Welcome to the October 2026 BTOG trainee newsletter, where we aim to provide you with updates relevant to trainees working in the thoracic oncology space. In terms of content, the trainee newsletter will seek to summarise the ‘hot-off-the-press’ landmark papers, list updates for trainee-led multicentre projects, and sign-post readers to useful educational/professional content.
Your BTOG trainee representatives:
- Michiala Cafferkey – Respiratory
- Paolo Davide d’Arienzo – Medical Oncology
- Kishen Patel – Clinical Oncology
- Tanya Patrick – Respiratory
You can contact the BTOG Trainee Representatives by email using info@btog.org
Oncology
PEACE: Evolutionary Characterisation of Lung Cancer Metastasis
Sonya Hessey, Abigail Bunkum, Ariana Huebner, Kerstin Haase, Kristiana Grigoriadis, Cristina Naceur-Lombardelli, … Mariam Jamal-Hanjani. Nature 2026
Background
Metastasis is the leading cause of cancer-related mortality (1). Lung cancer contributes more metastatic cases than any other cancer type, a consequence of its high incidence, late-stage presentation, and substantial relapse rates following curative surgery (2). Understanding the mechanisms that enable cancer cells to migrate from the primary site to other tissues could inform strategies to prevent this process.
Although metastatic dissemination is a transient process that cannot be directly observed in patients, its history can be reconstructed retrospectively by tracing the evolutionary relationships between tumour subclones across primary and metastatic sites using DNA sequencing.
The TRACERx (TRAcking Cancer Evolution through therapy) lung study (3), which prospectively profiles multi-region primary non-small cell lung cancer (NSCLC) tumours, has been strategically paired with the national, multicentre research autopsy programme PEACE (Posthumous Evaluation of Advanced Cancer Environment NCT03004755), enabling tumour sampling from diagnosis through to death. Together, these studies provide one of the most comprehensive evolutionary maps of NSCLC metastasis generated to date.
Using 501 longitudinally collected primary and metastatic tumour samples from 24 patients with NSCLC enrolled in both the TRACERx and PEACE studies, whole-exome sequencing data were used to reconstruct tumour evolutionary histories and metastatic migration patterns from diagnosis to death. Whole-exome sequencing enables comprehensive cataloguing of somatic mutations across tumour regions, while multi-region and multi-site sampling captures the spatial heterogeneity of disease that a single biopsy would miss. The sequencing dataset encompassed approximately 70% of radiologically identified metastases detected before death, paired with multi-region sampled primary tumours. The cohort was broadly representative of patients with operable NSCLC.
Key Results
- Metastases diverge substantially from the primary tumour. Metastatic genomes show substantial divergence from their primary tumour of origin, acquiring further driver alterations and whole-genome doubling events following dissemination. Most driver alterations were still found in primary tumours, but metastases acquired additional mutations with potential biological consequences.
- Multiple primary subclones seeded metastases. In 62.5% of patients, multiple primary tumour subclones disseminated, each founding a distinct metastasis. This event was substantially under detected in prior analyses relying on limited biopsy sampling (4, 5), reinforcing the importance of comprehensive spatial sampling to understand disease biology.
- More metastases were seeded by other metastases than by the primary. Over half of the metastases sampled were seeded by other metastases. In 88% of patients, seeding subclones originated from both the primary tumour and established metastases. This metastasis-to-metastasis cascade indicates that in advanced disease, metastases are a key source of onward metastatic seeding.
- Metastases established early in the disease course are more likely to seed. Metastases detected on the first relapse scan were significantly more likely to seed further spread than those identified only at autopsy. Seeding metastases were larger, harboured more subclonal diversity, and had accrued more somatic alterations.
- Spread respects anatomical boundaries. Most metastatic migrations started and ended in the same anatomical cavity. Intrathoracic metastases (lung and mediastinal lymph nodes) preferentially seeded further intrathoracic metastases, while intrathoracic-to-extrathoracic seeding was infrequent. The small subset of subclones that breached the thoracic boundary showed widespread dissemination and a higher burden of somatic copy-number alterations, pointing to chromosomal instability as a possible facilitator of extrathoracic spread.
- Treatment-associated mutations accumulate in metastases, but actionable targets remain in the trunk. New driver mutations arose in metastases in 83% of patients, many linked to prior chemotherapy. Of clinical relevance, no actionable mutations were detected uniquely in metastasis-specific subclones. Established targetable drivers (e.g. EGFR, KRAS) were consistently detected in the primary tumour, supporting the continued relevance of primary tumour molecular profiling to guide therapy selection.
This study reveals that metastasis in NSCLC is a deeply individualised biological process, shaped by time in situ, anatomical location, and chromosomal instability. Metastatic disease is not static; it is biologically diverse, spatially organised, and continuously evolving. Appreciating this complexity is fundamental to developing more effective strategies for preventing relapse, monitoring disease progression, and ultimately improving outcomes for patients with advanced NSCLC.

