, Ostoros G, Cobo M et al. First-line gefitinib in Caucasian EGFR mutation-positive NSCLC patients: a phase-IV, open-label, single-arm study. Br J Cancer. 2014 Jan 7;110(1):55-62.
- Thress KS, Brant R, Carr TH et al. EGFR mutation detection in ctDNA from NSCLC patient plasma: A cross-platform comparison of leading technologies to support the clinical development of AZD9291. Lung Cancer. 2015 Dec;90(3):509-15
- Jenkins S, Yang J, Ramalingam S et al. Plasma ctDNA analysis for detection of EGFR T790M mutation in patients (pts) with EGFR mutation-positive advanced non-small cell lung cancer (aNSCLC) J Thoracic Oncology 2016; 11 (4) Supplement, S153–S154
- Oxnard GR, Thress KS, Alden RS et al. Association Between Plasma Genotyping and Outcomes of Treatment With Osimertinib (AZD9291) in Advanced Non-Small-Cell Lung Cancer. J Clin Oncol. 2016 Jun 27. pii: JCO667162. [Epub ahead of print]