PLAsma Genomic Testing in Patients With Advanced Non-Small Cell Lung Cancer: The PLAN Study

Background Genomic Testing in Non-Small Cell Lung Cancer Tissue genotyping of therapeutically actionable alterations is a standard diagnostic requirement for patients with advanced non-squamous NSCLC. Tissue genotyping represents a gold standard, with both a sensitivity and specificity for identifying common genomic alterations of >90%. NSCLC represents a genomically diverse cancer. Detection of genomic alterations has progressed over the past decade, now including detection of mutations such as KRAS, EGFR, ERBB2, BRAF, MET and fusions including ALK, ROS1, NTRK and RET genes, as well as co-occurring mutations on tissue samples. The ability to detect these mutations has led to discovery of molecularly defined subsets of NSCLC, and a rapid growth in the development of targeted therapies. Actionable oncogenic driver mutations, which lead to uncontrolled cell growth, are detected in up to 64% of lung adenocarcinomas. The most commonly observed oncogenic driver mutation in NSCLC is KRAS, observed in up to 30% of cases, which has led to the development of agents such as sotorasib and adagrasib to target KRAS G12C mutations after failed first line systemic therapies. The development of EGFR tyrosine kinase inhibitors (TKIs) such as osimertinib, gefitinib or erlotinib, and ALK directed therapy with crizotinib, alectinib and ceritinib has led to significant improvement in survival in patients with NSCLC over the past decade, and shifted the focus to developing improved detection methods of these mutations for first-line targeted therapy. Each actionable mutation detected has led to the rapid expansion of available therapeutic options beyond those mentioned above and changed the landscape of lung cancer treatment.

The majority of NSCLC is diagnosed at an advanced stage (stage III or IV), and it is often diagnosed with a core biopsy or cytology specimens. Oncology and pathology societies recommend that molecular testing turnaround times not exceed 10 working days yet up to 30% of these specimens are insufficient to allow for tissue genotyping necessitating repeat biopsy. Patients may thus clinically deteriorate while waiting for genomic test results. In those patients who have insufficient samples on initial biopsy and require repeat intervention, this can prolong the time from diagnosis to treatment by more than 28 days. The methods by which these mutations are detected are continually evolving, including DNA sequencing, DNA allele-specific testing, RNA sequencing, fluorescent in situ hybridization (FISH) to detect fusion genes, and next generation sequencing (NGS). Both DNA and RNA sequencing have been shown to detect actionable mutations, but certain large panels can miss small exon skipping or fusion alterations, and thus a sequential approach may be warranted. In certain mutations, RNA sequencing can detect a higher proportion of MET exon 14 skipping. Frequently a hotspot panel is used to detect a range of commonly occurring mutations, and if negative for these mutations, single gene sequencing can be done for less common mutations, which is time consuming and can further delay treatment. NGS has become the gold-standard method to detect actionable mutations, and was recommended by the European Society for Medical Oncology (ESMO) in 2020 in advanced non-squamous NSCLC, as well as in other malignancies including cholangiocarcinoma, prostate and ovarian cancers. NGS provides a broader test platform, and can assess many more potential gene alterations. The MOSCATO trial assessed the use of high throughput genomic analyses including NGS and showed improved outcomes in advanced cancer. Specifically in NSCLC, NGS has shown significantly improved outcomes, by identifying patients who will derive the greatest benefit from targeted therapies, therefore reducing toxicity related to ineffective therapies with certain mutations.

Plasma genotyping in non-small cell lung cancer Plasma genotyping, or 'liquid biopsy,' is a relatively new technology in which circulating tumour DNA (ctDNA), shed into the bloodstream by tumours, is detectable by next generation sequencing. This technology, while associated with a slightly lower sensitivity (80%), has >96% concordance for tissue testing of EGFR, ALK, ROS1 and BRAF mutations. In addition, the median turnaround time is just 7-10 days, without the associated risks and costs of invasive biopsy. Recent studies have shown the dynamic nature of tumour cells including the development of resistance mechanisms in EGFR-mutated tumours. As such, many cancer centres, including some based in Ireland, have used liquid biopsy as a means of surveillance for these resistance mechanisms over time rather than submitting a patient to multiple repeat biopsies. More recently many centres in North America and Europe have utilised liquid biopsy for assessing genomic alterations for targeted therapies in lung cancer upfront at the time of tissue sampling, thereby reducing the time to treatment commencement. This approach is supported by several high-level papers, including the NILE study, a prospective study of 282 patients with advanced untreated NSCLC that showed concordance between tissue and plasma genotyping and a significantly faster turnaround time from 15 to 9 days. A single centre study in the USA showed that concurrent plasma and tissue testing in NSCLC increased the detection of actionable mutations by 15%, and hence increased the number of patients receiving targeted therapies.