Comment from the Authors
The PEACE autopsy programme, integrated with the TRACERx Lung study, created a unique opportunity to examine tumour evolution across the full disease course, from diagnosis through to death. The longitudinal, multi-site exome sequencing resource generated through co-recruitment to both studies enabled patterns of metastatic spread and clonal evolution in advanced NSCLC to be reconstructed with greater resolution than previously achieved. This work was possible because patients who participated in TRACERx altruistically consented to PEACE to help future patients with metastatic cancer.
Our findings provide a framework for understanding metastatic progression in NSCLC. The primary tumour initiates metastasis by seeding metastases in distant sites. More often than not, these metastases are seeded by distinct primary tumour cell populations, and hence are clonally distinct from one another. Once established, metastatic disease operates as a dynamic, self-propagating network that continues to evolve under diverse anatomical and therapeutic pressures.
The temporal and spatial complexity of metastatic spread and evolution highlighted in this study underscores the importance of considering the full metastatic burden in future metastasis research. For example, supplementing tissue sampling with serial plasma ctDNA analysis may support more comprehensive and dynamic characterisation of metastatic disease. Of clinical significance, the observation that metastases can serve as reservoirs for further metastatic spread suggests that metastasis-directed interventions, including surgery and local ablative therapies, may disrupt this process.
We hope this work, together with its associated publicly available data, will support further advances in this field and ultimately improve outcomes for patients with metastatic cancer
Sonya Hessey MBChB MRes PhD MRCP
NIHR Academic Clinical Lecturer
Cancer Metastasis Lab, UCL Cancer Institute
Medical Oncology Registrar
University College London Hospitals NHS Trust
Mariam Jamal-Hanjani BSc MBBS PhD FRCP
Professor of Cancer Genomics and Metastasis
Cancer Metastasis Laboratory, UCL Cancer Institute
Medical Oncology Consultant in Translational Lung Oncology
University College London Hospitals NHS Trust
References
- Dillekås H, Rogers MS, Straume O. Are 90% of deaths from cancer caused by metastases? Cancer medicine 2019;8(12):5574-76.
- Zappa C, Mousa SA. Non-small cell lung cancer: current treatment and future advances. Translational lung cancer research 2016;5(3):288.
- Jamal-Hanjani M, Wilson GA, McGranahan N, et al. Tracking the evolution of non–small-cell lung cancer. New England Journal of Medicine 2017;376(22):2109-21.
- Al Bakir M, Huebner A, Martínez-Ruiz C, et al. The evolution of non-small cell lung cancer metastases in TRACERx. Nature 2023;616(7957):534-42.
- McPherson A, Roth A, Laks E, et al. Divergent modes of clonal spread and intraperitoneal mixing in high-grade serous ovarian cancer. Nature genetics 2016;48(7):758-67.
Respiratory
TACTIC
Medical thoracoscopy with talc poudrage and indwelling pleural catheter insertion versus medical thoracoscopy with talc poudrage alone for patients with symptomatic malignant pleural effusion (TACTIC): a randomised, controlled phase 3 trial
Lung cancer accounts for 30% of malignant pleural effusions. Suspected malignant pleural effusions often require both diagnosis to identify the underlying cause and symptom control through pleural intervention. Medical thoracoscopy and pleural biopsy has better sensitivity compared to cytological analysis from pleural fluid and talc poudrage at the time of medical thoracoscopy can offer good pleurodesis rates, but requires an inpatient hospital stay. Indwelling pleural catheters offer an ambulatory option for pleural fluid drainage but no multi-centre randomised trial has considered IPC in addition to medical thoracoscopy and talc poudrage.
The authors of the TACTIC study sought to compare thoracoscopic talc poudrage plus IPC to thoracoscopic talc poudrage in a UK multi-centre, randomised controlled trial for patients with symptomatic malignant pleural effusions. Eligible participants either required talc poudrage as part of their usual care or talc poudrage was chosen by patient and operator as a preferred pleurodesis option. Primary outcomes analysed over the 4 weeks post-procedure were: 1) total length of hospital stay in days (including re-admission) and 2) breathlessness assessed by VAS dyspnoea scale. Secondary outcomes included: chest pain VAS score, days in hospital over the 12 weeks post-procedure, days until appropriate for discharge, and pleurodesis success at 4 and 12 weeks post-procedure. Analyses adjusted for WHO performance status, cancer diagnosis and baseline breathlessness where relevant. Pleurodesis was defined as absence of significant effusion (occupying 1/3 hemithorax on chest x ray or more than 2 rib spaces and 4cm depth on ultrasound) and no clinical need for further procedure. Health care resource use (number of healthcare contacts over 12-week follow-up), health-related quality of life assessed by questionnaires, and cost-effectiveness were also measured.
