Document resource
Introduction The concept of oligometastatic disease has reshaped the management paradigm for many solid tumours, suggesting that some patients with limited metastatic burden may achieve long-term disease control or even cure with aggressive local therapy. In non-small cell lung cancer (NSCLC), advances in imaging, radiotherapy (RT) and systemic therapy have converged to make stereotactic ablative radiotherapy (SABR) a feasible and increasingly used option for these patients. Yet, as systemic treatments evolve rapidly—most notably targeted therapies for oncogene-driven disease and immunotherapy for driver-negative tumours—the role and timing of SABR have become more complex and less uniform across molecular subtypes.Early multihistology studies1 2 suggested a survival benefit from ablating limited metastatic disease, prompting enthusiasm for local consolidative therapy (LCT) in metastatic NSCLC. However, emerging data now indicate that NSCLC with actionable oncogenic drivers, particularly EGFR mutations, behaves differently from molecularly unselected disease. Oncogene-driven cancers may remain controlled by targeted therapies for prolonged periods, creating distinct windows where LCT could further improve outcomes. In contrast, driver-negative, immunotherapy-treated and/or chemotherapy-treated patients exhibit more heterogeneous and often diffuse metastatic behaviour, with questions around the value of routine SABR in this setting.At the same time, refinements in central nervous system (CNS)-active systemic therapies have altered how clinicians manage brain oligometastases. Emerging efforts to define biomarkers, radiomic signatures and genomic predictors may guide the future selection of patients most likely to benefit from metastasis-directed therapy.In this narrative review, we examine the evolving role of SABR in oligometastatic NSCLC, with a particular focus on the growing divergence between EGFR-mutated and driver-negative disease. We synthesise the growing evidence and highlight ongoing trials that may redefine practice. Our goal is to provide a clinically practical framework for integrating SABR into an era of increasingly personalised systemic therapy.Methods for narrative review Methods A structured search strategy was undertaken to identify published and ongoing clinical trials evaluating local consolidative or locally ablative therapy in patients with oligometastatic NSCLC.Trial registry search The primary search for ongoing trials was conducted using the ClinicalTrials.gov database. Search terms were selected to capture three core concepts: (1) NSCLC, (2) oligometastatic or limited metastatic disease and (3) definitive local therapies. These included stereotactic body radiotherapy (SBRT), SABR, surgery and other local ablative techniques. The search was restricted to interventional clinical trials to ensure inclusion of prospectively designed studies. To focus on the metastatic setting relevant to oligometastatic disease, trials referring to early-stage (stage I–II) or locally advanced (stage III) NSCLC were excluded. Recruitment status filters were applied to identify trials that were recruiting, active but not recruiting, not yet recruiting or enrolling by invitation, allowing capture of ongoing and imminently relevant studies. Downloaded trial records were screened at the level of trial title, brief summary, eligibility criteria and intervention description. Trials were excluded if local therapy was optional, incidental or delivered exclusively with palliative intent, or if the study population did not clearly include oligometastatic or limited metastatic disease. As registry entries may lag peer-reviewed reporting, trial status and outcomes were cross-checked in PubMed and major conference proceedings (American Society of Clinical Oncology/European Society of Medical Oncology (ESMO)/American Society for Radiation Oncology/International Associated for the study of Lung Cancer (IASLC) when available.Literature search for published trials A literature search was performed using PubMed to identify trials with published results relevant to LCT in oligometastatic NSCLC. The PubMed search strategy combined controlled vocabulary and free-text terms including ‘oligometastatic’, ‘non-small cell lung cancer’, ‘local consolidative therapy’, ‘local ablative therapy’, ‘stereotactic radiotherapy’, ‘SBRT’ and ‘SABR’. Searches were focused on studies published within the preceding 5 years to capture contemporary evidence generated in the era of immunotherapy and modern targeted therapies. Reference lists of key randomised trials and recent narrative reviews were manually screened to identify additional relevant studies not captured through database searching. This manual citation review was particularly used to ensure inclusion of influential trials conducted before the widespread adoption of the term ‘oligometastatic’ in trial titles or abstracts, as well as studies reporting mature survival outcomes from earlier trial cohorts.Study selection and data categorisation All