Cancer metastasis is the leading cause of cancer-related mortality, and most patients with metastases from solid tumors have historically been considered incurable. Here, we discuss the evolution of our understanding of the oligometastatic state with an emphasis on the view that cancer metastasis represents a spectrum of disease. We highlight several recently published prospective clinical trials demonstrating improvements in cancer-specific outcomes with the utilization of metastasis-directed local therapies. We discuss biological aspects of oligometastases, including genetic, epigenetic, and immune determinants of the metastatic spectrum. Finally, we propose future considerations regarding clinical trial design for patients with oligometastatic disease.

Cancer metastasis is the leading cause of cancer-related mortality, and most patients with metastases from solid tumors are considered incurable (1). Toward the last quarter of the 20th century, certain exceptions were found: For example, metastatic germ cell tumors exhibited high cure rates when treated with multi-agent systemic therapy (2). An increased understanding of signaling pathways influencing tumor progression resulted in the development and successful application of targeted therapies, first in the setting of hormonal blockade in breast and prostate cancer and later in a broader context (3, 4). Immunotherapy also demonstrated initial successes with rare cures when applying IL-2 in patients with metastatic renal cell carcinoma (RCC; ref. 5). More recently, a minority of patients with metastatic melanoma and other patient subsets are likely cured with immune checkpoint blockade (ICB; refs. 6, 7). However, it is important to note that although these systemic therapies have improved outcomes, the majority of patients with metastatic disease either do not respond or experience disease progression after initial response.

In 1995, a different conceptual paradigm was introduced—that of oligometastasis (8). This concept proposed that an intermediate state exists between locoregionally confined and widely disseminated malignancy. Some initial data that provided support for the oligometastatic hypothesis actually preceded its formulation by decades and included surgical reports that demonstrated long-term survival in a significant minority of patients treated with resection of metastases from a variety of primary solid tumors (9, 10). More recently, numerous prospective trials of ablative therapy have demonstrated meaningful improvements in cancer-specific outcomes in patients with limited metastatic disease. Our understanding of the molecular underpinnings of the oligometastatic state is increasing in step. This review aims to summarize the key clinical and biologic data supporting the existence of the oligometastatic state before delving into a detailed discussion of the opportunities that lie ahead.

Several surgical series have demonstrated prolonged survival following resection of a limited number of metastases from a variety of primary solid tumors. In the setting of colorectal cancer metastasis to the liver, a retrospective cohort comprising 1,001 patients demonstrated a 10-year overall survival (OS) rate of 22% following hepatic resection (11). Another series of 929 patients demonstrated a 10-year cancer-specific survival (CSS) of 23% in the same setting (12). In an international, multi-institutional cohort of 5,206 cases of lung metastasectomy of a variety of histologies, the 5- and 15-year actuarial survivals after complete metastasectomy were 36% and 22%, respectively (13). Another cohort of 575 patients undergoing 708 lung metastasectomies, mostly for metastatic carcinomas and sarcomas, demonstrated a similar 5-year actuarial survival of 46% (14). More recently, a systematic review of 757 patients with oligometastatic non–small cell lung cancer (NSCLC) treated with either surgery or radiotherapy demonstrated a 5-year OS of 29% (15). Numerous smaller series with similar findings have been published (9, 10). Taken in aggregate, these results clearly demonstrate that a subset of patients with metastases from solid malignancies can be cured and/or experience long-term survival after metastasis-directed localized therapies (see Table 1).

Table 1.

Selected studies demonstrating predictors of oligometastatic disease.

