Efficacy and safety of metastasis-directed therapy for oligometastatic cancer: a systematic review and meta-analysis of randomized controlled trials Authors: David Chen, Ealia Khoshkish, Jane Jomy, Kristen Arnold, Joshua Yi, David Palma, Mitchell Liu, Xiang Y Ye, Zhihui Liu, Gabriel Boldt, Philip Wong, Srinivas Raman Published: 2026 ABSTRACT AbstractBackgroundMetastasis-directed therapy (MDT) is a promising therapeutic strategy in oligometastatic cancer, but its safety and efficacy remains uncertain. This systematic review and meta-analysis aims to evaluate the efficacy and safety of MDT interventions in randomized controlled trials (RCT) of oligometastatic cancer.MethodsThis study was prospectively registered in PROSPERO (CRD420251075947). A literature search of MEDLINE, Embase, and Cochrane CENTRAL was conducted from inception to June 9, 2025. Eligible RCTs compared MDT interventions with standard of care in patients with oligometastatic cancer (≤5 lesions). Hazard ratios (HRs) for overall survival (OS) and progression-free survival (PFS), and adverse event (AE) rates were pooled using random-effects meta-analysis.ResultsThirty seven records representing 3,500 participants were included. Subgroup analyses showed upfront MDT improved both PFS (HR = 0.45, 95% CI 0.32–0.64) and OS (HR = 0.49, 95% CI 0.38–0.64), while consolidation MDT improved PFS only (HR = 0.56, 95% CI 0.40–0.78). In de novo oligometastatic disease, MDT improved PFS (HR = 0.50, 95% CI 0.32–0.78) but not OS, while recurrent oligometastatic disease benefited in both PFS (HR = 0.57, 95% CI 0.42–0.78) and OS (HR = 0.70, 95% CI 0.52–0.93). MDT was associated with an increased odds of Grade ≥ 2 adverse events (OR = 1.77, 95% CI [1.30–2.41]).ConclusionThe addition of MDT to standard of care improves OS and PFS in oligometastatic cancer patients, with a moderate increase in adverse events. MDT is a valuable component of multimodal therapy for oligometastatic disease, warranting further study of optimal timing, patient selection, and long-term outcomes.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13014-026-02853-6. Introduction The oligometastatic paradigm, postulated by Hellman and Weichselbaum in 1995, describes an intermediate metastatic state where patients presenting with limited metastatic burden may represent a distinct clinical state from widely disseminated, metastatic disease [1]. It has been hypothesized that patients with a limited number of metastases may be amenable to metastasis directed therapies (MDT) such as surgery or ablative radiotherapy that could potentially achieve long-term disease control and prolong survival [2]. Multiple phase II randomized trials reported in the 2010s provided proof-of-concept that MDT can improve outcomes in oligometastatic cancer. In the landmark SABR-COMET trial, patients with up to 5 metastases from various primary tumors were randomized to standard palliative care versus standard care plus stereotactic ablative radiotherapy (SABR) to all metastases. SABR-COMET demonstrated an improvement in progression free survival (PFS) and trend towards improved overall survival (OS) favoring MDT, with 5-year OS nearly doubling from 18% to 42% in the SABR experimental arm [3]. Subsequent trials evaluating MDT across different cancer types have shown mixed results. While some studies in oligometastatic prostate [4, 5] and non-small cell lung cancers [6, 7] suggest potential benefits, including delayed disease progression and improved survival, other studies in oligometastatic breast cancer have reported no association with survival [8]. This variability likely stems from differences in tumor biology, patient populations, trial designs, and imaging techniques used for patient selection. Moreover, while trials generally report acceptable toxicity rates, the potential for high-grade adverse events (AE) highlight the importance of quantifying the treatment-associated risks of MDT to inform personalized treatment selection. To address these knowledge gaps, this systematic review and meta-analysis of randomized controlled trials assessing MDT in oligometastatic cancer aims to describe the landscape of MDT RCTs and evaluate the safety and efficacy of the addition of MDT. We also explore various factors influencing treatment outcomes including treatment timing (up-front vs. consolidation), metastatic chronology (de novo vs. recurrent disease), and primary tumor site. Methods This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO) under identifier CRD420251075947. Given the nature of this systematic review and meta-analysis, ethics approval and participant consent were not required. Clinical trial number: not applicable. Search