Welcome to the May 2025 BTOG trainee newsletter, where we aim to provide you with updates relevant to trainees working in the thoracic oncology space.
We summarise the ‘hot-off-the-press’ landmark papers, list updates for trainee-led multicentre projects, and sign-post readers to useful educational/professional content.
Please do use the BTOG trainee group-chat on Slack to continue your thoracic cancer conversations in the meantime. Feel free to contact the BTOG trainee representatives, Anand and Gerard, at anand.sundaralingam@ndm.ox.ac.uk or g.walls@qub.ac.uk
Clinical Oncology
NARLAL2
Image source Acta Oncologica
le radiotherapy (RT) with concurrent chemotherapy and adjuvant immunotherapy is the gold standard treatment for unresectable NSCLC, disease control rates are suboptimal. In particular, the cumulative incidence of locoregional relapse is 25% with state of the art treatment, however in practice, most patients may receive an inferior radiotherapy and drug combination or sequence associated with a locoregional failure rate closer to 50%. Beyond reductions in quality of life metrics, as local recurrence often precludes further treatment through increasing frailty, there is a strong independent association between local recurrence and reduced overall survival. There is thus an urgent need to improve local control following treatment with radiotherapy (RT).
NARLAL2 is an open-label multicentre Scandinavian phase 3 RCT in patients with ECOG performance status 0–1, who have locally advanced NSCLC (i.e. stage IIB–IIIB, TNM v7). The key intervention in this study was the evaluation of a PET/CT-guided approach to dose escalation. Patients were randomised to receive either standard 66 Gy in 33 fractions with 2 cycles of concurrent platinum/vinorelbine chemotherapy (+/- 2 cycles of induction), or a dose-escalated treatment where 66 Gy was given to all the disease plus an integrated boost for PET-avid tumour regions (up to 95 Gy to the primary and up to 75 Gy to involved lymph nodes), without breaching the radiation dose tolerance of nearby organs-at-risk (OARs). Only nodes ≥4cm were eligible for PET-based dose escalation. A key feature of the study was that both the standard and dose-escalated RT plans were prepared for all patients prior to randomisation, so that the radiation doses to normal lungs in the dose-escalated plan could be manipulated to be no higher than the standard plan. Adjuvant durvalumab was introduced to the study protocol following the global adoption of the PACIFIC regime, and 20% patients received this in each arm.
In terms of radiotherapy planning, a 5mm margin was used for gross tumour volume (GTV) to clinical target volume (CTV) expansion (editing off bony/cartilaginous structures and vessels) and 4–10mm margin was then applied to create the planning target volume (PTV), as per local protocol. Planning was performed on 4DCT scans and treatment was delivered with intensity modulated RT approaches (including volumetric arc RT, VMAT). Crucially, there was a central RT quality assurance programme.
The current analysis reported the 6-month safety data for all 350 patients that were randomised between January 2015 and March 2023. The mean doses to the GTV primary and GTV nodes were 88 Gy and 71 Gy in the experimental arm respectively. No statistically significant differences were observed in early adverse events between the two arms. In escalated and standard arms, grade 2 esophagitis was 26% vs 28%, grade 3 esophagitis 4% vs 7%, grade 2 pneumonitis 20% and 16%, and grade 3 pneumonitis 6% and 4%, respectively. The patient, tumour and RT plan characteristics of the two groups were very well balanced, adding confidence that the encouraging safety profile observed for PET-guided boost is a true reflection of how robust planning approaches can lead to safe dose escalation in locally advanced NSCLC.
NARLAL2 had a pragmatic protocol design, necessitating only a baseline PET-CT, as per current standards, and not requiring mid-treatment re-planning, which would have had considerable feasibility and resource implications. The safety profile is reassuring that if the primary endpoint is met (locoregional control) in future analyses, then this treatment would be deliverable in the UK setting should it be adopted. Of note, in terms of the systemic therapy used, the induction/concurrent vinorelbine was given differently than in UK practice, being administered orally three times weekly (50mg) every week.