trials identified through the registry and literature searches were reviewed for eligibility based on disease stage, study design and the mandated use of definitive local therapy. Eligible trials were categorised according to systemic therapy backbone (chemotherapy, immunotherapy, targeted therapy or combinations thereof), local therapy modality (RT, surgery, ablation or multimodality approaches) and clinical disease context, including de novo oligometastatic disease, oligopersistent disease following systemic therapy and oligoprogressive disease.The oligometastatic state in NSCLC Oligometastatic NSCLC is best viewed as a pragmatic concept describing a clinical state of limited visible metastatic burden, rather than a distinct ‘transitional’ biology on a fixed path to widespread dissemination. The hypothesis that some patients with a small number of metastases might achieve long-term control (or cure) with aggressive local therapy predates modern imaging and is often traced back to surgical oncology principles described as early as the 1930s; contemporary formulations, popularised by Hellman and Weichselbaum, 3 renewed interest in this concept. Importantly, many patients labelled ‘oligometastatic’ likely harbour additional microscopic metastases below current detection thresholds, meaning definitions depend heavily on imaging sensitivity and technical treatability. Guideline thresholds therefore vary: ESMO4 generally uses up to five lesions provided all sites are amenable to radical local therapy; National Comprehensive Cancer Network5 similarly cites three to five metastases with emphasis on curative-intent feasibility; and European Organisation for Research and Treatment of Cancer/IASLC6 propose up to five metastases in up to three organs (excluding mediastinal nodes), explicitly requiring that every lesion can be safely ablated. These differences reflect ongoing uncertainty about whether oligometastases should be defined primarily by lesion count and distribution, or by underlying biology that we cannot yet reliably measure.Synchronous oligometastatic disease describes limited metastases present at diagnosis, whereas metachronous oligometastatic relapse (oligorecurrence) occurs after prior curative-intent therapy7 and is often associated with better outcomes. An individual patient data meta-analysis by Ashworth et al showed that timing of metastatic presentation (metachronous vs synchronous), alongside nodal status, were major predictors of survival after comprehensive treatment, supporting the clinical relevance of this distinction.8 As systemic therapies have improved, newer terms such as oligopersistent/oligoresidual (few persisting lesions after response) and oligoprogressive (few escaping sites on an otherwise effective regimen, but potentially with numerous metastases overall) have emerged, but proliferating terminology can blur clinical decision-making when it does not map to a clear management pathway. A pragmatic approach is to keep the framework centred on whether limited sites represent potentially curable disease amenable to comprehensive ablation, versus focal resistance where local therapy is used to maintain an effective systemic agent.The lack of harmonisation across definitions similarly complicates the interpretation of clinical evidence. Trials frequently enrol heterogeneous populations (combination of synchronous, metachronous, oligopersistent and oligoprogressive presentations) making direct comparisons difficult and limiting the generalisability of reported outcomes. Imaging advances add another layer of complexity: positron emission tomography (PET)/CT and brain MRI can detect small metastases that previously went unnoticed, sometimes upstaging patients who would have formerly been considered oligometastatic, while conversely allowing more accurate identification of truly limited disease in others. Numerical lesion thresholds are convenient for trial eligibility but are a blunt surrogate for biology. Outcomes after LCT are strongly influenced by metastatic tempo/velocity9—the pace at which new sites appear—alongside genomic context and systemic therapy effectiveness. Practically, oligometastatic NSCLC is best treated as a spectrum: decisions should integrate burden and treatability with molecular profile, response kinetics and an estimate of future trajectory, to identify patients most likely to benefit from consolidative local therapy.Local consolidative therapy in EGFR-Mutated NSCLC EGFR mutations represent one of the most common actionable molecular alterations in advanced NSCLC, occurring in approximately 10–15% of patients in Western populations and up to 30–50% of patients of East Asian ancestry, 10 and are strongly linked to air pollution.11 The widespread use of EGFR tyrosine kinase inhibitors (TKIs) has fundamentally altered the clinical course of EGFR-mutant NSCLC, with many patients now experiencing prolonged disease control and survival measured in years rather than months. As survival with EGFR TKIs improves, progression commonly emerges as limited-site