StudyClassifierKey findings
Fong et al. (11) Clinical Developed a preoperative scoring system for predicting recurrence after hepatic resection of metastatic colorectal cancer, which included lymph node positivity, DFI from primary to metastases <12 months, number of hepatic metastases >1, largest metastasis >5 cm, and carcinoembryonic antigen level 
Pastorino et al. (13) Clinical Large clinical cohort of lung metastasectomy patients demonstrated improved survival for patients with germ cell tumors, DFI ≥36 months, and single site of metastatic disease 
Ashworth et al. (15) Clinical Systematic review of patients with NSCLC with 1 to 5 synchronous or metachronous metastases treated with local therapy that classified patients on recursive partitioning analysis into 3 risk groups: low risk with metachronous metastases, intermediate risk with synchronous metastases and no lymph node involvement, and high risk with synchronous metastases and lymph node involvement 
Turajlic et al. (44) Genetic Classification of primary RCCs and matched metastases by somatic copy-number alterations and intratumoral heterogeneity identified a group with intermediate, oligometastatic potential that demonstrated high intratumoral heterogeneity 
Iacobuzio-Donahue et al. (31) Genetic SMAD4 loss associated with widely disseminated rather than locally progressive pancreatic carcinoma (present in 78% of patients with polymetastases vs. 22% without) 
Pitroda et al. (43) Epigenetic Similarity network fusion (SNF) analyses of mRNA and miRNA in patients treated with hepatic resection for limited metastases from colorectal cancer demonstrated 3 groups: “canonical,” “immune,” and “stromal” 
  The combination of SNF group and clinical risk score predicted clinical outcomes better than either classifier alone 
Oshima et al. (59) Epigenetic Induction of 14q32 miRNA expression in vitro abrogated liver metastasis development in vivo without affecting cell viability 
  In the clinical setting (hepatic metastases from colorectal cancer), 14q32 miRNA expression directly correlated with overall survival 
Lussier et al. (63) Epigenetic Enhanced function of miR-200c in vitro increased metastatic burden in vivo, and miR-200 family overexpression correlated with polymetastases clinically 
Van den Eynde et al. (42) Immune The ImmunoScore (semiquantitative representation of CD3+ and CD8+ T-cell infiltration of tumor specimens) of the least immune-infiltrated colorectal cancer metastasis was associated with clinical outcome 
Pitroda et al. (43) Immune Two of the SNF groups in the aforementioned study exhibited immune-related signatures 
StudyClassifierKey findings
Fong et al. (11) Clinical Developed a preoperative scoring system for predicting recurrence after hepatic resection of metastatic colorectal cancer, which included lymph node positivity, DFI from primary to metastases <12 months, number of hepatic metastases >1, largest metastasis >5 cm, and carcinoembryonic antigen level 
Pastorino et al. (13) Clinical Large clinical cohort of lung metastasectomy patients demonstrated improved survival for patients with germ cell tumors, DFI ≥36 months, and single site of metastatic disease 
Ashworth et al. (15) Clinical Systematic review of patients with NSCLC with 1 to 5 synchronous or metachronous metastases treated with local therapy that classified patients on recursive partitioning analysis into 3 risk groups: low risk with metachronous metastases, intermediate risk with synchronous metastases and no lymph node involvement, and high risk with synchronous metastases and lymph node involvement 
Turajlic et al. (44) Genetic Classification of primary RCCs and matched metastases by somatic copy-number alterations and intratumoral heterogeneity identified a group with intermediate, oligometastatic potential that demonstrated high intratumoral heterogeneity 
Iacobuzio-Donahue et al. (31) Genetic SMAD4 loss associated with widely disseminated rather than locally progressive pancreatic carcinoma (present in 78% of patients with polymetastases vs. 22% without) 
Pitroda et al. (43) Epigenetic Similarity network fusion (SNF) analyses of mRNA and miRNA in patients treated with hepatic resection for limited metastases from colorectal cancer demonstrated 3 groups: “canonical,” “immune,” and “stromal” 
  The combination of SNF group and clinical risk score predicted clinical outcomes better than either classifier alone 
Oshima et al. (59) Epigenetic Induction of 14q32 miRNA expression in vitro abrogated liver metastasis development in vivo without affecting cell viability 
  In the clinical setting (hepatic metastases from colorectal cancer), 14q32 miRNA expression directly correlated with overall survival 
Lussier et al. (63) Epigenetic Enhanced function of miR-200c in vitro increased metastatic burden in vivo, and miR-200 family overexpression correlated with polymetastases clinically 
Van den Eynde et al. (42) Immune The ImmunoScore (semiquantitative representation of CD3+ and CD8+ T-cell infiltration of tumor specimens) of the least immune-infiltrated colorectal cancer metastasis was associated with clinical outcome 
Pitroda et al. (43) Immune Two of the SNF groups in the aforementioned study exhibited immune-related signatures 

The authors of the colorectal and pulmonary metastasectomy studies performed multivariate analyses to identify prognostic clinical factors. These were consistent across studies and included number of metastatic tumors, tumor size, tumor marker levels, metastasis disease-free interval (DFI), synchronicity of metastases, primary histology, control of the primary tumor, as well as the presence of lymphatic and/or multi-organ metastases. The identification of these factors has been important for several reasons.

First, risk stratification systems combining these factors further differentiate patient subsets by outcome. For example, in one of the colorectal cancer cohorts, the group with the most favorable versus the least favorable risk classifier experienced a 5-year CSS of 64% versus 2%, respectively (12). In the NSCLC cohort, those patients with metachronous distant-only metastases had a 5-year OS of 47.8% as compared with patients presenting with synchronous lymphatic and distant metastases, whose 5-year OS was only 13.8% (15). Second, the fact that it is these specific factors rather than others that predict clinical outcome has biological implications and has driven translational insights into the nature of oligometastasis. Finally, one of these clinical factors (number of lesions) has been used as enrollment criteria for completed and ongoing prospective clinical trials testing the benefit of local therapy in oligometastatic disease from a variety of primary tumors.

The site-agnostic, prospective phase II SABR-COMET trial randomized 99 patients with controlled primary tumors and ≤5 metastases to standard of care alone versus standard of care plus stereotactic body radiotherapy (SBRT), a form of ablative radiotherapy, to all sites (16, 17). After a median follow-up of 26 months, there was a 13-month improvement in OS in the SBRT arm. In the setting of oligometastatic NSCLC (defined as ≤3 metastases), a prospective phase II trial with a median follow-up of 38.8 months that randomized patients to local consolidative therapy (surgery or radiotherapy, including SBRT) + maintenance therapy versus maintenance therapy alone demonstrated a 2-year improvement in median survival in the SBRT arm (P = 0.017; ref. 18). In a phase II study, randomizing patients with colorectal cancer with up to 9 liver metastases to systemic treatment + local treatment (radiofrequency ablation and/or resection) versus systemic treatment alone, 8-year OS was 35.9% versus 8.9%, respectively (P = 0.01; ref. 19). More recently, the phase II ORIOLE study in oligometastatic prostate cancer (defined as ≤3 metastases) randomized 54 men 2:1 to SBRT versus observation; the median progression-free survival (PFS) was not reached in the SBRT group versus 5.8 months in the observation group (HR, 0.30; 95% confidence interval, 0.11–0.81; P = 0.002; ref. 20). Several other prospective studies in prostate cancer and NSCLC have been published with similar results (21).

Two other important points must be noted. The first is that not all data have been positive to date. For example, a recently published (albeit underpowered) randomized trial of pulmonary metastasectomy versus active monitoring in patients with metastatic colorectal cancer did not demonstrate a benefit to local therapy (22). The second is that the above data are specifically focused on the benefits of local therapy in either the de novo or metachronous oligometastatic state rather than in the setting of oligoprogressive disease, in which patients with metastatic disease experience progression in a limited number of sites. Studies have demonstrated a benefit to local ablation in this setting as well, for example, in patients with oligoprogressive EGFR-mutant NSCLC while on crizotinib or erlotinib (23), though we will limit most of our subsequent discussion to de novo or metachronous oligometastasis.