strategy and eligibility criteria A comprehensive literature search was performed across Ovid MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) databases from their inception up to June 9, 2025. The search terms included variations of oligometastasis (“oligometastasis”, “oligorecurrence”, “oligoprogressive”, “oligoresistant”, “limited metastasis”, “low volume metastasis”) combined with terms indicating randomized controlled trials (RCTs) (“randomized controlled trial”, “randomization”, “randomisation”). The reference lists of identified studies were also screened to ensure completeness. Eligible studies had to fulfill the following criteria: (i) any phase randomized controlled trials; (ii) evaluation of metastasis-directed therapy (MDT) in at least one trial arm; and (iii) inclusion exclusively of patients with oligometastatic or oligoprogressive cancer, both categorized as oligometastatic cancer going forward. Oligometastatic cancer was operationally defined based on the trial’s explicit inclusion criteria for oligometastases or a maximum number of five treated metastatic lesions. Exclusion criteria were non-English language publications, non-randomized studies (including case reports, case series, cohort studies, and case-control studies), studies evaluating solely non-therapeutic aspects such as feasibility or economic analyses without survival outcomes, and studies not limited exclusively to oligometastatic cancer patients. To address redundancy, only the most recent publication with the longest median follow-up was selected from studies with overlapping patient cohorts. For inclusion in the efficacy meta-analysis, studies had to report hazard ratios (HRs) for OS or PFS derived from Cox proportional hazards regression analyses comparing the addition of MDT intervention in the treatment arm to a control arm. For the safety meta-analysis, studies needed to report the total number of participants experiencing Grade 1 or 2 and Grade 3 or 4 CTCAE Version 5 AE in both the intervention and control arms. Study selection and data extraction Study selection processes, including abstract screening and full-text review, were independently by two reviewers (J.Y., K.A., J.J., E.K.), with disagreements resolved through consensus discussions and arbitration by a third reviewer (D.C.). Data extraction was conducted independently by pairs of reviewers (J.Y., K.A., J.J., E.K.) to extract study attributes, including trial identifiers, trial design features (phase, sample size, interventions in treatment and control arms), biological characteristics (primary tumor site, metastatic lesion sites, maximum number of treated lesions, de novo or recurrent disease, and therapy timing), reported efficacy outcomes (OS, PFS), AE counts, and standardized reported study endpoints. Definitions of intervention timing used in this review are described in Supplementary Table 1. Missing data were coded as “Not Specified” (NS). Additionally, risk of bias with RoB 2 was independently assessed by pairs of reviewers (J.Y., K.A., J.J., E.K.), with any disagreements resolved through consensus discussion involving the third reviewer (D.C.). The clinical outcomes in the combined, long-term follow up analysis of the STOMP and ORIOLE trials reported in Deek et al. [9] was used in place of the clinical outcomes reported in the STOMP [10] and ORIOLE [4] trials for the meta-analysis of efficacy outcomes. Primary tumor site-specific efficacy outcomes reported in the CURB [11] trial were used for the primary tumor site subgroup meta-analysis of efficacy outcomes. Statistical analysis Descriptive statistics were used to summarize the characteristics, primary endpoints, risk of bias assessments, and reporting quality of included studies. For the quantitative meta-analysis of the efficacy of the addition of MDT, pooled HRs and 95% confidence intervals (CIs) for OS and PFS were calculated using a random-effects model with the inverse-variance weighting method. Subgroup analyses of efficacy were conducted based on factors including therapy timing, metastasis timing, primary tumor site, maximum number of metastases, and median PFS duration. The 12-month cutoff for control-arm median PFS duration was selected as a pragmatic, clinically interpretable one-year threshold that approximately separated studies with shorter versus longer baseline disease control. Reported rates of Grade 2 to 4 and Grade 3 to 4 adverse events were grouped together as the rate of Grade ≥ 2 AEs. For the quantitative meta-analysis of the safety of the addition of MDT, pooled odds ratios for Grade 1 to 2 and Grade ≥ 2 AEs were calculated using a random-effects model with the inverse-variance weighting