NARLAL2 is a promising study which suggests that dose escalation may still be a viable concept in NSCLC, after multiple negative phase II–III studies over the last decade, most notably RTOG 0617. Interestingly, significant dose escalation was achieved without significant increase in OAR doses. Furthermore NARLAL2 has robust central RT QA. Whilst the primary endpoint (locoregional control) is awaited, the schedule outlined could be an exciting new technical RT approach for patients with LA-NSCLC. Follow-on studies may involve an adaptive option as part of radiotherapy treatment plan given the previously reported impact of tumour shrinkage and the impact on OAR doses in NARLAL2. Finally, future studies in this area may not only include adjuvant immunotherapy, but perhaps may include neoadjuvant immunotherapy with chemotherapy.
Professor Gerry Hanna
Marie Curie Chair of Clinical Oncology
Trinity College Dublin, the University of Dublin
Consultant in Radiation Oncology
St Luke’s Radiation Oncology Network
Medical Oncology
HARMONi-2
Image from IASLC Lung Cancer News
There has been a rapid expansion of immunotherapy-based treatment options for advanced NSCLC, but durable benefit remains limited. Mirroring the development of combination cytotoxic regimens, several groups are investigating whether combining immunotherapy with novel means of VEGF (vascular endothelial growth factor) inhibition may enhance outcomes, by targeting two ‘hallmarks of cancer.’ Bispecific monoclonal antibodies are relatively novel constructs that enable simultaneous targeting of immune evasion and tumour vasculature cancer hallmarks, and represent a theoretically promising combination.
HARMONi-2 is a randomised, double-blind, phase III multicentre RCT performed in China to evaluate a bispecific agent targeting both programmed cell death receptor 1 (PD-1) and VEGF, known as ivonescimab. Patients who were ECOG 0–1 and have advanced PD-L1-positive (PD-L1 >1%) NSCLC not habouring an actionable genomic alteration, were randomised 1:1 to receive 3-weekly ivonescimab or pembrolizumab for two years (or until disease progression or unacceptable toxicity). The primary endpoint was progression-free survival (PFS), and events were determined by blinded independent review.
At the first planned interim analysis with a median follow-up of 9 months, among the 398 patients enrolled, the objective response rate, disease control rate and progression- free-survival favoured ivonescimab (ORR: 51% vs 39%; DCR: 91% vs 71%; mPFS: 11.1 vs 5.8 months; HR PFS 0.51 (95%CI 0.37–0.78, p<0.0001). The benefit in PFS favoured of ivonescimab after stratifying for histology, clinical stage, and PD-L1. In addition, a subgroup analysis demonstrated the benefit in PFS was similar across both PD-L1 strata (1–49; ≥50% and histology (squamous, non-squamous), in favour of the investigational arm.
However, this benefit was balanced with a higher degree of toxicity. Treatment-related grade 3+ adverse events were nearly double in patients receiving ivonescimab compared with pembrolizumab (29% vs 16%), however discontinuation rates were fewer than with pembrolizumab (2% vs 3%). There was a few additional expected toxicities noted in the ivonescimab group compared with pembrolizumab, related to the nature of VEGF inhibition, including proteinuria (any-grade 31% vs 10%) and hypertension (any-grade 16% vs 3%). Grade 3+ immune-related adverse events were similar for ivonescimab and pembrolizumab (7% vs 8%).
HARMONi-2 is the first RCT of ivonescimab in the first-line setting for NSCLC. It is also the first to demonstrate a benefit in PFS on interim analysis, compared with pembrolizumab. While thoracic oncology trials integrating bevacizumab classically have excluded patients with squamous histology owing to risk of tumour-related haemorrhage, this study demonstrated VEGF targeting could be safely achieved in this population, using this novel bi-specific agent. If the positive signal from this interim analysis of ivonescimab is borne out in the full analysis, bi-specific therapies targeting VEGF and PD-L1 may represent a promising future treatment option in advanced NSCLC.