As well as the oft needed requirement for repeat biopsy due to a lack of tumour tissue for NGS, patients with advanced NSCLC often have a poor ECOG performance status which can limit the possibility of primary invasive procedures or repeat tissue sampling, and in some cases, the location of the tumour site may not be amenable to biopsy. In the event that tissue can be retrieved by invasive procedure, storage of these samples, for example with formalin fixation, can lead to false positive results on next generation sequencing. Invasive sampling is often associated with high cost, due to the number of clinicians, and supportive staff available. Blood sampling is minimally invasive during the initial diagnosis, bypasses the time required to schedule bronchoscopy or radiologically-guided biopsies and allows immediate processing of samples. Although the use of plasma genotyping has expanded in recent years, it is likely that it would remain an additional tool alongside tissue genotyping, as tissue samples can provide morphology and elucidate primary sites of disease.

Rationale We propose this study to evaluate the feasibility of a plasma-based circulating tumour DNA mutation testing pathway using NGS and initiated at the Rapid Access Lung Cancer Clinic (RALCC) for patients with suspected NSCLC in Ireland. This proof of principle initiative aims to establish a robust patient pathway for systematic somatic mutation testing in patients with NSCLC in Ireland using plasma-based testing. Plasma will be tested for circulating tumour DNA mutations using a validated NGS-based assay, at one of two testing laboratories in Ireland. Proving feasibility through a clinical trial in Ireland is crucial to inform successful applications for authorisation of liquid biopsy to the National Clinical Cancer Programme and to inform clinical trials that aim to identify novel therapies for Irish patients. We believe an upfront plasma-based pathway would lower median turnaround time in the Irish context. Liquid biopsy testing of patients in the RALCC for tumour mutations should ensure that more patients have access to precision medical therapies. Moreover, this approach is likely to significantly improve the identification rate of patients with NSCLC actionable genomic alterations, with resultant benefits for these patients in terms of cancer treatment. This study will also report on the currently unknown frequency, characteristics, disease course, and treatment patterns of somatic mutations in an Irish population with NSCLC.

Hypothesis Upfront plasma-based tumour genotyping of patients with advanced NSCLC with the goal of guiding treatment selection is feasible to implement and will reduce the time to treatment compared with tissue genotyping.

Planned Analysis This proof of principle initiative aims to establish a robust patient pathway for systematic somatic mutation testing in patients with NSCLC in Ireland using plasma-based testing. Plasma will be tested for circulating tumour DNA mutations using a validated NGS-based assay, at one of two testing laboratories in Ireland. The Cancer Molecular Diagnostics (CMD) laboratory at St James's Hospital will perform the tumour test on using an in- house Roche AVENIO ctDNA Expanded panel (Basel, Switzerland). The Histopathology Department at Beaumont Hospital will use an in-house Roche AVENIO ctDNA Expanded panel on the Illumina MiSeq NGS platform (California, USA). Both assays will be validated on a training tumour set. In parallel to ctDNA testing, tumour biopsy tissue testing will be carried out as per standard of care. Liquid biopsy results will be reviewed by the treating clinician for treatment decision purposes however the treatment decision will remain at the discretion of the treating medical oncologist.

It is recognised that liquid and tissue biopsy both have an excellent specificity (>95%). However, liquid biopsy may have a lower sensitivity in the region of 80%. As such, there may be genomic alterations identified on tissue biopsy not identified on liquid biopsy and less likely - vice-versa. In that event, the genotyping which has reported a genomic alteration will be considered the most clinically relevant results as the likelihood of a false positive result is very low (\< 5%). However, final treatment decision will remain at the discretion at the treating medical oncologist.