To confirm the benefit of local ablation in this setting, numerous phase III randomized control trials investigating standard of care ± local therapy for oligometastatic disease are ongoing. The results of these trials are eagerly awaited and should increase our insight into this clinical state. However, it is important to note that the definition of oligometastasis used as an inclusion criterion on these trials is usually based on number of metastases. These selection criteria may result in the enrollment of patients who have few metastases but biologically aggressive disease while excluding those with a greater number of more indolent lesions. Taken together, these two tendencies could lead to the dilution of the benefits of local therapy in this setting. At the same time, consideration of the biology implied by the other prognostic factors such as DFI and tumor marker levels can inform future trial design and secondary trial endpoints that should provide the field with more robust ways to define and track the oligometastatic state.

Primary tumors are made up of subpopulations of clonogens with different propensities to undergo the crucial steps of the metastatic cascade (24, 25). This fact, combined with discoveries implicating specific molecular pathways in each of these steps, has led to significant interest in characterizing the tumor microenvironment in which these subpopulations first develop and the ways in which it shapes the metastatic phenotype (26, 27). Recent reformulations of the classic “seed and soil” hypothesis acknowledge the importance of the primary tumor “soil” and the complex tumor–host interactions that determine metastatic proclivity (28). At the same time, there has been an increasing understanding of the dichotomous role of host innate and adaptive immunity in the development of metastatic disease (29). A greater understanding of oligometastasis as part of a broader spectrum of metastasis may shed light on the involvement of these mechanisms. Using the clinical factors first identified in the surgical series outlined above, emerging preclinical and translational data are beginning to make inroads into this very question. We therefore now turn to this area of research.

Tumor burden

The number of metastases has long been considered one of the defining features of oligometastasis and features prominently in the recent ASTRO and EORTC classification recommendations (30). It is frequently used as a marker of the oligometastatic state both in clinical trial enrollment as well as in translational studies investigating the biology of metastatic disease. One early rapid autopsy series correlated patterns of failure with pathogenic mutations in 76 patients with locally advanced or metastatic pancreatic cancer (31). It found that SMAD4—an important mediator of TGF-β signaling, which in turn is important in the epithelial–mesenchymal transition (EMT) and invasiveness—was differentially lost from patients who died from tumors that underwent widespread metastatic dissemination (78% loss) as compared with those who died from local disease without metastases (22% loss). SMAD4 has been implicated in other aspects of pancreatic cancer biology and progression in other clinical cohorts (32, 33), and these data provide a concrete example of how the biology of the underlying tumor can alter the metastatic phenotype (see Table 1).

More recent clinical data shed light specifically on the biology of the oligometastatic state and demonstrate the importance of an accurate determination of burden of disease using novel imaging techniques. The prospective phase II ORIOLE study in oligometastatic, hormone-sensitive prostate cancer (defined as ≤3 metastases) demonstrated significant benefits to 6-month PFS with SBRT over observation (20). Several secondary outcomes involving the use of molecular imaging and liquid biopsy correlatives lend additional insights. The first is that 36 participants underwent prostate-specific membrane antigen (PSMA)–targeted PET—which detects metastases more successfully than conventional imaging—before study therapy. Because of blinding to the PSMA-PET results, 16 of 36 patients treated with SBRT had baseline PET-avid metastases that were not treated. The proportion of men with untreated lesions with progression at 6 months was 38% compared with 5% for those with no untreated lesions (P = 0.03); the rate of new distant metastases at 180 days was 62.5% versus 15.8% (P = 0.006), respectively. This suggests that emerging technologies may allow for a true characterization of the number of lesions in oligometastatic disease and that this in turn can improve patient selection for local ablative therapies.

The authors also defined a high-risk circulating tumor DNA-based mutation signature by CAPP-seq; high-risk genes included ATM, BRCA1/2, KRAS, PTEN, RB1, and TP53, among others. Interestingly, when stratifying by the presence of these high-risk mutations, SBRT improved outcomes over observation in the high-risk mutation negative subgroup but not in the high-risk mutation positive subgroup. Given that the clinical benefit seen on prospective trials of locally ablative therapies in patients with presumed oligometastatic disease is the strongest evidence to date of the existence of the oligometastatic state, the lack of benefit in patients with high-risk mutations in this cohort of patients with 1 to 3 metastases implies that underlying biology trumps apparent clinical risk factors in some situations. These data also suggest the future possibility of using genetic classifiers to stratify patients for local ablation.

Although there is a growing body of evidence tying molecular features to tumor number in the oligometastatic state, another important marker of tumor burden—tumor size—awaits definitive study. However, it is suggestive that the presence of a metastatic tumor >5 cm was a negative prognostic factor in the colorectal hepatic metastasectomy series and large tumor size is associated with worse local control following definitive local ablative therapy in NSCLC (>5 cm) and hepatocellular carcinoma (>4 cm; refs. 34, 35). One plausible explanation behind the association between tumor size and oligometastasis invokes the host immune system. An increasing number of studies in melanoma and NSCLC have demonstrated that the response to ICB (anti–PD-1/PD-L1 and/or anti-CTLA4 monoclonal antibodies) is inversely correlated to tumor size. In metastatic melanoma, variable size cutoffs of <5 or <10.2 cm have been found to predict response (36, 37), and some of the greatest benefits of ICB have been found in the adjuvant setting in both melanoma and NSCLC, when the tumor burden is microscopic and therefore at its lowest (38, 39). Preclinical studies are also in line with this hypothesis: ICB response in melanoma is directly associated with the ratio of activated cytotoxic T cells to tumor burden. In fact, tumor burden in one study correlated with the ability of ICB to reinvigorate antitumor cytotoxic T cells (40). Though these results in aggregate are correlative rather than causal, they suggest a complex heterotypic tumor–host interaction that may influence the metastatic spectrum.