failure rather than diffuse dissemination; in contemporary osimertinib series, most progressions are oligoprogressive (70–80%),12 13 prompting interest in local consolidative approaches such as SABR to attempt to reduce tumour burden, delay resistance-driven progression and extend the benefit of effective targeted therapy.Despite the biological rationale for integrating LCT with EGFR-targeted therapy in oligometastatic NSCLC, prospective randomised evidence has only recently matured (table 1), with two phase III trials showing possible benefits in progression-free survival (PFS) and overall survival (OS), although in different settings. In SINDAS,14 upfront SABR to all sites added to first-generation EGFR TKIs improved both PFS and OS compared with TKI alone (median OS 25.5 vs 17.4 months), although key limitations included exclusion of brain metastases. Similarly, Sun et al15 reported that adding concurrent thoracic RT to EGFR TKIs improved survival outcomes versus TKI alone in ‘oligo-organ’ metastatic disease, extending median PFS from 10.6 to 17.1 months and median OS from 26.2 to 34.4 months, although in a broad population that included many patients above classic oligometastatic lesion-count thresholds.Table 1Summary of selected published trialsTrialSetting/biologySystemic backboneLocal therapySample size (randomised)Effect size (HRs)Key limitationsOligometastatic—synchronousSINDAS (14Wang XS, JNCI 2023; phase III)EGFR-mutated synchronous oligometastatic NSCLC (≤5 mets; ≤2/organ; no brain mets)First-generation EGFR TKI (gefitinib/erlotinib/icotinib)Upfront RT/SABR to all disease (incl primary+involved nodes)133PFS HR 0.22 (95% CI 0.17 to 0.46); OS HR 0.44 (0.28–0.68)First-gen TKI era, excludes brain mets; trial stopped early after interim efficacy signal.Local treatment of primaryNROG-002 (15Sun H, JCO 2025; phase III)EGFR-mutated ‘oligo-organ’ metastatic NSCLCFirst-line EGFR TKI (Icotinib)Upfront thoracic RT (60 Gy to primary+positive nodes); RT to other mets at clinician discretion118PFS HR 0.57 (p=0.004); OS HR 0.62 (p=0.029)Higher severe TRAEs with RT+TKI (11.9% vs 5.1%)Oligometastatic—oligoresidual (postinduction)16Peng P, Radiother Oncol 2023 (phase II)Stage IV EGFR-sensitive, stable or partial response after initial EGFR-TKI 3 moFirst-generation EGFR TKI (gefitinib/erlotinib/icotinib)SBRT 30–50 Gy/5 fractions to protocol-defined sites62PFS HR 0.52 (0.31–0.89); OS HR 0.53 (0.30–0.95)LCT for responders, closed early due to slow accrualNorthStar (Elamin, ESMO 2025; phase II randomised)EGFR-mutant locally advanced/metastatic, non-progressing after osimertinib inductionosimertinibLCT (surgery and/or RT); ‘comprehensive’ versus ‘partial’ reported119PFS HR 0.66 (p=0.025); OS not mature (per report)Mixed biological populations; long-term toxicity and OS results pendingNRG-LU002 (25Iyengar, ASCO 2024; phase II/III)Limited metastatic (≤3 extracranial lesions) after initial first-line therapy; predominantly IO-era populationMaintenance systemic therapy (most had IO-containing induction)LCT (RT and/or surgery) + maintenance215PFS HR 0.93 (0.66–1.31); OS HR 1.05 (0.70–1.56)Negative trial in a modern systemic-therapy context. Heterogeneous oligometastatic population (synchronous/induced)Oligoprogressive diseaseCURB (47Tsai CJ, The Lancet 2024; phase II)Oligoprogressive NSCLC (≤5 progressive lesions) on/after systemic therapy; majority driver-negativeStandard-of-care systemic therapy continuedSBRT to progressing lesionsNSCLC: 59NSCLC: PFS HR 0.41 (0.22–0.75)Mixed systemic backbone; trial stopped early after interim; increased grade ≥2 AEs in SBRT armSTOP (Schellenberg, IJROBP 2025. Phase II)(1–5 progressing lesions on active systemic therapy; trial initially NSCLC-only but expanded to mixed solid tumours—dissatisfaction with omitting ablative treatments for patients in SOC armSystemic therapy continuation versus SABR to all progressing lesions+SOCSBRT to progressing lesionsNSCLC: 40(HR: not reported); median PFS 8.4 vs 4.3 mo; p=0.91; median OS 31.2 vs 27.4 mo; p=0.22Control-arm contamination. 35% in SOC arm withdrew or received high-dose/ablative therapiesAEs, adverse events; ASCO, American Society of Clinical Oncology; ESMO, European Society of Medical Oncology; gen, generation; IJROBP, International Journal of Oncology, Biology, Physics; IO, Immuno-oncology; JCO, Journal of Clinical Oncology; JNCI, Journal of the National Cancer Institute; LCT, local consolidative therapy; mo, months; NSCLC, non-small cell lung cancer; OS, overall survival; PFS, progression-free survival; RT, radiotherapy; SABR, stereotactic ablative radiotherapy; SBRT, stereotactic body radiotherapy; SOC, Standard of care; TKI, tyrosine kinase inhibitor; TRAEs, Treatment related adverse events.Phase II data and contemporary osimertinib-era evidence reinforce the same message while shifting attention toward selection and sequencing. In a randomised clinical trial by Peng et al,16 adding SBRT after an initial 3-month period of disease control on EGFR TKI improved outcomes versus TKI alone (median PFS ∼17.6 vs 9.0 months; HR ∼0.52; median OS ∼33.6 vs 23.2 months; HR ∼0.53), supporting a pragmatic ‘test-of-time’ approach that enriches for more favourable biology. NORTHSTAR17 extended the consolidative paradigm into the third-generation TKI era, with osimertinib plus LCT improving PFS compared with osimertinib alone (25.3 vs 17.5 months; HR ∼0.66) and exploratory analyses suggested greater benefit with comprehensive rather than partial consolidation, without significant excess toxicity (three cases of LCT-specific grade three adverse events: pneumonitis/arterial injury/empyema). Overall, the key message is that LCT can extend TKI benefit in selected EGFR-mutant patients, while the remaining uncertainties centre on patient selection, timing and how comprehensive treatment needs to be.A biological rationale for local ablation in TKI-treated EGFR-mutant NSCLC comes from tumour evolution. Phylogenetic studies comparing primaries and metastases support the idea that a small number of ‘key’ deposits can act as founders that later seed additional metastases; if so, eradicating limited sites could interrupt onward spread rather than simply debulk disease.18 In the TKI era, this also fits with the clinical pattern of persistent or residual lesions that may contain distinct subclones and serve as reservoirs for resistance, consistent with branched evolutionary models described in sequencing studies of osimertinib resistance.19 Viewed this way, RT to persistent lesions is not just cytoreduction but a biologically motivated attempt to remove potential sources of future resistance, complementing systemic intensification strategies such as chemotherapy or amivantamab-based approaches.A key counterbalance to any discussion of LCT is treatment-related harm—most importantly radiation pneumonitis when thoracic SABR is delivered alongside targeted therapy. A 2024 systematic review/meta-analysis20 which included studies using local SBRT with EGFR TKIs, severe treatment-related pneumonitis (grade ≥3) occurred in 3.6% of those receiving local radical/palliative RT (subgroup analysis). Importantly, sequencing mattered: severe pneumonitis was higher with concurrent TKI–RT delivery (4.9%) than with sequential approaches (0.4%). Fatal pneumonitis was rare (pooled estimate 0.1%), underscoring why thoracic SABR when combined with TKIs should be reserved for situations where expected benefit is clear and delivered with strict dose–volume constraints, careful attention to baseline lung risk factors and proactive toxicity surveillance.Much of the prospective evidence for LCT in EGFR-mutant metastatic NSCLC predates routine first-line osimertinib, which has superior systemic and intracranial efficacy and may narrow the incremental gain from upfront local therapy. As a result, optimal timing and selection for LCT in the osimertinib era remain unclear, and current guidance reflects this—ESMO does not recommend routine radical local therapy outside oligoprogression given limited randomised data and the expectation of persistent micrometastatic disease requiring ongoing TKI. In practice, LCT is therefore increasingly framed as a means to control limited resistant sites and prolong an effective systemic backbone, particularly in oligoprogressive scenarios. Thoracic RT combined with EGFR TKIs also demands caution due to pneumonitis risk, reinforcing the importance of careful selection and strict dose–volume constraints.Local consolidative therapy in Driver-Negative, Immunotherapy-Treated NSCLC Immune checkpoint inhibitors have substantially improved outcomes for patients with advanced, driver-negative NSCLC, therefore LCT in driver-negative oligometastatic NSCLC must be interpreted through the lens of modern systemic control ( table 1). Early prospective and phase II data suggest that a highly selected minority can enter a curative-intent pathway after (chemo)-immunotherapy: in one prospective analysis, 91% proceeded to definitive local therapy (resection or chemoradiation), with particularly favourable outcomes in those undergoing surgery.21 Multiple retrospective series similarly support an association between LCT and improved outcomes. In a single-centre cohort of 251 patients with oligometastatic NSCLC treated with pembrolizumab22 (with chemotherapy when indicated), the addition of LCT was associated with longer median PFS (13.97 vs 10.08 months; HR 0.64) and longer median OS (30.67 vs 21.97 months; HR 0.53). Single-arm phase II strategies have also reported durable disease control when local ablative treatment is delivered upfront followed by pembrolizumab,23 with median PFS of 19.1 months and median OS of 41.6 months; however, generalisability is constrained by the predominance of very low-burden disease, as 28 of 45 patients had a solitary metastasis. A further phase II multimodality approach (SABRcure24) using chemo-immunotherapy followed by SABR/RT and durvalumab maintenance achieved a high objective response rate (71.9%) with median PFS of 10.4 months, but wide CIs and single-arm design limit inference regarding incremental benefit.The most