Ivonescimab is a novel bispecific agent targeting both VEGF and PD-1, that has demonstrated encouraging results in a phase III RCT in China only, favouring ivonescimab over pembrolizumab, in the first-line PD-L1+ population. This represents an important finding, as pembrolizumab is an accepted standard of care for PDL1 >50% NSCLC worldwide. The role of this agent in the PDL1 1-49% population remains controversial, as most patients will be treated with an immunotherapy-based combination incorporating chemotherapy. While these data are encouraging, both PFS and OS results will be important to assess in the first-line setting, and will be required in order to change future practice in certain jurisdictions worldwide. The results of this trial challenge the dogma that VEGF inhibition is not suitable for patients with squamous NSCLC, and indicates that this limitation may be specific to certain agents such as bevacizumab, while new therapeutic constructs may not be associated with unacceptable toxicity in this population. Next steps in the development of this compound include appropriate global phase III trials of this agent both alone and in combination with chemotherapy, in the first-line setting, guided by PD-L1 expression. The thoracic oncology field awaits these results, with breath that is bated.
Professor Jarushka Naidoo
Professor of Medical Oncology and Consultant Medical Oncologist
Beaumont RCSI Cancer Centre, Dublin, Ireland
Lung Cancer Lead, Cancer Trials Ireland
Adjunct Professor of Oncology, Johns Hopkins University
Respiratory
SUMMIT
Image from Wikimedia Commons
Low-dose CT screening in an at-risk population has been shown to reduce lung cancer related mortality. Several pilot programmes emerged throughout England and have shown promising results and targeted lung cancer screening was recommended by the UK National Screening Committee in 2022. The pilots have informed the design of the national programme, but variation in invitation strategies, eligibility and nodule management protocols has made applicability for a population wide programme more challenging and further research was needed specifically, for a UK population.
In March’s edition of The Lancet Oncology, Bhamani and colleagues, present the findings from the 1st round of CT screening (baseline) from the SUMMIT study. This was a prospective, longitudinal cohort study of low-dose CT screening offered to individuals aged 55-77, recorded as a current smoker in their primary care record at any point in the preceding 20 years. The study sent postal invitations to residents living in north-central and north-east London, followed by a telephone-based lung-cancer risk assessment. Those deemed to be at sufficient risk (PLCOM2012 6-year risk > 1.3% or meeting 2013 USPSTF criteria) were then invited to take part in CT screening.
Participants would have a non-contrast, thin collimation low-dose CT at baseline and would then be randomised to receiving a CT vs no CT at year 1 (if the baseline scan was normal), followed by all participants going on to having a CT at year 2. Lung nodules would be monitored according to the BTS standards with slight modifications. The primary outcome measures of interest included proportion of eligible participants who underwent CT screening, incidence of screen-detected lung cancers and early vs late-stage diagnoses. The study also conducted a cell-free nucleic acid based multi cancer early detection blood test, but this will be reported on separately.
Between April 2019 and May 2021, 18,581 participants attended a lung health check. 15,269 were deemed to meet eligibility criteria for CT screening, of whom 13,035 (85%) chose to proceed and results were available for analysis in 12,733. 7353 (57.6%) were male, 10,665 (83.5%) were white and 4062 (31.8%) came from socio-economic group 1 (most-deprived). The baseline CT identified a lung cancer in 261 (2%) of participants, with 163 of these detected after the initial scan and the remaining 98 after nodule follow up scans. The number needed to screen to detect one lung cancer was 49 (95% CI 43.4-55.4). 8 participants developed an interval lung cancer within 12 months of a negative baseline scan, a risk of 0.07%. At 12-months the episode sensitivity of low-dose CT screening for detecting lung cancer was 97%, whilst the specificity was 95.2%, with a false positive rate of 4.8%. 207 (79.3%) of the detected lung cancers were stage I or II, whilst 37 (14.2%) were stage III and 16 (6.1%) were stage IV. 26 (10%) of screen detected lung cancers had synchronous primary cancers. 201 (77%) of participants with detected lung cancer were offered surgery and surgery was performed in 75.7% of those with stage III disease. In total, 241 resections were conducted in 231 participants with a benign resection rate of 11.6% (28 resections in 27 patients).