Interestingly, direct evidence of the interplay of the immune system and tumor burden is strongest in the setting of tumor number rather than tumor size. Using the ImmunoScore, which is a semiquantitative representation of CD3+ and CD8+ T-cell infiltration of colorectal tumor specimens that has been externally validated (41), researchers investigated outcomes in a cohort of patients with metastatic colorectal cancer treated with curative resection ± preoperative systemic therapy (42). Combining the ImmunoScore of the least infiltrated lesion with number of metastases resulted in improved predictions of outcome than number of metastases alone. Specifically, those patients with a high ImmunoScore and 1 to 3 metastases fared better than those with either a high ImmunoScore and ≥4 metastases or a low ImmunoScore. In the setting of the ORIOLE trial, greater peripheral baseline T-cell receptor clonality was associated with composite endpoint progression in the SBRT arm, and SBRT resulted in significantly more expanded clones than observation (20). Taken together, this suggests that host–tumor immune interactions can be harnessed in the setting of local ablation of oligometastatic disease (see Table 1).

Pace of progression

The potential influence of tumor immune-related signaling on the metastatic spectrum is further illustrated by a study investigating the pace of disease progression and other cancer-specific outcomes in colorectal cancer (43). Hepatic metastases from 134 patients with colorectal cancer treated with perioperative chemotherapy and surgical resection underwent detailed molecular characterization. Although primary tumor consensus molecular subtypes based on genomic analysis did not predict outcomes, a clustering analysis of parallel mRNA and miRNA data was able to stratify patients into three markedly different risk categories. When these biologically defined groups (canonical, immune, and stromal) were incorporated with a previously validated clinical risk score, they performed better than either classifier alone. The distant metastasis-free survival was 59 months versus 35 months versus 13 months in the low-risk, intermediate-risk, and high-risk composite groups, respectively, and metastatic recurrences were limited (≤3 additional metastases) in number in 100% of low-risk patients, 87% of intermediate-risk patients, and 34% of high-risk patients. Importantly, the prognostic import of these molecular subtypes was externally validated in a separate clinical cohort. Interestingly, two of the risk categories had alterations in immune signaling pathways, with the immune subtype constituting the majority of the low-risk group and demonstrating molecular signatures related to T-cell activation and interferon signaling (see Table 1).

Other groups are investigating different aspects of metastatic biology over time in large cohorts of patients and thereby are inherently incorporating the concept of DFI into their biologic definition of oligometastatic disease. The Tracking Cancer Evolution through Therapy (TRACERx) project is using genomic and proteomic sequencing to characterize the evolution of several solid tumors, including RCC. In a publication matching 575 primary RCCs with 335 metastases, metastatic propensity was measured as a function of both number and time (44). The investigators defined a rapid progression group (multiple sites of disease progression within 6 months of nephrectomy) and an attenuated progression group (single-site progression <6 months or multisite progression >6 months), quantifying their intratumoral heterogeneity and somatic copy-number alterations (SCNA). The attenuated progression group demonstrated high intratumoral heterogeneity, whereas the rapid progression group demonstrated low heterogeneity in combination with high levels of SCNA (see Table 1).

Detailed study of the attenuated progression group yielded further insights. The first was that patients in this group had metastatic disease that was amenable to control through further local therapy, whether surgical or radiotherapeutic. The second was that tumors within the attenuated progression group were enriched for specific evolutionary subtypes that involved PBRM1 mutations, further evidence that specific molecular aberrations can predict metastatic propensity in a clinical cohort. The third pertinent finding was that the course of these patients was one of initial oligometastatic progression followed by an increase in metastatic efficiency leading to disseminated metastases. This may suggest one reason why some series suggest that synchronous metastatic disease predicts for a worse outcome than metachronous disease: Perhaps the former patients simply have aggressive disease that has more in common biologically with a polymetastatic presentation rather than an oligometastatic one.

Epigenetic markers and specifically miRNA expression patterns are also linked with DFI and tumor number. Our group investigated a cohort of 63 patients with 1 to 5 metastases at time of treatment with lung metastasectomy (45). Patients were stratified into groups based on rate of progression following local therapy, with the low-rate group experiencing <0.6 new metastases per year, the intermediate-rate group experiencing 0.6 to 3.6 new metastases per year, and the high-rate group experiencing >3.6 new metastases per year. Downregulation of miRNAs associated with tumor-suppressor functions was found in the high-rate progression group and this was validated in a separate report (45). Other publications highlight the importance of the 14q32 miRNA locus, which is located in a region of the genome whose loss is associated with the metastatic phenotype in the TRACERx study discussed above (ref 46; see Table 1).

The optimization of inclusion criteria and endpoints for oligometastasis clinical trials

The major ongoing randomized clinical trials of locally ablative therapy to all metastases enroll patients primarily based on number of sites and generally limit this to either ≤3 or ≤5. Though such broad inclusion criteria are beneficial for rapid accrual, this results in a heterogeneous study population. For example, PCS IX, a phase II/III trial enrolling patients with castrate-resistant prostate cancer with 1 to 5 metastases, does not limit patients to those with bony metastases alone (NCT02685397). NRG LU002, a phase III trial enrolling patients with NSCLC with 1 to 3 metastases, does not limit patients to those without synchronous lymphatic metastases (NCT03137771). Given that expected 5-year OS can differ by approximately 15% between these two NSCLC patient groups and that local therapy is expected to have less impact on patients with polymetastatic prostate cancer with visceral disease, current inclusion criteria may dilute the benefits of local therapy and result in the premature rejection of the entire paradigm (15). These issues with patient heterogeneity are further amplified when studies enroll patients with multiple different cancer types. Indeed, the advisability of pooling data on disparate diseases with varying biology to make overarching statements regarding the role of local ablation in metastatic malignancy is unclear. Furthermore, ignoring the numerous clinical predictors of oligometastasis at both enrollment and progression limits our ability to gain further insight into the oligometastatic and oligoprogressive states.