Talc poudrage with 3-4g sterile talc was performed in accordance with local policy and BTS guidance. Participants in the IPC insertion arm were assessed for discharge from the day of procedure according to study protocol and those without IPC were admitted for 48-72 hours. Participants were followed up for 12 weeks or until death.
124 of 254 eligible patients from 11 UK hospitals were randomised 1:1 to thoracoscopic pleurodesis either with or without IPC insertion. 58 participants received IPC and 57 received standard of care (no IPC). 46% (n=57) of participants had pleural mesothelioma, 28% (n=35) had lung cancer and 10% (n=13) had breast cancer. There was no difference in median length of stay in hospital over the 4 weeks post-procedure (IPC group: 1 day (IQR 1–3, 95% CI 1–2) vs no IPC group: 2 days (1–3, 95% CI 1–2), p=0.26). 4-week breathlessness scores were also not different (IPC group: median 14·0 mm (IQR 8·8–32·4) vs no IPC group: 19·6 mm, (8·1–38·7); p=0·26 unadjusted, p=0·77 adjusted, median difference –3·5 (95% CI –9·56–2·25)), although both groups had improved breathlessness scores compared to baseline (median decrease from baseline IPC group: 27.4mm (IQR 5-50.4) vs no IPC group: 19.3 (10.6-28.9)). A post-hoc analysis excluding 35 participants with protocol deviations (either not randomised to intended group or did not follow the trial discharge protocol) revealed a shorter length of hospital stay for the IPC group (IPC group: 1 day (IQR 1–2, 95% CI 1–2) vs non-IPC group: 2 days (2–5, 1–2); p=0·025 unadjusted, p=0·014 adjusted)), but there remained no difference in breathlessness scores.
The group without IPC had higher chest pain VAS scores on day 1, (IPC group: median 24·7 mm (IQR 5·0–42·0) vs no IPC: 36·0 mm (11·5–67·5), median difference –10·0 mm (95% CI 0 to –24·5); p=0.041), but not any other time point. Median length of hospitalisation over 12 weeks post-procedure did not differ (IPC group: 2 days (IQR 1–4); non-IPC group: 2 days (1–6); p=0·71 unadjusted, p=0·98 adjusted), nor did pleurodesis rates at 4 weeks (40% vs 40%; p=0.94) or 12 weeks (IPC group: 53%, non-IPC group: 42%; p=0.27). The IPC group had lower rates of repeat intervention by 12 weeks (IPC group: 3%, non-IPC group: 34%; p<0.0001). There were no differences in rates of healthcare contact or quality of life scores between the two groups. However, the IPC group had an additional cost per QALY gained of £178,874 (above the range of £20,000-30,000 typically considered cost-effective), and the additional initial cost of the procedure was £1406 more than standard of care (without IPC).
The authors conclude that thoracoscopic talc poudrage with IPC insertion did not exhibit different length of hospital stay, breathlessness control or chest pain compared to thoracoscopic talc poudrage without IPC insertion. However, they suggest that patients prioritising fewer hospital attendances or invasive pleural interventions may consider IPC insertion alongside thoracoscopic talc poudrage to achieve this.

Expert Commentary
Thoracoscopic talc poudrage offers definitive effusion control but requires an inpatient admission, whereas an IPC offers this through day-case insertion and ambulatory drainage. Retrospective series have suggested combining the two might offer the best of both worlds, and TACTIC (1) is the first randomised trial to test this.
In relation to the trial’s co-primary outcomes, adding an IPC to talc poudrage did not shorten hospital length of stay, nor did it improve dyspnoea compared to thoracoscopic talc poudrage alone. However, it was just as safe, was not more painful and reduced the need for further pleural procedures (3% vs 34%). Both arms improved breathlessness beyond the minimal important difference.
Pleurodesis success in the control arm at twelve weeks was 42%, substantially lower than 78% in TAPPS (2). This partly reflects population differences, but also differing definitions of pleurodesis. TACTIC required both radiological clearance and freedom from further intervention, with TAPPS requiring only the latter. A stable, asymptomatic residual effusion is a success in TAPPS but a failure in TACTIC. Time to pleurodesis in the IPC group was slower than in the rapid-pleurodesis literature (3) (4), with a median time to IPC removal of 42 days. However TACTIC was not a study of rapid pleurodesis and its data aligns closely with the talc-via-IPC arm of IPC-PLUS (5), where 51% achieved pleurodesis by day 70.
Of note, 26 of 62 control group patients were discharged within 48 hours contrary to protocol, reflecting an early-discharge approach many clinicians practise. A post-hoc per-protocol analysis excluding early discharges showed a significantly shorter stay in the IPC arm (median 1 day [IQR 1–2] vs 2 [IQR 2–5]).
Whilst the IPC was not cost-effective and places a burden on district-nursing teams, for the patient keen to avoid a hospital admission and minimise the chance of a repeat pleural procedure, an IPC alongside poudrage is an entirely reasonable option, provided that pleurodesis success with swift catheter removal is not oversold.