The authors conclude that a low-dose CT screening programme with a nodule management protocol based on the BTS guidelines had a high sensitivity for detecting lung cancer at 12 months, with a relatively low false-positive rate and their study reinforces the findings from other screening trials, with a stage shift in the lung cancers diagnosed with greater rates of curative intent treatment.
The four UK nations are at different stages of implementing targeted low-dose CT (LDCT) screening for lung cancer following the UK National Screening Committee’s (NSC) positive recommendation in 2022. England is in the most advanced position thanks to NHS England’s Targeted Lung Health Check (TLHC) programme, which launched prior to the UK NSC’s recommendation. The programme currently covers around a third of the eligible population in England with plans to extend to full coverage in the form of a national screening programme in the coming years. Whilst evidence of the benefit of lung cancer screening is secure, questions have remained about the most effective and efficient ways to deliver screening (O’Dowd 2023). A number of UK studies attempting to answer these questions have run in parallel to the TLHC programme, with SUMMIT being one of the largest and most influential. SUMMIT has already generated a number of informative publications on various aspects of the screening pathway, including the use of telephone-based assessment (Dickson 2022) and important evidence about when to intervene for small growing lung nodules (Creamer 2023). Results relating to a multicancer early detection blood test used in the study are awaited.
This paper adds to the body of UK-based evidence on lung cancer screening implementation. Despite disruption due to the COVID-19 pandemic, SUMMIT delivered lung cancer screening efficiently and at scale. Pragmatic evidence-based protocols for LDCT findings were key, avoiding the over-investigation of clinically irrelevant findings which has been a difficulty in previous lung screening activity. The lung cancer detection rate, stage-shift and surgical resection rate for screen-detected lung cancers were in keeping with the known benefits of screening, whilst data on potential harms was also reassuring with a relatively low benign resection, false-positive and interval cancer rate. This London-based study used static CT scanners, in contrast to many other UK lung cancer screening activities that have utilised community-based mobile scanners. There are advocates for both approaches, and it is likely that the optimal approach will vary depending on factors such as the characteristics of the target population, geography and population density.
Dr Sinan Eccles
Respiratory Consultant
Royal Glamorgan Hospital
South Wales
Pathology
IASLC Staging Project for Lung Cancer: Recommendation to introduce spread through air spaces as a histologic descriptor in the 9th edition of TNM classification of Lung Cancer
Image courtesy of Mike Sheaf (arrows indicates ‘STAS’)
“Spread through air spaces” (STAS) is a histological description where lung cancer tumour cells are identified microscopically beyond the edge of the main tumour, and in the adjacent alveolar parenchyma. This phenomenon was first termed STAS in 2015 but descriptions were noted in the medical literature as far back as 1987. In recent times it has attracted much interest as it appears to be a common finding (reported in 20-40% of surgical specimens) and appears to be associated with poor prognosis, across all histological subtypes. Therefore there have been calls to include it within the pathological staging of lung cancers.
In this article from the Journal of Thoracic Oncology by Travis and colleagues, the staging and prognostic factors committee of the international association for the study of lung cancer (IASLC) investigated whether STAS could be a useful addition to pathological staging. All cases of NSCLC, with pathological stage I N0M0R0 submitted to the global IASLC database between 2016-2019 were included. Adenocarcinoma in situ, minimally invasive adenocarcinoma, neuroendocrine tumours and those who received neoadjuvant therapy were excluded from the study.
4061 cases met the inclusion criteria and had the presence of absence of STAS included in the report. STAS was found in 930 (22.9%) of cases. STAS occurred more often in older patients (>65), in sublobar vs lobar resections and in adenocarcinoma vs squamous cell carcinoma. Recurrence free survival (RFS) at 3-years differed between patients with STAS and those without (80% vs 91%, p < 0.001), as did overall survival (OS) at 3-years (88% vs 95%, p < 0.001). A multivariate analysis to account for confounders such as age, sex, histological subtype, extent of resection and others indicated that STAS remained an independent predictor for poor prognosis (RFS: HR 1.62, 95% CI 1.28-2.06, p < 0.001, OS: HR 2.74, 95% CI 2.09-3.58, p < 0.001). Importantly STAS remained an independent predictor when compared to existing histological descriptors used in pathological staging (lymphatic/vascular invasion and visceral pleural invasion).