method. Absolute incidence rates of Grade 1 to 2 and Grade ≥ 2 AE were normalized using the Freeman–Tukey double-arcsine transformation and then pooled using a random-effects model with the inverse-variance weighting method. Grade 5 events were not meta-analyzed due to the paucity of available data. Statistical heterogeneity among studies was evaluated using the I2 statistic, where I2 values of 50–75% indicated moderate heterogeneity and I2 values above 75% indicated substantial heterogeneity. Egger’s test was used to assess publication bias [12]. Trim-and-fill method was used to correct pooled estimates for publication bias [13]. The certainty of evidence for each primary outcome was assessed using the GRADE methodology, evaluating risk of bias, inconsistency, indirectness, imprecision, and publication bias [14]. All hypothesis tests were two-sided and used a p-value of 0.05 as the threshold for statistical significance. Statistical analyses were performed using metafor version 4.8. Results The systematic review search included 2,051 articles. Abstract screening of 1,541 unique articles followed by full text screening of 264 articles yielded a final search result of 37 articles (Supplementary Fig. 1). The 37 articles are comprised of 22 full text articles and 15 conference abstracts that collectively represent 29 unique studies that evaluated MDT in 3,500 participants with oligometastatic cancer [4–11, 15–41]. Risk of bias assessment of the 22 full text articles demonstrated that 14 (61%) articles had some concerns of overall bias (Supplementary Fig. 2). No studies were identified as having a serious concern of overall bias. The overall risk of bias assessment was driven predominantly by some concerns in the domains of deviations from intended interventions (n = 18, 82%) and measurement of the outcome (n = 10, 46%). Most trials were phase II trials (n = 27, 93%) that evaluated the addition of radiation to unspecified standard of care (n = 7, 24%), hormonal therapy (n = 5, 17%), observation (n = 3, 10%), chemotherapy (n = 2, 7%), and targeted agents (n = 2, 7%) (Supplementary Table 3). Included studies evaluated upfront therapy (n = 16, 55%) and consolidation therapy (n = 8, 28%), with a minority of studies evaluating oligoprogression (n = 3, 10%). Enrolled participants were most commonly experiencing recurrent oligometastasis (n = 16, 55%) compared to de novo oligometastasis (n = 6, 21%). Common primary tumor sites include the prostate (n = 12, 41%), lung (n = 8, 28%), any solid tumor (n = 5, 17%), and breast (n = 4, 14%). The maximum number of treated metastases for enrolled participants included 5 (n = 15, 52%), 3 (n = 11, 38%), and 4 (n = 3, 10%). The most prevalent primary endpoints reported in the literature included PFS (n = 24, 83%), OS (n = 20, 69%), toxicity (n = 16, 55%), local control (n = 13, 45%), and use of subsequent therapy (n = 12, 41%). Among included trials, PET/CT was incorporated into baseline staging in 12 trials (41%), whereas 3 trials (10%) used conventional imaging only, and 14 trials (48%) did not specify their baseline staging strategy The addition of MDT for patients with oligometastatic cancer demonstrated improved PFS (HR = 0.52, 95% CI=[0.42–0.65], I2 = 75%) across 2,137 participants in 22 studies. Likewise, the addition of MDT for patients with oligometastatic cancer also demonstrated improved OS (HR = 0.70, 95% CI=[0.56–0.89], I2 = 44%) across 1,611 participants in 16 studies. Egger’s test for publication bias was significant for PFS (p = 0.022) and non-significant for OS (p = 0.85). Trim-and-fill correction of publication bias confirmed that the addition of MDT is indeed associated with improved PFS (HR = 0.70, 95% CI=[0.55–0.90], I2 = 82%) (Supplementary Table 4). Subgroup analysis based on intervention timing revealed that upfront therapy is associated with improved PFS (HR = 0.45, 95% CI=[0.32–0.64], I2 = 78%, n = 10) and OS (HR = 0.49, 95% CI=[0.38–0.64], I2 = 0%, n = 6), while consolidation therapy is associated with improved PFS (HR = 0.56, 95% CI=[0.40–0.78], I2 = 70%, n = 8) but not OS (HR = 0.89, 95% CI=[0.60–1.31], I2 = 49%, n = 6) (Figs. 1A, 1B). MDT of oligoprogressive lesions was not associated with improved PFS (HR = 0.75, 95% CI=[0.37–1.51], I2 = 79%, n = 2) or OS (HR = 0.88, 95% CI=[0.56–1.38], I2 = 0%, n = 2). Stratified by de novo and recurrent oligometastasis timing, we observed that the addition of MDT to the treatment of de novo oligometastasis is associated with improved PFS (HR = 0.50, 95% CI=[0.32–0.78], I2 = 83%, n = 6) but not OS (HR = 0.74, 95%CI=[0.45–1.20], I2 = 72%, n = 5), while improved PFS (HR = 0.57, 95% CI=[0.42–0.78], I2 = 78%, n = 11) and OS (HR = 0.70, 95% CI=[0.52–0.93], I2 = 31%, n = 8) was observed in the recurrent oligometastatic setting (Figs. 2A, 2B). Meta-analysis of progression free survival and overall survival stratified by intervention timing. A. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of progression free survival. B. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of overall survival. *the clinical outcomes in the combined, long-term follow up analysis of the STOMP and ORIOLE trials reported in Deek et al. [9] was used in place of the clinical outcomes reported in the STOMP [10] and ORIOLE [4] trials for the meta-analysis of efficacy outcomes Meta-analysis of progression free survival and overall survival stratified by metastasis timing. A. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of progression free survival. B. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of overall survival Across primary tumor sites, we observed that only prostate (HR = 0.47, 95% CI=[0.31–0.72], I2 = 68%, n = 7) and lung (HR = 0.57, 95% CI=[0.44–0.75], I2 = 47%, n = 7) showed significant effect of MDT on PFS, while no association with PFS was observed in breast cancer (HR = 0.75, 95% CI=[0.49–1.14], I2 = 60%, n = 3) (Fig. 3A). No association with OS was observed across any standalone primary tumor site subgroup (Fig. 3B). When subgrouped based on median PFS duration in the control trial arm, we observed that trials with a baseline of less than 12 months median progression free survival demonstrated a positive association between the addition of MDT and overall survival (HR = 0.67, 95% CI=[0.51–0.88], I2 = 35%, n = 10) (Fig. 4). Trials with a baseline of greater than 12 months median progression free survival did not show an association between the addition of MDT and overall survival (HR = 0.71, 95% CI=[0.40–1.26], I2 = 52%, n = 5) (Fig. 4). Meta-analysis of progression free survival and overall survival stratified by organ site of the primary tumor. A. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of progression free survival. B. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of overall survival Meta-analysis of overall survival stratified by median progression free survival. Forest plot of the pooled and study-specific hazard ratios and 95% confidence intervals of overall survival In total, 6 trials with 473 participants reported arm-specific rates of any Grade 1 or 2 AE, 7 trials with 803 participants reported arm-specific rates of any Grade ≥ 2 AE, and 7 trials with 479 participants reported arm-specific rates of any Grade 3 to 4 AE. The addition of MDT in the experimental intervention arm was not associated with increased odds of Grade 1 or 2 events (OR = 1.50, 95% CI=[0.87–2.59], I2 = 33%, n = 6) (Fig. 5A). Egger’s test for publication bias was significant for Grade 1 or 2 AEs (p = 0.016). Trim-and-fill correction estimated that the addition of MDT was associated with increased odds of Grade 1 to 2 AEs (OR = 1.98, 95% CI=[1.14–3.43], I2 = 37%, n = 9). The addition of MDT in the experimental intervention arm was associated with increased odds of Grade ≥ 2 events (OR = 1.82, 95% CI=[1.34–2.46], I2 = 0%, n = 14) (Fig. 5B). Egger’s test for publication bias was not significant for Grade ≥ 2 AEs (p = 0.13). Any Grade 1 or 2 (n = 6) AE occurred in 40% of participants in the experimental arm and 26% of participants in the control arm. Any Grade 3 or 4 (n = 7) AE occurred in 18% of participants in the experimental arm and 8% of participants in the control arm (Supplementary Table 5). Meta-analysis of rates of any adverse event stratified by adverse event grade. A. Forest plot of the pooled and study-specific odds ratios and 95% confidence intervals of Grade 1 or 2 adverse events. B. Forest plot of the pooled and study-specific odds ratios and 95% confidence intervals of Grade 2 to 4 adverse events and Grade 3/4 adverse events Using the GRADE approach, we concluded with a high certainty of evidence that the addition of MDT improves OS and moderate certainty that the addition of MDT improves PFS after downgrading for publication bias. Furthermore, we concluded with a low certainty of evidence that the addition of MDT is associated with Grade 1–2 AEs after downgrading for publication bias and imprecision and moderate certainty that the addition of MDT is associated with Grade ≥ 2 AEs after downgrading for imprecision (Supplementary Table 6). Discussion In this systematic review and meta-analysis of 29 studies including 3,500 patients, we found that the addition of metastasis-directed therapy to standard of care is associated with an improvement in OS and PFS for patients with oligometastatic cancer. Our results confirm and extend the promising signals observed in earlier