Re‐re‐biopsy? How much tumor is enough in the era of precision oncology?

Precision oncology is the concept of molecular profiling of tumors to identify targetable mutations.1 This underpins the idea of personalized medicine. In order to identify such alterations, it has been necessary to sample and analyze tumor tissue obtained through biopsy. Next-generation sequencing (NGS) and other comparable methods have made it possible to quickly analyze these tumor samples for molecular targets. Molecular profiling can now also be performed on blood samples, which is termed liquid biopsy. The treatment paradigm for non-small cell lung carcinoma (NSCLC) continues to evolve, influenced heavily by precision oncology. The understanding of molecular mutations that promote oncogenesis has led to the development of several efficacious therapies that target these driver mutations. Under treatment pressure, tumors commonly develop further mutations as a mechanism of resistance. These mutations may then be targeted with other therapeutic agents. In NSCLC a range of actionable driver mutations have been identified. The most common of these is a mutation of the epidermal growth factor receptor (EGFR) gene. EGFR mutations are present in 32% of NSCLC patients, and up to around 50% in selected Asian populations.2, 3 Historically, patients with metastatic NSCLC harboring a sensitizing EGFR mutation such as an exon 19 deletion or L858R point mutation were treated with a first-generation (1G) tyrosine kinase inhibitor (TKI) such as gefitinib4 or erlotinib,5 or a second generation (2G) TKI such as afatinib.6 At the time of progression, re-biopsy of a progressive or new tumor is important to assess for a mechanism of resistance, such as secondary EGFR T790M mutation. This assumed added importance to the results of the AURA3 study. Mok et al. reported an improvement in progression-free survival (PFS) when osimertinib was used following 1G/2G EGFR-TKIs in comparison to platinum and pemetrexed chemotherapy when a T790M mutation was found.7 Osimertinib became the preferred first-line treatment paradigm, with the subsequent FLAURA trial in 2018 demonstrating its superiority over gefitinib or erlotinib, with a significantly longer PFS and overall survival, in the first-line setting.8 Osimertinib also has significant central nervous system (CNS) penetration resulting in better control of de novo CNS metastases or delaying occurrence of CNS progression. Despite this, patients in a large part of the Asia-Pacific region continue to have 1G TKI as first-line treatment, due to the lack of reimbursed access to osimertinib in the first-line setting. As such, re-biopsy looking for EGFR T790 mutation remains a widespread clinical practice to see who may benefit from second-line osimertinib.9, 10 There is sufficient data demonstrating up to 50% of patients have T790M mutation at progression on 1G EGFR TKI. 11, 12 However it is not understood whether repeat biopsy after further therapy can identify more patients with T790M mutations, who had a previously negative result. Wang et al. recently reported a repeat re-biopsy rate of 58% in NSCLC patients after 1G/2G EGFR TKI who initially tested negative for T790M and had some other therapy in the interval. They reported their findings on the efficacy and safety of osimertinib in NSCLC patients with an EGFR T790M mutation, depending on when this mutation was detected. They assessed patients who progressed on a 1G/2G EGFR TKI and compared those who received osimertinib after their first re-biopsy demonstrated T790M mutation versus those who received osimertinib after a repeat re-biopsy following another therapy.13 They demonstrated the potential benefits of re-biopsy in this population. Underpinning Wang et al.’s study was the notion that regardless of when T790M mutation was detected, there was benefit in using osimertinib. This suggests repeated or serial re-biopsy might improve outcomes for patients. Firstly the overall detection rate of T790M was improved by repeat re-biopsy. It is unclear whether this is due to the sensitivity of the testing assay or whether T790M mutations subsequently emerge from the selection pressure of further therapy. The rate of T790M mutation detected was 60% and is comparable to other real-world studies.3, 14 Notably the T790M second re-biopsy rate was 35.7%. The authors conclude that repeat re-biopsy can therefore increase the T790M mutation positivity rate. Secondly, the efficacy of osimertinib did not differ greatly regardless of when it was initiated. The objective response rate was 72.1% for re-biopsy patients and 62.5% for repeat re-biopsy. This compares well to the 71% response rate in AURA3. Median survival also did not differ significantly between the two groups, of 20 months and 19 months respectively. This was somewhat lower than AURA3's reported 26.8 months in a trial-selected population.15 Tissue biopsy at the time of disease progression is a good oncological practice and is considered the gold standard.16 Unfortunately, the practicality of obtaining tissue biopsy can be challenging. The issue of practicality is demonstrated in Wang et al.’s finding that only 41% of patients had tissue biopsy as the method of re-biopsy, while the majority (59%) of patients underwent liquid biopsy. What are these practical challenges to tissue biopsy? They include a lack of easily accessible tumor biopsy sites, repeated failures of tissue acquisition, safety concerns from physicians, and perceived invasiveness from patients. Tissue biopsy does not always provide a result. A large analysis of American lung cancer patients found that 46% of patients required a re-biopsy in their diagnostic work-up.17 Separate studies have found up to 23% of tissue lung cancer biopsies retrieve inadequate samples, and between 14% and 30% of the tissue is inadequate for