practice-influential evidence is the randomised NRG-LU002 trial25 (table 1), which did not show a clinically meaningful benefit for routine LCT after induction chemo-immunotherapy in non-progressing oligometastatic NSCLC (PFS HR 0.93; OS HR 1.05) and reported higher grade ≥3 pneumonitis with LCT (10% vs 1%). Interpretation should be cautious because the study population included both synchronous oligometastatic and ‘induced’ limited metastatic states, and subgroup contributions are unclear pending full publication. Against this neutral randomised signal, retrospective datasets in patients achieving limited residual disease after first-line programmed cell death protein-1/programmed death-ligand 1 (PD-L1) therapy suggest that post-induction LCT may be associated with improved outcomes, but these findings remain hypothesis-generating given potential for selection bias.26 27 A further prospective benchmark is provided by ETOP CHESS,28 which tested a protocolised multimodality strategy in synchronous oligometastatic, non-driver-mutated NSCLC (induction carboplatin/paclitaxel plus durvalumab with metastasis-directed SBRT, followed—if non-progressing—by definitive treatment of the primary and durvalumab maintenance). In the published analysis, the observed 1-year PFS rate was 33%, while 1-year OS was 74.9%; treatment-related grade ≥3 adverse events occurred in 34% with no grade 5 events. Although the trial did not meet its primary endpoint, it is important because it demonstrates the feasibility and safety of combining immunotherapy-chemotherapy with metastasis-directed SBRT, and it highlights real-world attrition and selection effects that likely influence outcomes in this setting.Alongside cytoreduction, an immune-rationale persists: MD Anderson29 30 and PEMBRO-RT31 suggest numerically higher response signals with RT plus pembrolizumab (especially in PD-L1-negative subsets), and pooled analyses report an OS signal, but definitive phase III evidence in oligometastatic-only NSCLC is lacking. Overall, the practical conclusion is that LCT should be selective—best positioned as an optimisation strategy for carefully staged patients with durable systemic control and safely ablatable targets, rather than a default add-on for all immunotherapy-treated patients.Management of brain oligometastases in EGFR-Mutated NSCLC Management of brain metastases in EGFR-mutated NSCLC is increasingly anchored by the high intracranial efficacy of osimertinib, which has shifted practice toward a systemic-first strategy in many patients. In FLAURA, 32 first-line osimertinib reduced the risk of CNS progression compared with first-generation TKIs and prolonged CNS PFS (not reached vs 13.9 months; HR 0.48), supporting initial medical management when lesions are small and clinically stable. However, CNS progression remained common over time due to resistance or sanctuary-site failure, so a long-term plan typically anticipates the later need for focal therapy, and close imaging follow-up is therefore required. In Imber et al33, the median time to best intracranial response exceeded 3 months, intracranial complete response rates were high and lesion-level local failure was low at 2 years, although cumulative CNS progression still accrued over follow-up. Clinically, this reinforces that early radiographic persistence is not necessarily failure and that short-interval MRI reassessment after starting osimertinib is reasonable in asymptomatic patients.Prospective consolidation data support systemic-first followed by local therapy. In the phase II study by Sampath et al,34 patients with stable or responding disease after initial osimertinib received radiation to persisting lesions (predominantly stereotactic radiosurgery (SRS)) followed by continued osimertinib, with encouraging PFS/OS and acceptable toxicity. By contrast, the argument for routine upfront SRS at diagnosis is less compelling. Retrospective series (eg, Zhou35) and meta-analytic summaries (Nepote36) suggest possible improvements in intracranial control and even OS in selected patients (eg, L858R mutation positive disease). The STARLET joint analysis37 pooled two small randomised phase II trials comparing upfront SRS plus osimertinib versus osimertinib alone in up to 10 brain metastases (median maximum metastasis diameter 1.1 cm, 35% had five or more metastases). Upfront SRS did not significantly improve intracranial PFS, while OS was numerically longer, with low rates of severe radionecrosis and frequent use of salvage SRS in the osimertinib-alone arm. Overall, STARLET supports osimertinib-first with selective salvage SRS as a reasonable default, reserving upfront SRS for higher-risk lesions where immediate durable local control is clinically critical.In current practice, a risk-adapted strategy is quite practical: osimertinib alone with close MRI surveillance is appropriate for small, asymptomatic lesions in non-critical locations, with SRS or surgery used for persistent lesions after initial systemic response or for focal CNS escape while maintaining an effective