A further limitation of current trial design is the use of endpoints not specifically tailored to the oligometastatic state. Given our current inability to accurately identify all sites of metastatic disease in most patients, the paradigm of local ablative therapy would be expected to result in modest but significant benefits in PFS and larger benefits in local control and OS. This is a pattern that has already been seen in the prospective phase II NSCLC trials and might be missed on larger randomized trials using PFS as the primary endpoint (18). It is also important to note that in many studies, patients frequently require and benefit from further local ablation to oligoprogressive disease. This may require the utilization of other clinical endpoints, such as metastasis-free survival or time to new metastasis. Finally, the inclusion of secondary endpoints focusing on biology (genomic, epigenetic, immunologic, etc.) is essential in order for the field to move to a combined clinicobiologic staging of metastatic disease (Fig. 1).

Figure 1.

Clinical and biologic determinants of oligometastasis.

Figure 1.

Clinical and biologic determinants of oligometastasis.

Close modal

The utilization of molecular imaging, liquid biopsy, and single-cell sequencing

Molecular imaging and liquid biopsy hold significant promise for identifying and tracking oligometastatic disease. In the setting of PET, an increasing number of useful tracers are in development (HER2, PD-L1, FAPI, etc.), demonstrate high diagnostic accuracy, and can potentially accurately identify oligometastatic patients (47–49). The utilization of PSMA PET in the ORIOLE trial is a case in point: Patients with metastatic prostate cancer who received SBRT to all PSMA-defined lesions experienced a distant metastasis-free survival of 29.0 versus 6.0 months in those who had PSMA-positive disease left untreated (20). Molecular imaging may also increase the number of patients diagnosed with oligometastatic disease and allow for the measurement of treatment response.

Liquid biopsy and single-cell sequencing are also important tools for the future (see reviews by De Michino and colleagues and Marine and colleagues, respectively, for detailed discussions; refs. 50, 51). The likely utility of liquid biopsy in identifying and tracking oligometastatic disease is suggested by several data points, including the observation that high tumor marker levels in the original colorectal cancer hepatectomy series predicted for worse clinical outcomes (12). Since then, studies across many tumor types have found reliable markers of both high initial tumor burden and residual disease, from EBV titers in nasopharyngeal carcinoma to circulating tumor cells in small-cell lung cancer (52, 53). Already validated cutoffs could serve as enrollment criteria for patients on oligometastasis trials, whereas the inclusion of such biomarkers in exploratory secondary endpoints could result in novel signatures with which to track metastatic disease. Single-cell sequencing, on the other hand, may eventually allow for the identification and prediction of mechanisms behind the therapeutic resistance that will invariably arise in many patients treated for oligometastatic disease (51, 54). This, in turn, may result in an improved ability to select patients for local ablation and/or combine this modality with targeted systemic therapies.

Local therapy and the role of systemic therapy for oligometastasis

As our ability to accurately diagnose oligometastatic disease increases and assuming the benefits of local therapy in this setting are borne out in the phase III data, an important question arises: What role will systemic therapy play in this setting? Truly oligometastatic disease that is treated with either surgery or SBRT, modalities that can potentially achieve high local control rates, should theoretically not require further therapy. Early data to support this contention were presented above: The 2.5 year freedom from new metastatic disease off all therapy following SBRT in oligometastatic prostate cancer (20). As increasingly effective risk classifiers are developed and if molecular imaging and liquid biopsy demonstrate no residual disease, should we subject patients to the toxicity of systemic therapy? Furthermore, even if such patients harbor residual disseminated tumor cells, these are by definition dormant and therefore unlikely to be eradicated with cytotoxic therapies. In fact, if immunosurveillance plays a role in keeping these nascent metastases in check, cytotoxic and in particular myelosuppressive therapies may actually have detrimental effects.

On the other hand, long-term follow-up of a subset of the trials discussed earlier suggests that local control and therefore the efficacy of local ablation may decrease over time. In particular, the updated results of the SABR-COMET trial after a median follow-up of 51 months demonstrate a local progression rate of 37% in SBRT-treated lesions (17). If this result is replicated as other trials mature, it suggests that SBRT alone may be insufficient to control disease. Perhaps it is the combination of SBRT and novel systemic therapies that will optimize outcomes. One interesting experimental approach is the combination of immunotherapy with partial-lesion SBRT for large metastases, as demonstrated on a recently published phase I clinical trial. Seventy-nine heavily pretreated patients with a variety of metastatic solid tumors received SBRT to two to four metastases followed by the administration of pembrolizumab (55). Importantly, metastases larger than 65 cubic centimeters received partial tumor irradiation and local control at 1 year did not differ between partially irradiated and completely irradiated lesions (89.5% overall; ref. 56). Several other early trials of novel systemic therapy–SBRT combinations, including tyrosine kinase inhibitors (NCT02893332) and immune checkpoint therapies have been published (57, 58). However, a number of questions—including the optimal sequencing of local ablation and systemic therapy—remain unanswered. Prospective trials will be necessary to shed light on these and other important questions.

The development of therapeutics that alter the metastatic phenotype

On the other end of the spectrum remains the pivotal issue regarding the best treatment approach for polymetastatic, rapidly progressive disease. The current approach is to apply successive lines of systemic therapy until resistant clones outcompete susceptible clones and drive further progression, a process that tends to accelerate with each new therapy. Though novel systemic therapies and in particular immunotherapies have shown significant promise in this setting, patients with high tumor burden in most histologic subtypes tend to have low response rates. An alternative approach is to develop therapies that decrease the virulence of the metastatic phenotype. Importantly, the current heavy utilization of PFS as an endpoint limits the ability to detect such effects and is another reason why novel endpoints in the metastatic disease setting should be investigated.