The authors conclude that STAS is an important histological descriptor that impacts prognosis and therefore recommend it for inclusion as a T histological descriptor for the 9th edition of the TNM classification, however, feel the data does not yet support its inclusion within the stage grouping criteria.
Optimum lung cancer patient management requires full assessment of an individual and their tumour, including accurate tumour staging. Currently the latter is generally achieved through the UICC TNM classification (9th edition). The T classifier refers to the extent of the primary tumour which relies fundamentally on location, size and invasion of local structures. Travis et al. analysed the ninth edition database and propose that spread through air spaces (STAS) should be included as an addition histologic descriptor in future TNM editions, to accompany lymphovascular invasion and visceral pleural invasion.
The concept and significance of STAS has caused a degree of controversy amongst pathologists for years. It may mean different things to different people although it is defined in the WHO blue book (Classification of Thoracic Tumours, 5th edition 2021) as ‘tumour cells within airspaces in the lung parenchyma beyond the edge of the main tumour.’ This publication suggests that it should be restricted to lung cancers, where the presence of STAS predicts both likelihood of recurrence (especially when STAS is present at the surgical margin) and worse clinical outcome. STAS has been seen in all major histological lung cancers types that have been investigated. However, STAS has to be distinguished from histological artefact – which is sometimes easier said than done. Potential mimics of STAS include jagged tissue edges, displaced tumour cells due to wet specimen or fixed tissue block handling, poor fixation and misinterpreted non-neoplastic pneumocytes or bronchial epithelial cells around a tumour. Strategies to differentiate between genuine STAS and artefact are not provided in the WHO book and each pathologist evaluating their lung cancer resection slides will need to make a subjective call on whether the STAS is genuine.
Many pathologists are convinced that STAS is a reliable, genuine and reproducible feature. On the other hand, there are pathologists who are not so sure about STAS. A recent PRO-CON debate on the subject (“Is spread through air spaces an in vivo phenomenon or an inducible artefact?”) took up many pages of the May volume of the JTO and we recommend taking time to look into these if you would like to know more about the controversy. Irrespective of whether STAS is considered genuine or artefactual and induced by either intra-operative manipulation or handling during macroscopic dissection in the laboratory, it does appear to be associated with post-operative recurrence and therefore has particular implications for sub-lobar resections.
The article by Travis et al. relies on a slightly different definition of STAS to that given above whereby, ‘lung cancer tumor cells are identified microscopically beyond the edge of the main tumor in the adjacent alveolar parenchyma’. Clearly, the same caveats apply with regard to mimics. As stated, worse recurrence-free and overall survival was found with univariate and multivariate analysis of the database. This is used to support their view that STAS should be included as a histologic descriptor for future TNM classifications, and it is hard to argue against this proposal. Interestingly, the group do not dictate exactly how STAS would fit into a new TNM classification system. This will rely on further evaluation of the updated database when developing the tenth edition of the TNM classification. It is possible that the presence of STAS will upstage the tumour (T) status (akin to visceral pleural invasion in small lung cancers) or that any changes might be applied to the residual tumour (R) status – time will tell.
It should be noted as well that others have applied the concept of STAS to metastatic tumour deposits to the lung and again found that it predicts recurrence and worse outcome as one might imagine so the definition of STAS might be widened to reflect this (see articles in the Pro-Con debate given above).
In conclusion, in our opinion it is difficult to argue against the inclusion of STAS as a histological descriptor in future TNM editions, as long as everyone is using STAS to mean the same thing.
Professor Michael Sheaff and Dr Lilian Edmunds, Consultant Cellular Pathologists, Barts Health NHS Trust.
BTOG Events
Other Thoracic Oncology Events
Discover other Thoracic Oncology conferences, study days and events from other organisations not related to BTOG.
The BTOG One-Stop Resource Hub is a comprehensive, searchable database of presentations and videos from BTOG education events (annual conferences, essential updates, study days, webinars, and podcasts), available exclusively to BTOG members.
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.