TKI. Upfront surgery/SRS remains preferred for lesions that are large or symptomatic, threaten critical structures (brainstem, posterior fossa, optic pathway) or where durable local control is paramount in patients expected to live long enough to experience multiple CNS events. The role of whole-brain RT is now largely restricted, informed by data such as QUARTZ32 and contemporary comparative evidence38 suggesting no meaningful survival advantage for whole brain radiotherapy (WBRT) in many settings, while stereotactic approaches for patients with up to 20 brain metastases can preserve outcomes with better neurocognitive and quality-of-life profiles compared with hippocampal-avoidance WBRT.START-NEW-ERA and the influence of upfront SABR in locally advanced NSCLC STereotactic Ablative RadioTherapy in NEWly Diagnosed and Recurrent Locally Advanced Non-Small Cell Lung Cancer Patients Unfit for ConcurrEnt RAdio-Chemotherapy (START-NEW-ERA) 39 explores an alternative local strategy for a clinically important subgroup of patients with unresectable locally advanced NSCLC who are unfit for standard concurrent chemoradiotherapy. In this non-randomised phase II study, SABR was delivered to the primary tumour and involved nodes using hypofractionated regimens (median prescription 45 Gy in five fractions to the primary and 40 Gy in five fractions to nodal disease), with a high proportion of ultracentral tumours (76% central primary disease, 82% node positive disease) and selective use of neoadjuvant chemotherapy (54%) and adjuvant durvalumab (14%). At a median follow-up of 63 months, local relapse-free survival remained durable (1-year, 3-year and 5-year local-recurrence-free survival 86%, 59% and 59%), OS was encouraging for this frail population (median OS 55 months; 5-year OS 50%), with no G3 or higher toxicity noted. These findings support the feasibility of SABR-based definitive treatment for locally advanced disease when conventional fractionation concurrent chemoradiotherapy is not deliverable.These results may reshape future oligometastatic management by improving durable intrathoracic control in patients who would otherwise relapse locoregionally, thereby shifting failure patterns toward distant relapse—often as limited-volume recurrence that is amenable to metastasis-directed therapy—and enriching for slower-tempo disease with longer metastatic latency. This supports a longer-term sequencing framework in which SABR is used at multiple points along the disease continuum: first as definitive thoracic therapy when concurrent chemoradiotherapy is not feasible, and later as metastasis-directed treatment for limited distant relapse or oligoprogression, with the proviso that this strategy requires rigorous surveillance imaging and careful attention to cumulative toxicity, particularly if reirradiation of thoracic or mediastinal structures becomes necessary. Short-course thoracic RT may also be attractive from an immune-sparing perspective, because shorter treatment courses and reduced exposure of circulating lymphocytes could lessen treatment-related lymphopenia.40SABR-COMET, SABR-COMET-10 and Lessons from Other Tumour Groups Metastasis-directed SABR has proof-of-principle support across tumour types from Stereotactic Ablative Radiotherapy for the Comprehensive Treatment of Oligometastases (SABR-COMET), 41 an international randomised phase II trial in patients with controlled primaries and one to five metastases, where adding SABR to standard palliative care improved long-term survival outcomes but also highlighted the importance of toxicity vigilance, with a small early signal of treatment-related mortality. With longer follow-up the survival signal persisted (5-year OS 42.3% vs 17.7% in controls), without late quality-of-life detriment, reinforcing that comprehensive ablation can be clinically meaningful in carefully selected patients when planning rigour and organ-at-risk constraints are prioritised.For NSCLC, the key cross-disease lessons are about how metastasis directed therapy (MDT) works: benefit appears most plausible when all visible disease can be safely treated, when modern staging reduces underdetection, and when the clinical objective is explicit (durable control/‘line extension’ vs curative intent). The ongoing SABR-COMET-1042 trial directly tests whether these principles extend beyond the classic one to five lesion boundary (4–10 metastases), and regardless of its result—underscores the central theme of this review: lesion count alone is an imperfect surrogate, and local therapy must be integrated with tumour biology, systemic therapy potency and metastatic tempo rather than applied as a uniform add-on.Other tumour groups illustrate how intent shapes benefit. In prostate cancer, STOMP43 and ORIOLE44 support MDT as a strategy to delay systemic escalation and emphasise the value of comprehensive ablation and modern imaging. In breast cancer, results are more context-dependent: NRG-BR00245 was neutral for routine metastasis ablation in newly oligometastatic disease, whereas TROG 20.03 AVATAR46 supports MDT to prolong an effective systemic line in oligoprogressive ER+/HER2− disease, and CURB47 highlights that heavily pretreated biology may blunt MDT benefitBiomarkers and molecular selection ctDNA and circulating tumour cells Circulating tumour DNA (ctDNA) is attractive in oligometastatic NSCLC because it offers a method to estimate occult tumour burden and identify minimal residual disease after local therapy—precisely the setting where imaging can underestimate risk. Across NSCLC, ctDNA dynamics (clearance vs persistence) have shown prognostic value and have been proposed as a means to personalise consolidation strategies, but prospective validation for decision-making in oligometastatic SABR pathways is still limited. 