TACTIC provides valuable data on the combined approach of thoracoscopic talc poudrage with IPC insertion. With symptom control equivalent irrespective of the treatment arm, clinicians can convey the relative advantages and disadvantages of each option more clearly, and guide patients towards the choice that best fits their priorities.
Dr Deepan Sivakumar
Consultant Respiratory Physician
Blacktown Hospital, New South Wales, Australia
Conjont Senior Lecturer
Western Sydney University, New South Wales, Australia
References
- Dipper A, Sundaralingam A, Hedley E, et al. Medical thoracoscopy with talc poudrage and indwelling pleural catheter insertion versus medical thoracoscopy with talc poudrage alone for patients with symptomatic malignant pleural effusion (TACTIC): a randomised, controlled phase 3 trial. Lancet Respir Med. 2026;14(4):341-349.
- Bhatnagar R, Piotrowska HEG, Laskawiec-Szkonter M, et al. Effect of thoracoscopic talc poudrage vs talc slurry via chest tube on pleurodesis failure rate among patients with malignant pleural effusions: a randomized clinical trial. JAMA. 2020;323(1):60-69.
- Reddy C, Ernst A, Lamb C, Feller-Kopman D. Rapid pleurodesis for malignant pleural effusions: a pilot study. Chest. 2011;139(6):1419-1423.
- Boujaoude Z, Bartter T, Abboud M, Pratter M, Abouzgheib W. Pleuroscopic pleurodesis combined with tunneled pleural catheter for management of malignant pleural effusion: a prospective observational study. J Bronchology Interv Pulmonol. 2015;22(3):237-243.
- Bhatnagar R, Keenan EK, Morley AJ, et al. Outpatient talc administration by indwelling pleural catheter for malignant effusion. N Engl J Med. 2018;378(14):1313-1322.
Medical Oncology
RMC-6236-001 lung cohort: phase I-II trial of daraxonrasib for pretreated RAS-mutant non-small-cell lung cancer
Approximately 30% of non-small-cell lung carcinomas (NSCLC) harbour gain-of-function point mutations in the KRAS oncogene. In NSCLC, mutations causing amino acid substitutions at residue 12 (referred to as ‘KRASG12X’ hereafter) account for ~80% of cases, while mutations at codons 13 and 61 each represent about 7% of KRAS-mutant NSCLC (1). Among KRASG12X mutations, KRASG12C is the most common, making up ~40% of KRAS-mutant NSCLC, and it is also the only subtype for which a first generation of inhibitors (sotorasib and adagrasib) have reached regulatory approval in multiple countries (2).
Daraxonrasib (RMC-6236) is a molecular glue compound that sequesters both mutant and wild-type RAS proteins (KRAS, NRAS, HRAS) with the abundant cytoplasmic protein cyclophilin A, therefore preventing their interaction with downstream effectors essential for RAS signalling pathways (3). Because it targets the active, GTP-bound form of mutant and wildtype RAS proteins ((ON) state), daraxonrasib is classified as a multi-selective RAS (ON) inhibitor.
Arbour and colleagues report here the NSCLC cohort of the phase I-II RMC-6236-001 study. This trial was published eight days after daraxonrasib received FDA approval for pretreated KRAS-mutant pancreatic carcinoma, based on findings from the phase III RASolute 302 trial (4). Patients with pretreated NSCLC harbouring mutations in RAS oncogenes, as determined by local testing, were enrolled; key exclusion criteria were the KRASG12C mutation, any prior RAS-targeting therapy, thoracic radiotherapy within two months of recruitment, interstitial lung disease, and untreated brain metastases. The study included dose-escalation (10 mg to 400 mg once daily) and dose-expansion phases.
The study enrolled 137 patients (median age 66, 52% male, 75% white). 136/137 received doses of up to 300 mg once daily (safety population). 99% had received prior platinum-based chemotherapy and anti-PD-(L)1 therapy, 20% had received docetaxel, and 29% had previously treated CNS metastases. The predominant mutational subtypes were KRASG12V (37%) and KRASG12D (30%), while 14% harboured RASnon-G12X mutations. Overall, this distribution appears consistent with that of KRAS-mutant NSCLC, accounting for the exclusion of KRASG12C.
Patients who received between 160 and 220 mg (59/136 patients) were grouped together for further analyses based on similar steady-state exposure (AUC 3350-3500). Grade ≥3 treatment-emergent adverse events (TEAEs) and treatment-related adverse events (TRAEs) occurred in 54% and 30% of patients, respectively. The incidence of grade ≥3 TRAEs rose with dose: 12% for ≤120 mg, 25% for 160–220 mg, and 45% for 300 mg. The most frequently reported adverse events included rash (90%), diarrhoea (73%), nausea (62%), vomiting (54%), stomatitis (40%), paronychia (27%), and AST rise (21%). One case of grade 4 drug-related pneumonitis, associated with concurrent Pneumocystis jirovecii pneumonia, was reported. The adverse event rates appear similar to those observed in the daraxonrasib arm (300 mg daily) of the phase III RASolute 302 trial, which investigated daraxonrasib vs chemotherapy in pretreated metastatic pancreatic carcinoma.