individual studies of oligometastatic cancer first reported in RCTs such as SABR-COMET, STOMP, and ORIOLE [3–5]. Improved survival outcomes support the incorporation of MDT as part of multidisciplinary management for carefully selected oligometastatic patients with tolerable toxicity profiles. Subgroup analyses revealed insights regarding optimal MDT timing. Up-front MDT, integrated at the first detection of metastatic disease, appeared to demonstrate the most pronounced improvements in both PFS and OS. In contrast, consolidation MDT after first-line systemic therapy was associated with improved PFS but had no association with OS. This result aligns with prior findings from lung cancer consolidation trials [7] and potentially underscores the value of early intervention, where treating all visible metastases as part of initial therapy may effectively limit metastatic disease before the development of resistant clones or widespread progression. This biologic rationale is supported by genomic studies of advanced prostate cancer demonstrating greater mutational complexity in metastatic and treatment-exposed disease and linking resistance-associated alterations with poorer outcomes, consistent with clonal diversification under treatment pressure [42]. If MDT is delayed until after systemic therapy, local ablation of known lesions may temporarily control metastatic disease but may not impact survival due to occult metastases. However, the strength of the conclusion is limited by the observation that patients who were treated with consolidation MDT likely had an increased burden of metastatic disease at presentation, reflecting a worse prognosis and higher risk for polymetastatic progression. We also observed no OS or PFS benefits from the addition of MDT to manage oligoprogressive lesions. We speculate that this result may reflect poor statistical power or could suggest that oligoprogressive lesions unresponsive to systemic therapy harbor resistance mechanisms that are not addressed by the addition of MDT [43]. However, specific populations, such as the lung cancer cohort in the CURB trial, did experience a PFS benefit, that was not seen in the other breast cancer cohorts. The role of MDT in oligoprogressive disease remains uncertain and requires larger prospective studies to clarify whether oligoprogressive subgroups may benefit from MDT, or if alternative strategies, such as earlier systemic therapy escalation, are warranted [44]. In regards to metastatic disease presentation, we observed that patients experiencing oligorecurrent disease derived improvement in PFS and OS, supporting the role of MDT at disease relapse. Conversely, MDT in patients presenting with de novo oligometastases yielded improved PFS without OS benefit. Although the lack of OS benefit in de novo oligometastatic disease may reflect a higher burden of occult disease and more aggressive tumor biology, this finding may also have been influenced by between-study differences in follow-up maturity and the systemic therapy backbones used in de novo cohorts, and should therefore be interpreted as hypothesis-generating rather than definitive.Our findings suggest that patients with oligometastatic recurrence after a disease-free interval may be the strongest candidates for MDT while patients with de novo oligometastatic disease may require alternative management strategies for improved survival outcomes, such as combined systemic therapy intensification alongside MDT. The efficacy of MDT may also vary by histology defined by primary tumor type. Our meta-analysis found that prostate and lung cancer patients with oligometastatic disease demonstrated PFS benefits from the addition of MDT, aligning with results from prior meta-analyses subgrouped based on histology [45]. In contrast, our pooled analysis of two breast cancer trials did not demonstrate PFS or OS benefits of MDT, suggesting variations in tumor biology and systemic therapy efficacy that could mitigate MDT’s impact. Breast cancer’s oligometastatic behavior may be unique given that most subtypes can often be durably controlled with multiple lines of effective systemic therapies [46]. Tumor biology defined by molecular biomarkers such as genomic alterations may play a role in determining which oligometastatic patients stand to benefit the most from MDT. In a pooled analysis of the STOMP and ORIOLE trials, patients with pathogenic alterations in ATM, BRCA1, BRCA2, RB1, or TP53 demonstrated shorter PFS from MDT compared to those without these mutations, implying that MDT alone is often insufficient for durable control if genomic driver mutations associated with aggressive tumor biology are present [9]. Ongoing