48 49 Baseline circulating tumour cells (CTCs) offer a complementary signal of systemic dissemination potential: prospective work in oligometastatic disease suggests that detectable CTCs can track disease control and may precede radiographic progression, supporting their role as a risk stratifier for ‘biologic oligometastatic’ versus early polymetastatic behaviour.50 Practical limitations remain for both assays (variable shedding by site, imperfect sensitivity—particularly for intracranial disease and uncertainty about what action to take when biomarkers are positive but imaging is stable), so at present ctDNA/CTCs are best positioned for trial enrichment and post-treatment surveillance research rather than routine decision rules for offering or withholding SABR.Immune and T-cell signatures (LONESTAR) If the next biomarker leap in driver-negative NSCLC is not genomic, it may be immunologic. Randomized Phase III Trial of Local Consolidation Therapy After Nivolumab and Ipilimumab for Immunotherapy-Naive Patients with Metastatic Non-Small Cell Lung Cancer (LONESTAR) ( NCT03391869)51 is a phase III programme evaluating dual-checkpoint immunotherapy (nivolumab/ipilimumab) with or without local consolidation therapy in advanced NSCLC, and it is notable not only for its clinical question but also for its embedded translational agenda. Correlative work linked to this trial has highlighted the potential importance of peripheral blood T-cell—higher T-cell diversity has been associated with clinical outcomes and toxicity risk on ipilimumab/nivolumab in metastatic NSCLC, supporting the idea that systemic immune ‘fitness’ may influence who benefits from intensification strategies including potential LCT. While these signals are hypothesis-generating, they point toward a plausible future in which SABR candidacy in immunotherapy-treated oligometastatic disease is partly guided by immune context rather than anatomy alone.Radiomics and future directions Radiomics offers a complementary, non-invasive approach to biological selection by extracting quantitative features from routine CT/PET imaging that may correlate with indolent versus aggressive metastatic behaviour and thereby help identify patients most likely to benefit from ablative therapy. 52 In a radiomics study of patients with oligometastatic NSCLC,53 PET-derived texture features were associated with OS and added prognostic information beyond routine clinical factors, supporting the idea that imaging can capture biological differences relevant to outcomes. However, radiomic models remain vulnerable to numerous factors including differences in scanners and acquisition protocols and few have undergone rigorous external validation for routine clinical use.Ongoing trials that may define the future, and their challenges Several ongoing studies are now positioned to clarify whether LCT should be routine, selective or avoided in the modern era of effective systemic therapy. TARGET-01 ( NCT0527784454) and TARGET-02 (NCT0527805255) are parallel phase II randomised trials designed to test whether adding comprehensive LCT improves outcomes beyond contemporary systemic therapy: TARGET-01 enrols EGFR-driven or ALK-driven oligometastatic NSCLC (≤5 lesions) and compares TKI alone versus TKI plus SBRT to all lesions (primary endpoint PFS), while TARGET-02 applies a similar strategy in driver-negative disease after non-progressing induction systemic therapy (typically four to six cycles), randomising patients with one to five metastatic lesions (excluding primary/regional nodes) to maintenance therapy with or without LCT to all metastatic sites and the primary. Within EGFR-mutant disease, prospective programmes are also testing how aggressively third-generation TKIs can be combined with locally ablative RT; a planned two-arm phase II trial of lazertinib with or without early locally ablative RT (NCT0516785156) evaluates whether early comprehensive ablation adds value on top of a modern third-generation EGFR TKI in treatment-naïve synchronous oligometastatic disease (≤5 metastases).Immunotherapy-era randomised trials will be particularly influential because they address whether LCT is additive after modern ICB-based systemic control is established. NIRVANA LUNG (NCT037747357) integrates early RT during chemoimmunotherapy, preferentially using