molecular characterisation18, 19 Concerns around safety are reasonable given that pneumothorax can occur in around 10% of percutaneous biopsies.17 Aside from practical challenges, tissue samples also lack the opportunity to examine tumor heterogeneity. A small area of a single lesion is not able to capture the dynamic evolution of a tumor between primary and metastatic sites. In this way, liquid biopsy may be advantageous and better capture the chronological emergence and intensity of resistance mutations. There is a need to overcome the barriers and limitations of tissue biopsy outlined above, and liquid biopsy looking at circulating-tumor DNA (ctDNA) poses a solution. Liquid biopsy might be preferred to tissue biopsy in some settings. Liquid biopsy could be considered if a patient is unfit for an invasive procedure, if there is insufficient material, or if it is unsafe to access further tissue. Liquid biopsy can also be complementary to histology and can be used for serial testing to track tumor molecular changes across all sites of disease without the need for invasive physical biopsy. For many of these reasons, liquid biopsy is gaining acceptability in clinical practice. It has high specificity but carries a low sensitivity and a 30% false negative rate. The concordance rate between tissue and plasma in one study approached 67%.20 ctDNA and also is less able to detect RNA fusion oncogenes such as ALK-fusion and RET-fusion. Finally, no published or consensus guidelines exist for ctDNA testing21 in a rapidly evolving field. Cost is a complex practicality to individual patients and more broadly to health economies. The ‘cost’ of a liquid biopsy and any potential therapeutic outcome may differ greatly depending on the region, national reimbursement scheme, and the patient's socio-economic demographics. In Asia-Pacific, there is variability across important aspects of lung cancer care that have financial implications. Variable factors include access to tissue biopsy procedures; availability of liquid biopsy; associated laboratory requirements of tissue and liquid biopsy; national reimbursement programs of targeted therapies; and availability of clinical trials. A Canadian cost-effectiveness analysis in 2022 assessed the addition of liquid biopsy to tissue testing in advanced NSCLC and concluded it did not greatly increase healthcare costs, whilst leading to more patients receiving appropriate therapy.22 Given the widespread adoption of liquid biopsy remains in its infancy, and the nuances of the Asia-Pacific region in particular, the health economic impact of liquid biopsy requires further exploration.23 Wang et al. analyzed a population of patients on first-line 1G/2G TKI. This treatment paradigm still exists in developing economies in the Asia Pacific. Can their study teach us a lesson about repeat re-biopsy in regions where we have access to first-line osimertinib? We learned that the mechanism of resistance following first-line osimertinib is heterogenous. In the FLAURA study of 91 patients with liquid biopsy post progression after first-line osimertinib, mutational heterogeneity exists with the detection of MET amplification (15%), EGFR-C797S (7%), acquired HER2 amplification, PIK3CA and RAS mutations (2%–7%).14 In addition, histological transformation to squamous cells is reported as up to 15% and small cells between 3% and 10%.24 Importantly, there is emerging therapies targeting MET amplification25, HER2 amplification26 and fourth-generation EGFR TKI27 that has activity against EGFR C797S mutation. As such there may be significant benefit in repeated re-biopsies to increase the detection rates of those targetable alterations in the post-osimertinib setting. This warrants further studies. We believe that Wang et al.’s findings are important for patients treated with 1G/2G EGFR TKI. Repeated re-biopsy looking for T790M mutation may be beneficial to enable treatment with osimertinib. This concept may also apply in patients treated with first-line osimertinib, although this hypothesis needs to be explored. Further studies describing the rate of re-biopsy at each point of progressive disease as well as the molecular and clinical outcomes of each re-biopsy will be informative. What is thought-provoking is the optimal way of obtaining the re-biopsy. Liquid biopsy offers a tempting companion test to tissue, and in select cases, may be used in place of tumor tissue. However, with escalating healthcare costs globally, this may not be a panacea routinely endorsed by governments in the Asia-Pacific until there is a decline in the cost of serial testing or comparable sensitivity to tumor tissue. For now, tumor tissue at each point of progression remains the standard of care for molecular profiling. The clinical context must be considered in a cohort of vulnerable patients with incurable malignancy so as to avoid invasive procedures and time toxicity associated with healthcare interactions. With each subsequent line of therapy, a clinician must also consider a variety of factors—how likely is it that tissue might yield an actionable target, the regional availability or reimbursement of a targeted therapy, and critically a patient's performance status for suitability of further therapies. Samuel Smith and Michael Boyer declare no conflict of interest. Steven Kao is an Editorial Board member of the Asia-Pacific Journal of Clinical Oncology and a co-author of this article. To minimize bias, they were excluded from all editorial decision-making related to the acceptance of this article for publication. Served on Advisory Boards: AstraZeneca; Pfizer; Boehringer; MSD; Novartis; Roche, Takeda, Amgen Honorarium partly to my institution: MSD, BMS, Roche, AstraZeneca, Pfizer, Boehringer, Takeda Research grant to my institution: AstraZeneca

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.