The actual existence of such therapeutics has been indirectly suggested by clinical observations of oligoprogression on targeted molecular therapies such as EGFR inhibitors in NSCLC (23). More direct support of the alteration of metastatic phenotype as a viable therapeutic strategy is provided by preclinical studies of epigenetic modifiers. In one series of experiments performed on the 14q32 locus miRNAs discussed above, investigators first discovered that transfection of metastatic cell lines with 14q32 locus miRNAs decreased metastatic lung colonization in murine models in vivo (59). The subsequent use of shRNAs to knockdown known 14q32 miRNA targets, led to reversal of a polymetastatic phenotype to an oligometastatic one. The group then demonstrated that DNA methylation controls the expression of 14q32-encoded miRNAs and treated HCT116 colorectal cells in vitro with decitabine, a commonly used antimetabolite in acute myelogenous leukemia and an inhibitor of DNA methylation. This induced 14q32 miRNA expression and abrogated the development of liver metastases in vivo without affecting cell viability (for a more in-depth analysis of the promise of epigenetic drugs in the treatment of solid tumors, see review by Morel and colleagues; ref. 60).

Other articles have recently demonstrated potential vulnerabilities in the molecular program regulating cancer metastasis. One series of experiments using 4-hydroxyacetophenone to activate non-muscle myosin-2C, alter actin organization, and inhibit the mechanical program of metastasis reduced metastatic burden in an in vivo model of colon cancer metastasis to the liver (61). Another set of experiments demonstrated the enhancement of CD44-mediated iron endocytosis during EMTs and found that interference with iron homeostasis using small-molecule inhibitors could alter epigenetic plasticity in tumor cells (62). Though this is preliminary work, the potential of such varied approaches and the subsequent possible long-term cancer-specific outcome benefit of combining these with locally ablative therapy should not be ignored.

The understanding of metastatic disease has increased in the 25 years since the oligometastatic state was first proposed. The original clinical predictors of oligometastatic disease continue to inform our investigations into the underlying biology of oligometastasis as well as current clinical trial design, though much remains to be learned. Safe and effective local ablation, new translational insights, and novel diagnostic technologies hold much promise. We believe that their creative application coupled with a bold willingness to question established paradigms in the treatment of metastatic disease will move the field forward even more rapidly in the 25 years to come.

S.P. Pitroda reports a patent for methods and kits for diagnosis and triage of patients with colorectal liver metastases pending. P.T. Tran reports grants and personal fees from RefleXion Medical and personal fees from Noxopharm, Janssen-Taris Biomedical, Myovant, and AstraZeneca outside the submitted work, as well as a patent for compounds and methods of use in ablative radiotherapy, and patent: 9114158 licensed to Natsar Pharm. R.R. Weichselbaum reports grants from Varian and Regeneron; other from Boost Therapeutics (stock), Immvira LLC (stock), Reflexion Pharmaceuticals (stock), Coordination Pharmaceuticals Inc. (stock), Magi Therapeutics (stock), Oncosenescence (stock), Aettis Inc. (consulting/advisory role), AstraZeneca (consulting/advisory role), Coordination Pharmaceuticals (consulting/advisory role), Genus (consulting/advisory role), Merck Serono S.A. (consulting/advisory role), Nano Proteagen (consulting/advisory role), NK Max America Inc. (consulting/advisory role), Shuttle Pharmaceuticals (consulting/advisory role), Highlight Therapeutics, S.L. (consulting/advisory role); and personal fees from AstraZeneca, Boehringer Ingelheim LTD, and Merck Serono S.A. outside the submitted work. No disclosures were reported by the other author.

The authors acknowledge the Ludwig Cancer Research Foundation and the Foglia Family Foundation for their funding support.