TEAEs led to dose interruption in 70% of patients, dose reduction in 35%, and treatment discontinuation in 11%. Patients receiving 160-220 mg daily experienced fewer dose interruptions and modifications compared to those receiving 300 mg daily. This corresponded to a mean dose intensity of 88% for the 160-220 mg group versus 75% for the 300 mg group.
Confirmed objective response rates (ORR) were 31% for the ≤120 mg group, 34% for the 160–220 mg group, and 37% for 300 mg, with disease control rates ranging from 80 to 85% across groups. Responders had tumours harbouring KRASG12D, KRASG12V, KRASG12A, KRASG13R, KRASQ61H; this represents the first clinical evidence in lung cancer that broad-spectrum RAS(ON) inhibition translates in tumour responses beyond KRASG12X mutants. The investigators explored whether co-mutational status affects anti-cancer activity, reporting response rates by TP53, STK11, and KEAP1 co-mutational status for the 160-220 mg dose group. Response rates were similar in patients with and without co-mutations in TP53 and STK11; only 4 tumours harboured a KEAP1 mutation (response observed in 1/4).
Work published as part of daraxonrasib preclinical activity profiling supports the notion that maintaining sustained suppression of oncogenic RAS signalling is paramount to achieve tumour regression (3). Compared with 300 mg daily, the 160-220 mg dose range achieved similar response rates with numerically lower rates of TRAEs and better dose intensity delivery in this population; 200 mg was therefore selected as the recommended dose for phase III trials in advanced NSCLC.
Additional post-hoc efficacy analyses were reported for the subgroup of platinum- and anti-PD-(L)1 pretreated, docetaxel-naïve patients who received daraxonrasib at doses of 160-220 mg daily (38 out of 136 patients). This subgroup mirrors the eligibility criteria of the ongoing phase III RASolve 301 trial, which is evaluating daraxonrasib against docetaxel in pretreated NSCLC. However, removing a large proportion of patients from the phase I/II cohort may impact the generalisability of these findings. In this subgroup, confirmed ORR was 42%, disease control rate 89%, median duration of response 11.5 months, median progression-free survival (mPFS) 8.3 months, and median overall survival (mOS) 16.0 months. Based on preclinical work suggesting less potent anti-cancer activity of daraxonrasib in RASnon-G12X mutant cancers, a further subgroup analysis focussed on RASG12X mutant tumours only (33 out of the previously mentioned 38), showing identical ORR, mPFS, mOS; outcomes for RASnon-G12X mutant cancers are not reported.
In summary, this study reported safety and anti-cancer activity for multi-selective RAS(ON) inhibition with daraxonrasib in a cohort of patients with pretreated RASnon-G12C mutant NSCLC for whom no targeted therapy options currently exist. Despite treatment-related toxicity not being inconsiderable, tolerability at the selected phase III dose compares reasonably with docetaxel in contemporaneous cohorts of patients with KRAS-mutant NSCLC: grade ≥3 TRAEs occurred in 25% of patients treated with daraxonrasib 160-220 mg daily against 28% and 46% in the docetaxel arms of CodeBreaK 200 and KRYSTAL-12, respectively (5,6). It should also be noted that phase III trials of daraxonrasib have incorporated prophylaxis of skin and GI adverse events to reduce the on-target toxicity burden from continuous inhibition of wild-type RAS proteins in normal tissues.
Whether KRAS mutant-selective inhibitors may provide similar or improved anti-cancer activity with a better toxicity profile relative to a multi-selective RAS inhibitor remains an open research question to date. For example, the KRASG12D (ON) selective inhibitor zoldonrasib demonstrated an ORR of 52% and a mPFS of 11.1 months in an a similarly designed phase I pretreated NSCLC cohort, with substantially lower rates of treatment-related skin and GI toxicity (rash 18% vs. 90%; diarrhoea in 30% vs 73%) (7) than in the phase I/II study of daraxonrasib. Ongoing studies will determine whether daraxonrasib may serve as a universal treatment strategy for KRAS-mutant NSCLC, or whether mutant-selective inhibitors will be preferred for the most prevalent mutational subtypes, with daraxonrasib reserved for rare or difficult-to-treat mutants.