biomarker-stratified trials, such as the phase II KNIGHTS trial (NCT06212583) of oligometastatic prostate cancer patients with DNA repair mutations randomized to intensified systemic therapy in combination with MDT versus MDT alone, reflects the emergence of biomarker-driven MDT personalized to the metastases’ molecular profile to maximize clinical benefit [47]. In addition, circulating tumor DNA (ctDNA) assays present an underexplored modality for prognostication and post-treatment surveillance of oligometastatic cancer recurrence. [48] Despite promising findings in pilot trials, the integration of ctDNA assays into clinical practice faces ongoing challenges due to the need for standardized assay methodologies, optimization of sampling intervals, and establishing clinically relevant thresholds for assay positivity[49]. While MDT offers a promising approach to improve survival in oligometastatic patients, the increase in both Grade 1 to 2 AEs after correcting for publication bias and the increase in Grade ≥ 2 AEs reinforces the importance of careful patient selection, vigilant safety monitoring, as well as optimization of MDT timing, sequence, and agents to reduce toxicities. Future research should aim to determine the populations most likely to benefit from MDT while judiciously considering which patient subgroups may be predisposed to adverse safety outcomes. In a discrete choice experiment of 870 US respondents with cancer experience, participants preferred greater certainty of survival benefit and were willing to accept delays in care access linked with disease progression to prioritize clinical benefit [50]. Patient tolerance of toxicity and efficacy thresholds may differ from physicians and vary temporally with the evolution of disease and symptom burden, underscoring the need for tailored communication about risk–benefit trade-offs of MDT [51, 52]. Ongoing research involving image-guided and intensity-modulated radiotherapy approaches will continue to improve the precision of dose delivery and tolerability without compromising efficacy. Even in settings where an OS benefit is not demonstrable, MDT may still confer patient-centered value by delaying progression and postponing escalation to a subsequent systemic line, thereby allowing patients to remain longer on a regimen they are currently tolerating and may be controlling other areas of disease [53]. Accordingly, absence of OS benefit should not be interpreted as absence of clinical benefit when QoL and time-to-next-therapy endpoints are incompletely reported. Ongoing prospective phase III trials, such as SABR-COMET-10 [41], are poised to establish the long-term survival benefit of MDT in cohorts with expanded lesion count thresholds that may alter the classical definition of oligometastasis beyond 5 metastases. Basket trials remain essential for establishing proof-of-principle across tumor types, but we speculate that disease-specific randomized studies can refine our understanding of how MDT should be integrated with contemporary systemic therapy within individual histologies. For example, integrating systemic therapies, including targeted therapy and immunotherapy, with MDT may also address occult micrometastases that can further improve treatment efficacy. Radiotherapy dose and fractionation schedules were highly heterogeneous across included trials and tailored by disease site and location-specific constraints, contributing to heterogeneity and potential variability in outcomes. Lastly, biomarker-driven approaches are important steps towards personalized MDT strategies for high-risk molecular subgroups. Redefining oligometastasis based on tumor biology rather than lesion counts, such as genomic, epigenetic, transcriptomic, and immunologic signatures that distinguish oligometastatic tumors with limited metastatic potential, is a promising future direction of research [54]. Taken together, our findings highlight both the promise and the complexity of MDT in oligometastatic cancer. The consistent improvement in PFS and OS across multiple RCTs affirms MDT as a clinically effective strategy, but the heterogeneity of outcomes across timing, disease setting, and tumor histologies underscores that further stratification of demographics who stand to benefit the most from MDT is needed. In particular, the favourable effect of upfront MDT and MDT in recurrent oligometastatic disease suggest that disease trajectory and treatment sequencing are critically associated with survival, while the lack of OS benefit in de novo oligometastases invites further discussion about whether systemic intensification is necessary in parallel with local ablation. The observation of increased adverse