SABR for oligometastatic patients and conventional conformal RT for more disseminated disease, explicitly testing whether upfront RT improves outcomes during checkpoint blockade. LONESTAR (NCT0339186951) is a phase III trial in stage IV NSCLC evaluating nivolumab/ipilimumab with or without LCT (surgery and/or RT) after induction treatment in non-progressing patients, with embedded biomarker work aimed at identifying predictors of benefit and harm.Finally, larger phase III frameworks will help define eligibility thresholds and clinical endpoints for LCT. ANDROMEDA (NCT0614107058) is a multicentre phase III trial randomising patients with stage IV oligometastatic NSCLC (≤5 metastases, excluding the primary and mediastinal nodes) to standard systemic treatment with or without added RT. SARON59 is a UK phase III trial (results anticipated 2026) comparing standard systemic anticancer therapy with or without RT (either conventional or SABR) in oligometastatic NSCLC (1–5 metastases in up to three organs). In the oligoprogressive setting, trials such as HALT (NCT03256981) in mutation-positive NSCLC on TKIs are evaluating whether ablating a limited number of resistant sites can extend the duration of benefit from an otherwise effective systemic regimen.Clinical Pearls for the practicing oncologist EGFR-mutated oligometastatic disease Think ‘biology+durability’. EGFR-mutated disease often has prolonged systemic control on TKIs, so SABR is most useful when you can safely treat all visible sites to prolong time on an effective TKI. The most applicable candidates are synchronous low-burden disease at diagnosis or oligopersistent disease after an initial TKI response, using rigorous staging and a comprehensive (all-sites) strategy when feasible. Frame goals clearly: this is usually about extending control and delaying resistance-driven switches, not replacing ongoing TKI.SABR in driver-negative/immunotherapy era Be selective, in the IO era, routine consolidation for all non-progressing oligometastatic patients is not supported by contemporary randomised data; consider SABR when there is a clear clinical reason for durable local control (organ-at-risk lesion, symptomatic/high-risk site, limited sites amenable to safe ablation) and when PET/brain MRI staging reduces the chance of occult polymetastatic biology.Practical guidance for CNS oligometastases Sequence matters. For small, asymptomatic EGFR-mutant brain metastases, starting osimertinib with close MRI surveillance is reasonable, with SRS reserved for persistent lesions after response or for focal CNS escape. Choose upfront SRS/surgery for large or symptomatic lesions, critical locations (brainstem/posterior fossa/optic pathways) or limited (often one to three) lesions in long-survival patients where durable intracranial control is a priority. CNS oligoprogression on TKI is a high-yield indication for focal SRS while continuing the effective systemic agent.Principles for safe ablative treatment alongside modern systemic therapy Safety and prioritisation of SABR alongside systemic treatment should be planned together: use rigorous staging and multidisciplinary review to decide whether the intent is comprehensive ablation or targeted optimisation, then deliver SABR only where the incremental benefit outweighs toxicity risk in the context of the systemic regimen. High-quality motion management and image guidance, strict adherence to organ-at-risk constraints (especially cumulative lung and proximal bronchial tree doses) and a willingness to de-escalate or omit marginal targets are essential—particularly when combining thoracic SABR with TKIs or immunotherapy, where pneumonitis risk can be amplified. Practically, SABR should be timed and targeted to preserve the value of systemic therapy (eg, ablate limited resistant/active sites to delay switching) or prevent the next clinically meaningful event (organ compromise or symptoms), rather than added reflexively because lesion number is low ( figure 1).Figure 1Summary illustration of the new oligometastatic divide in NSCLC. NSCLC, non-small cell lung cancer; SABR, stereotactic ablative radiotherapy; TKI, tyrosine kinase inhibitor.Conclusion SABR has an established role in selected oligometastatic NSCLC, but its value is increasingly determined by systemic therapy context and tumour biology rather than lesion count alone. In EGFR-mutated disease, the most consistent rationale is preserving durable TKI benefit by ablating limited-volume sites when comprehensive treatment is feasible. In driver-negative, immunotherapy-treated disease, routine consolidation is not supported by contemporary randomised evidence and SABR should be reserved for carefully staged patients where local control is likely to change the next clinical event or extend time on an effective systemic regimen. Across subtypes, future progress depends on rigorous trial design with modern systemic backbones, mandated PET/MRI staging and biomarker-integrated selection to identify the patients most likely to benefit from comprehensive local therapy.