Expert Commentary
Before visiting the detail of this study, it is worth first stepping back to remind everyone that daraxonrasib represents another momentous step forward for (K)RAS targeting. While allele-specific KRASG12C inhibitors trailblazed in this ‘undruggable’ space after ~40 years of effort, the development of pan-(K)RAS inhibitors was considered an extremely high risk endeavour – owing to decades of preclinical evidence that has demonstrated the vital physiological role of RAS in normal tissue, such as embryonic lethality that occurs when the KRAS gene is genetically deleted from mice. There are also no known cases of humans with germline homozygous deletion of KRAS, suggesting they don’t survive in utero. Crossing this ‘red line’ for drug development, it is therefore surprising that we don’t observe more toxicity with daraxonrasib. A 25% rate of grade 3 treatment-related toxicities (no grade 4 or 5 TRAEs) is something that we would be able to manage in the clinic, although we await results from the much larger RASolve-301 phase III population before making any definitive conclusions.
From here, I believe we can project very optimistically: (i) the example of other molecular subsets of NSCLC, such as EGFR and ALK, suggests that daraxonrasib will just be the first step towards a much higher quality of pan-(K)RAS drugs in 5-10 years’ time; and (ii) it can’t be overstated that effective targeting of KRASnon-G12C mutations with daraxonrasib is an incredible advance – underlined by the near doubling of overall survival recently observed in patients with pretreated advanced pancreatic cancer (RASolute 302).
There are at least two very interesting clinical signals in this NSCLC report that are worth highlighting:
- There is little drop-off in activity of daraxonrasib at its lower doses. This is an important pattern that we have also observed with early phase trials of KRASG12C As clinicians, this knowledge can influence our inclination and confidence to dose reduce. It can also assist our projections on feasibility of lung cancer trial designs (with and without combination), given that a KRAS-mutant NSCLC population will always be affected by more frailty and co-morbidity than patient with cancers harbouring other molecular or histological subtypes. In this context, a recommended phase II-III dose of 200mg once daily appears a pragmatic choice.
- Figures on duration of response, median PFS and median OS suggest that daraxonrasib offers very encouraging activity for a first-time drug approach, easily surpassing what was observed at the same stage for first generation KRASG12C Given its distinct mechanism of action to KRASG12C inhibitors such as divarasib and olomorasib, I do not view that the best clinical application of daraxonrasib will always be in direct competition with these drugs. Our lab and others have previously shown the importance of wildtype RAS signalling (targeted by daraxonrasib) as a key mechanism of resistance to allele-specific inhibition, a finding that was validated in patient-derived xenograft (PDX) models by demonstrating clear combination benefits of the two KRAS drug approaches (https://doi.org/10.1158/2159-8290.CD-25-1616). If toxicity permits (and that is currently a big ‘if’), then dual pan-(K)RAS + allele-specific KRAS inhibition will become increasingly relevant to explore in the pre-treated, and perhaps first-line, advanced patient setting.
To finish, it should be noted that NSCLC patients affected by KRAS mutations can often do very well with standard-of care (chemo) immunotherapy, which remains our peri-operative and first-line advanced treatment in this molecular setting. The future of daraxonrasib, and all (K)RAS inhibitors, is therefore going to be shaped by their potential to safely combine with PD-(L)1 inhibitors such as pembrolizumab and durvalumab. Ongoing trials such as SUNRAY-01, SUNRAY-02, and KRYSTAL-7 will begin to definitively answer this question in the next 1-2 years.
Prof Colin R. Lindsay
Group Leader, RAS Lab, Division of Cancer Sciences, The University of Manchester
Consultant Medical Oncologist, Lung Unit, The Christie NHS Foundation Trust
References
- Judd J, Karim NA, Khan H, et al. Characterization of KRAS mutation subtypes in non-small cell lung cancer. Molecular Cancer Therapeutics 2021;20(12):2577-84. doi: 10.1158/1535-7163.MCT-21-0201
- Frisch A, Martin E, Kim SY, et al. KRAS mutated NSCLC: past, present, and future directions in a rapidly evolving landscape. Oncologist 2025;30(6) doi: 10.1093/oncolo/oyaf153
- Jiang J, Jiang L, Maldonato BJ, et al. Translational and Therapeutic Evaluation of RAS-GTP Inhibition by RMC-6236 in RAS-Driven Cancers. Cancer Discov 2024;14(6):994-1017. doi: 10.1158/2159-8290.CD-24-0027
- O’Reilly EM, Wainberg ZA, Hendifar AE, et al. Daraxonrasib or Chemotherapy in Previously Treated Metastatic Pancreatic Cancer. N Engl J Med 2026;395(4):325-37. doi: 10.1056/NEJMoa2605555 [published Online First: 20260531]
- de Langen AJ, Johnson ML, Mazieres J, et al. Sotorasib versus docetaxel for previously treated non-small-cell lung cancer with KRASG12C mutation: a randomised, open-label, phase 3 trial. The Lancet 2023;401(10378):733-46. doi: 10.1016/S0140-6736(23)00221-0
- Barlesi F, Yao W, Duruisseaux M, et al. Adagrasib versus docetaxel in KRAS(G12C)-mutated non-small-cell lung cancer (KRYSTAL-12): a randomised, open-label, phase 3 trial. Lancet 2025;406(10503):615-26. doi: 10.1016/S0140-6736(25)00866-9