events with MDT, though moderate, is also clinically relevant depending on the disease burden of oligometastatic cancer and patient goals of care. Thus, our critical review supports MDT as an integral component of multimodality oligometastatic cancer care, but also emphasizes the importance of patient selection, careful discussion of risk–benefit trade-offs, and the integration of MDT with systemic and biomarker-driven approaches to maximize durable benefit. We acknowledge several limitations of our review. First, our meta-analysis is based on aggregate, trial-level data instead of individual patient data. Second, the substantial heterogeneity in the primary PFS analysis (I2 = 75%) suggests that the pooled effect should be interpreted as an average across an oligometastatic population with heterogeneity from but not limited to tumor histology, metastatic chronology, intervention timing, and systemic therapy backbone. We explored known subgroups that may explain study heterogeneity, but we acknowledge that residual variability can remain. Third, we were unable to examine finer subgroups of oligometastatic disease that may differ in prognosis and responsiveness to MDT. For example in prostate cancer, this may include first-line versus later-line oligoprogression or hormone-sensitive versus castration-resistant prostate cancer, or modern molecular-imaging-defined versus conventionally defined oligometastatic states. We acknowledge that our subgroup findings should be considered hypothesis-generating rather than definitive. For example, the lack of OS benefit in de novo oligometastases compared to OS benefit in oligorecurrence associated with MDT should ideally be confirmed with patient-level meta-analysis or prospective, phase III trials with long-term follow-up. Fourth, our review included RCTs of oligometastatic cancer defined as 5 metastases or fewer, precluding extrapolation of the study findings to metastatic disease with more than 5 metastases. Fifth, the primary outcomes of this review were measures of clinical safety and efficacy, although we acknowledge that outcomes such as quality of life [55] and cost-effectiveness [56] should be considered as part of multifactorial evaluations of MDT utility [57]. Sixth, treatment intent was not uniformly reported across trials, which limited our ability to perform a formal subgroup analysis by radical or palliative treatment intent. Although our study identified a small increase in adverse events, the heterogeneity in reporting precluded any conclusions about the timing or chronicity of these events. Conclusion Patients with oligometastatic cancer represent a unique clinical subgroup amenable to MDT that can improve clinical outcomes. This systematic review and meta-analysis demonstrated that the addition of metastasis-directed therapy prolongs PFS and OS in oligometastatic patients along with an increase in moderate toxicities. Up-front incorporation of MDT and treatment of oligorecurrences yielded the strongest survival benefits among subgroups, indicating that early and aggressive intervention is associated with positive clinical outcomes. Future research aims to evaluate the long-term outcomes of MDT as well as define biomarker-based patient cohorts and combination treatment strategies that may harbor the greatest clinical benefit associated with MDT. In summary, metastasis-directed therapy has emerged as an effective and safe component of multidisciplinary cancer management for patients with limited metastatic disease, offering the potential for prolonged survival in select cases. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Supplementary Material 6 Supplementary Material 7 Supplementary Material 8 Author contributions DC conceptualized, design the methodology, conducted the investigation and formal analysis, curated the data, wrote the original manuscript draft, prepared visualizations, and supervised the research. EK, JJ, KA, and JY conducted the investigation and revised the manuscript draft. XY, ZL, and GB conducted the investigation, formal analysis, and revised the manuscript draft. DP, ML, and PW revised the manuscript draft. SR wrote the original manuscript draft and supervised the research. Funding None. Data availability All data are included in the study materials at the time of study publication. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Jane Jomy and Kristen Arnold equal contribution to this work. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Supplementary Material 5 Supplementary Material 6 Supplementary Material 7 Supplementary Material 8 Data Availability Statement All data are included in the study materials at the time of study publication.