1.
Seyfried
TN
,
Huysentruyt
LC
. 
On the origin of cancer metastasis
.
Crit Rev Oncog
2013
;
18
:
43
73
.
2.
Einhorn
LH
,
Williams
SD
,
Troner
M
,
Birch
R
,
Greco
FA
. 
The role of maintenance therapy in disseminated testicular cancer
.
N Engl J Med
1981
;
305
:
727
31
.
3.
Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Early Breast Cancer Trialists' Collaborative Group
.
Lancet
1992
;
339
:
71
85
.
4.
Brawer
MK
. 
Hormonal therapy for prostate cancer
.
Rev Urol
2006
;
8
:
S35
47
.
5.
Rosenberg
SA
. 
Interleukin 2 for patients with renal cancer
.
Nat Clin Pract Oncol
2007
;
4
:
497
.
6.
Garon
EB
,
Hellmann
MD
,
Rizvi
NA
,
Carcereny
E
,
Leighl
NB
,
Ahn
MJ
, et al
Five-year overall survival for patients with advanced non–small cell lung cancer treated with pembrolizumab: results from the phase I KEYNOTE-001 study
.
J Clin Oncol
2019
;
37
:
2518
27
.
7.
Larkin
J
,
Chiarion-Sileni
V
,
Gonzalez
R
,
Grob
JJ
,
Rutkowski
P
,
Lao
CD
, et al
Five-year survival with combined nivolumab and ipilimumab in advanced melanoma
.
N Engl J Med
2019
;
381
:
1535
46
.
8.
Hellman
S
,
Weichselbaum
RR
. 
Oligometastases
.
J Clin Oncol
1995
;
13
:
8
10
.
9.
Mountain
CF
,
McMurtrey
MJ
,
Hermes
KE
. 
Surgery for pulmonary metastasis: a 20-year experience
.
Ann Thorac Surg
1984
;
38
:
323
30
.
10.
Foster
JH
. 
Survival after liver resection for secondary tumors
.
Am J Surg
1978
;
135
:
389
94
.
11.
Fong
Y
,
Fortner
J
,
Sun
RL
,
Brennan
MF
,
Blumgart
LH
. 
Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases
.
Ann Surg
1999
;
230
:
309
18
.
12.
Rees
M
,
Tekkis
PP
,
Welsh
FK
,
O'Rourke
T
,
John
TG
. 
Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a multifactorial model of 929 patients
.
Ann Surg
2008
;
247
:
125
35
.
13.
Pastorino
U
,
Buyse
M
,
Friedel
G
,
Ginsberg
RJ
,
Girard
P
,
Goldstraw
P
, et al
Long-term results of lung metastasectomy: prognostic analyses based on 5206 cases
.
J Thorac Cardiovasc Surg
1997
;
113
:
37
49
.
14.
Casiraghi
M
,
De Pas
T
,
Maisonneuve
P
,
Brambilla
D
,
Ciprandi
B
,
Galetta
D
, et al
A 10-year single-center experience on 708 lung metastasectomies: the evidence of the "international registry of lung metastases
.
J Thorac Oncol
2011
;
6
:
1373
8
.
15.
Ashworth
AB
,
Senan
S
,
Palma
DA
,
Riquet
M
,
Ahn
YC
,
Ricardi
U
, et al
An individual patient data metaanalysis of outcomes and prognostic factors after treatment of oligometastatic non–small cell lung cancer
.
Clin Lung Cancer
2014
;
15
:
346
55
.
16.
Palma
DA
,
Olson
R
,
Harrow
S
,
Gaede
S
,
Louie
AV
,
Haasbeek
C
, et al
Stereotactic ablative radiotherapy versus standard of care palliative treatment in patients with oligometastatic cancers (SABR-COMET): a randomised, phase 2, open-label trial
.
Lancet
2019
;
393
:
2051
8
.
17.
Palma
DA
,
Olson
R
,
Harrow
S
,
Gaede
S
,
Louie
AV
,
Haasbeek
C
, et al
Stereotactic ablative radiotherapy for the comprehensive treatment of oligometastatic cancers: long-term results of the SABR-COMET phase II randomized trial
.
J Clin Oncol
2020
;
38
:
2830
8
.
18.
Gomez
DR
,
Tang
C
,
Zhang
J
,
Blumenschein
GR
 Jr
,
Hernandez
M
,
Lee
JJ
, et al
Local consolidative therapy vs. maintenance therapy or observation for patients with oligometastatic non–small cell lung cancer: long-term results of a multi-institutional, phase II, randomized study
.
J Clin Oncol
2019
;
37
:
1558
65
.
19.
Ruers
T
,
Van Coevorden
F
,
Punt
CJ
,
Pierie
JE
,
Borel-Rinkes
I
,
Ledermann
JA
, et al
Local treatment of unresectable colorectal liver metastases: results of a randomized phase II trial
.
J Natl Cancer Inst
2017
;
109
:
djx015
.
20.
Phillips
R
,
Shi
WY
,
Deek
M
,
Radwan
N
,
Lim
SJ
,
Antonarakis
ES
, et al
Outcomes of observation vs stereotactic ablative radiation for oligometastatic prostate cancer: The ORIOLE phase 2 randomized clinical trial
.
JAMA Oncol
2020
;
6
:
650
9
.
21.
Ost
P
,
Reynders
D
,
Decaestecker
K
,
Fonteyne
V
,
Lumen
N
,
De Bruycker
A
, et al
Surveillance or metastasis-directed therapy for oligometastatic prostate cancer recurrence: a prospective, randomized, multicenter phase II trial
.
J Clin Oncol
2018
;
36
:
446
53
.
22.
Treasure
T
,
Farewell
V
,
Macbeth
F
,
Monson
K
,
Williams
NR
,
Brew-Graves
C
, et al
Pulmonary metastasectomy versus continued active monitoring in colorectal cancer (PulMiCC): a multicentre randomised clinical trial
.
Trials
2019
;
20
:
718
.
23.
Weickhardt
AJ
,
Scheier
B
,
Burke
JM
,
Gan
G
,
Lu
X
,
Bunn
PA
 Jr
, et al
Local ablative therapy of oligoprogressive disease prolongs disease control by tyrosine kinase inhibitors in oncogene-addicted non–small cell lung cancer
.
J Thorac Oncol
2012
;
7
:
1807
14
.
24.
Li
Y
,
Tang
ZY
,
Ye
SL
,
Liu
YK
,
Chen
J
,
Xue
Q
, et al
Establishment of cell clones with different metastatic potential from the metastatic hepatocellular carcinoma cell line MHCC97
.
World J Gastroenterol