- Riess J, Haura EB, Yaeger R, et al. Abstract CT021: Preliminary safety and clinical activity of zoldonrasib (RMC-9805), an oral, RAS(ON) G12D-selective, tri-complex inhibitor in patients with previously treated KRAS G12D non-small cell lung cancer (NSCLC). Cancer Research 2026;86(8_Supplement):CT021-CT21. doi: 10.1158/1538-7445.AM2026-CT021
Thoracic Surgery
Consensus Definition of Stage III Non-Small-Cell Lung Cancer (NSCLC): Technical Resectability to Standardise Inclusion Criteria for Clinical Trials — EORTC-LCG Collaboration
Approximately 30% of all patients diagnosed with non-small-cell lung cancer (NSCLC) present with locally advanced disease – stage III (1,2). Stage III NSCLC represents a heterogeneous group of disease based on the TNM staging which reflects prognosis (3). This diversity has long posed a major challenge in designing and interpreting clinical trials, often leading to inconsistencies in patient selection and trial outcomes. In response, the European Organisation for Research and Treatment of Cancer (EORTC) Lung Cancer Group, together with several international thoracic societies, convened a multidisciplinary panel of experts to create a consensus definition of “technical resectability” for stage III NSCLC. The goal is to standardise inclusion criteria for prospective trials and guide clinical decision-making.
Previous studies and trials have inconsistently defined what constitutes “technically resectable” disease, particularly regarding nodal stations (especially N2 disease), tumour invasion of mediastinal structures, and chest wall involvement. This has resulted in poor comparability between trials, limited external validity, and difficulties translating findings into clinical practice. The EORTC-LCG initiative therefore sought to provide a shared framework.
This collaborative effort used a modified Delphi process involving 36 experts including thoracic surgeons, medical oncologists, radiation oncologists, radiologists, and pathologists from across Europe. Structured consensus rounds were used to collate expert opinion and achieve agreement on definitions, supported by a comprehensive literature review and detailed consideration of anatomical boundaries and surgical techniques (4-6). Consensus was defined by a 75% agreement. There was a unanimous consensus that decisions around resectability should be based on PET–CT, brain MRI, and invasive mediastinal staging and discussed with experienced thoracic surgeons within the broader MDT.
The panel focused on defining technical resectability — the physical possibility of achieving a complete (R0) resection which has been defined previously (7). Key components include:
Primary Tumour (T descriptors):
-
- T3 lesions are resectable if there is no involvement of vital mediastinal structures.
- T4 (size or satellite) tumours are resectable.
- T4 (invasive) tumours limited to resectable structures (e.g. carina, great vessels, recurrent laryngeal nerve, vertebral body) may be considered potentially resectable in specialised centres in highly selected cases.
- Diffuse mediastinal invasion or encasement of major vessels are unresectable.
Nodal Disease (N descriptors):
-
- Single-station N2 disease are resectable.
- Multistation N2 that is not bulky or invasive are potentially resectable.
- Bulky N2 (defined as >2.5-3cm in short axis) disease tends to be unresectable.
- N3 disease (contralateral mediastinal or supraclavicular nodes) are unresectable.
The panel clarified boundaries of resectability in the context of mediastinal involvement but particularly emphasized the role of reconstructive surgery in expert centres. The group also recommends that future trials explicitly document anatomical details, staging methods, and multidisciplinary discussions used to determine resectability. Investigators are urged to stratify stage III subgroups (e.g. T3N1 vs. T4N2) and report whether surgical decisions were made after induction therapy or at initial presentation. The consensus proposes a structured decision algorithm that divides stage III NSCLC into technically resectable disease – where an R0 resection is feasible, potentially resectable disease – where surgery may be possible after induction treatment or in select patients, and unresectable disease.
This EORTC-LCG consensus is an important step towards standardising international clinical research on locally advanced NSCLC. By distinguishing “technical resectability” from broader clinical operability, the framework promotes consistency across treatment centres and strengthens the comparability of trial outcomes. It also helps multidisciplinary collaboration— enabling more coherent integration of surgery into multimodal treatment strategies involving chemotherapy, radiation and immunotherapy. Further data collection, especially on the nature of N2 disease, is needed to refine the definition.

Expert Commentary
The management of stage III NSCLC is undergoing a profound transformation. The emergence of neoadjuvant and perioperative immunotherapy, targeted therapies, and increasingly sophisticated multimodality treatment strategies has challenged many traditional concepts that have historically guided treatment selection. Among these, the definition of resectability remains one of the most important—and one of the most debated.
The recent EORTC consensus on technical resectability is therefore both timely and highly relevant. By bringing together experts from different disciplines, this initiative provides a much-needed framework to standardise how resectability is defined, particularly in the context of clinical trials. This is critical, as inconsistent definitions have historically limited comparisons across studies and complicated interpretation of outcomes in locally advanced disease.
One of the greatest strengths of this work is its recognition that resectability cannot be viewed as a purely surgical concept. Accurate pre-treatment staging, including PET-CT, brain MRI, and invasive mediastinal assessment when appropriate, is fundamental to any meaningful evaluation of resectability. Equally important, technical resectability alone does not determine the optimal treatment strategy. The final decision must also incorporate medical operability, considering patient fitness, comorbidities, anticipated functional outcomes, and patient preferences. Ultimately, the question is not only whether a tumour can be removed, but whether surgery offers the best overall oncological outcome for that individual patient.
The consensus also reinforces the importance of expertise. This is particularly relevant for patients with complex T4 tumours requiring extended resections and for those presenting with borderline scenarios such as bulky or multistation N2 disease. In these settings, treatment decisions should be made within experienced multidisciplinary teams and, whenever possible, in high-volume specialist centres with expertise in advanced thoracic oncology surgery and multimodality care. Access to expert surgical assessment—and, when appropriate, a second opinion—may substantially influence treatment options, trial eligibility, and ultimately patient outcomes. Furthermore, treatment recommendations are often shaped by country-specific regulatory approvals, reimbursement policies, and access to novel therapies, highlighting the value of a common international framework.
Finally, important questions remain, particularly regarding the definition of bulky N2 disease. As modern systemic therapies achieve increasingly impressive rates of nodal downstaging and pathological response, future research may need to reconsider whether traditional size-based anatomical definitions remain sufficient. Greater granularity in characterising nodal burden, biological behaviour, and treatment response may be required to refine patient selection and ensure that potentially beneficial multimodality strategies are not inadvertently excluded in the era of precision oncology and immunotherapy.
Dr Cecilia Pompili MD PhD FACS FEHA
Associate Professor in Psychosocial Oncology and Thoracic Surgeon
University of Hull, UK
BTOG Steering Committee Member
References
- Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin 2022; 72: 7–33.
- Chansky K, Detterbeck FC, Nicholson AG, et al. The IASLC Lung Cancer Staging Project: external validation of the revision of the TNM stage groupings in the eighth edition of the TNM Classification of Lung Cancer. J Thorac Oncol 2017; 12: 1109–21
- Rami-Porta R, Nishimura KK, Giroux DJ, et al. The International Association for the Study of Lung Cancer Lung Cancer Staging project: proposals for revision of the TNM stage groups in the forthcoming (ninth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2024; 19: 1007–27.
- Xenophontos E, Giaj Levra N, Durieux V, et al. Definition of resectable stage III non-small cell lung cancer: a systematic review from EORTC lung cancer group. Lung Cancer 2025; 207: 108671.
- Houda I, Bahce I, Dickhoff C, et al. An international and multidisciplinary EORTC survey on resectability of stage IIInon-small cell lung cancer. Lung Cancer 2025; 199: 108061.
- Brandão M, Prisciandaro E, Xenophontos E, et al. Definition of resectable stage III non-small cell lung cancer (NSCLC) for inclusion in clinical trials: a clinical case review by a pan-European multidisciplinary expert panel led by the EORTC Lung Cancer Group. Lung Cancer 2025; 209: 108798
- Rami-Porta R, Wittekind C, Goldstraw P, et al. Complete resection in lung cancer surgery: proposed definition. Lung Cancer 2005; 49: 25–33.
BTOG Events
Thoracic Oncology Conferences, Study Days and Other Events - please note that this list includes events from other organisations not related to BTOG - inclusion on this list does not suggest endorsement by BTOG.

Cancer Research UK Lung Cancer Centre of Excellence Conference
4th-6th November 2026
Manchester, UK

BIR Annual Congress 2026
12th-13th November 2026
Birmingham, UK

BTS Winter Meeting 2026
25th-27th November 2026
London, UK

ETOP Annual Meeting 2026
27th-28th November 2026
Zurich, Switzerland

ETOP Residential Workshop 2026
27th-28th November 2026
Zurich, Switzerland
2027

ASCO Annual Meeting 2027
4th- 8th June 2027
Chicago, IL
Also included are Useful Core Resources: educational resources, guidelines, and other materials relevant to UK thoracic oncology professionals from organisations worldwide.
You can choose a speciality and then filter by tumour type, theme, and sub-category if you wish.
You can also search text in titles and descriptions of all our resources. You can search by keyword(s) e.g. ‘lung cancer screening’, or presenter surname, e.g. Popat.
Subscribe
The BTOG Trainee eNews is currently sent out quarterly to trainees. If you don't receive it directly, please subscribe below.
Unsubscribe
The BTOG Trainee eNews is currently sent out quarterly to trainees. Select the box below and click 'Submit' to stop receiving the email.