2001
;
7
:
630
6
.
25.
Shindo-Okada
N
,
Takeuchi
K
,
Nagamachi
Y
. 
Establishment of cell lines with high- and low-metastatic potential from PC-14 human lung adenocarcinoma
.
Jpn J Cancer Res
2001
;
92
:
174
83
.
26.
Chiang
AC
,
Massagué
J
. 
Molecular basis of metastasis
.
N Engl J Med
2008
;
359
:
2814
23
.
27.
Nguyen
DX
,
Massagué
J
. 
Genetic determinants of cancer metastasis
.
Nat Rev Genet
2007
;
8
:
341
52
.
28.
Fidler
IJ
,
Poste
G
. 
The "seed and soil" hypothesis revisited
.
Lancet Oncol
2008
;
9
:
808
.
29.
Gonzalez
H
,
Hagerling
C
,
Werb
Z
. 
Roles of the immune system in cancer: from tumor initiation to metastatic progression
.
Genes Dev
2018
;
32
:
1267
84
.
30.
Lievens
Y
,
Guckenberger
M
,
Gomez
D
,
Hoyer
M
,
Iyengar
P
,
Kindts
I
, et al
Defining oligometastatic disease from a radiation oncology perspective: an ESTRO-ASTRO consensus document
.
Radiother Oncol
2020
;
148
:
157
66
.
31.
Iacobuzio-Donahue
CA
,
Fu
B
,
Yachida
S
,
Luo
M
,
Abe
H
,
Henderson
CM
, et al
DPC4 gene status of the primary carcinoma correlates with patterns of failure in patients with pancreatic cancer
.
J Clin Oncol
2009
;
27
:
1806
13
.
32.
Ahmed
S
,
Bradshaw
AD
,
Gera
S
,
Dewan
MZ
,
Xu
R
. 
The TGF-beta/Smad4 signaling pathway in pancreatic carcinogenesis and its clinical significance
.
J Clin Med
2017
;
6
:
5
.
33.
Wang
F
,
Xia
X
,
Yang
C
,
Shen
J
,
Mai
J
,
Kim
HC
, et al
SMAD4 gene mutation renders pancreatic cancer resistance to radiotherapy through promotion of autophagy
.
Clin Cancer Res
2018
;
24
:
3176
85
.
34.
Kuo
HT
,
Que
J
,
Lin
LC
,
Yang
CC
,
Koay
LB
,
Lin
CH
. 
Impact of tumor size on outcome after stereotactic body radiation therapy for inoperable hepatocellular carcinoma
.
Medicine
2017
;
96
:
e9249
.
35.
Parker
SM
,
Siochi
RA
,
Wen
S
,
Mattes
MD
. 
Impact of tumor size on local control and pneumonitis after stereotactic body radiation therapy for lung tumors
.
Pract Radiat Oncol
2019
;
9
:
e90
e7
.
36.
Joseph
RW
,
Elassaiss-Schaap
J
,
Kefford
R
,
Hwu
WJ
,
Wolchok
JD
,
Joshua
AM
, et al
Baseline tumor size is an independent prognostic factor for overall survival in patients with melanoma treated with pembrolizumab
.
Clin Cancer Res
2018
;
24
:
4960
7
.
37.
Robert
C
,
Ribas
A
,
Hamid
O
,
Daud
A
,
Wolchok
JD
,
Joshua
AM
, et al
Durable complete response after discontinuation of pembrolizumab in patients with metastatic melanoma
.
J Clin Oncol
2018
;
36
:
1668
74
.
38.
Antonia
SJ
,
Villegas
A
,
Daniel
D
,
Vicente
D
,
Murakami
S
,
Hui
R
, et al
Overall survival with durvalumab after chemoradiotherapy in stage III NSCLC
.
N Engl J Med
2018
;
379
:
2342
50
.
39.
Eggermont
AM
,
Chiarion-Sileni
V
,
Grob
JJ
,
Dummer
R
,
Wolchok
JD
,
Schmidt
H
, et al
Prolonged survival in stage III melanoma with ipilimumab adjuvant therapy
.
N Engl J Med
2016
;
375
:
1845
55
.
40.
Huang
AC
,
Postow
MA
,
Orlowski
RJ
,
Mick
R
,
Bengsch
B
,
Manne
S
, et al
T-cell invigoration to tumour burden ratio associated with anti–PD-1 response
.
Nature
2017
;
545
:
60
5
.
41.
Pages
F
,
Mlecnik
B
,
Marliot
F
,
Bindea
G
,
Ou
FS
,
Bifulco
C
, et al
International validation of the consensus immunoscore for the classification of colon cancer: a prognostic and accuracy study
.
Lancet
2018
;
391
:
2128
39
.
42.
Van den Eynde
M
,
Mlecnik
B
,
Bindea
G
,
Fredriksen
T
,
Church
SE
,
Lafontaine
L
, et al
The link between the multiverse of immune microenvironments in metastases and the survival of colorectal cancer patients
.
Cancer Cell
2018
;
34
:
1012
26
.
43.
Pitroda
SP
,
Khodarev
NN
,
Huang
L
,
Uppal
A
,
Wightman
SC
,
Ganai
S
, et al
Integrated molecular subtyping defines a curable oligometastatic state in colorectal liver metastasis
.
Nat Commun
2018
;
9
:
1793
.
44.
Turajlic
S
,
Xu
H
,
Litchfield
K
,
Rowan
A
,
Chambers
T
,
Lopez
JI
, et al
Tracking cancer evolution reveals constrained routes to metastases: TRACERxrenal
.
Cell
2018
;
173
:
581
94
.
45.
Lussier
YA
,
Khodarev
NN
,
Regan
K
,
Corbin
K
,
Li
H
,
Ganai
S
, et al
Oligo- and polymetastatic progression in lung metastasis(es) patients is associated with specific microRNAs
.
PLoS One
2012
;
7
:
e50141
.
46.
Uppal
A
,
Wightman
SC
,
Mallon
S
,
Oshima
G
,
Pitroda
SP
,
Zhang
Q
, et al
14q32-encoded microRNAs mediate an oligometastatic phenotype
.
Oncotarget
2015
;
6
:
3540
52
.
47.
Henry
KE
,
Ulaner
GA
,
Lewis
JS
. 
Human epidermal growth factor receptor 2-targeted PET/Single-photon emission computed tomography imaging of breast cancer: noninvasive measurement of a biomarker integral to tumor treatment and prognosis
.
PET Clin
2017
;
12
:
269
88
.
48.
Kratochwil
C
,
Flechsig
P
,
Lindner
T
,
Abderrahim
L
,
Altmann
A
,
Mier
W
, et al
(68)Ga-FAPI PET/CT: tracer uptake in 28 different kinds of cancer
.
J Nucl Med
2019
;
60
:
801
5
.
49.
Niemeijer
AN
,
Leung
D
,
Huisman
MC
,
Bahce
I
,
Hoekstra
OS
,
van Dongen
G
, et al
Whole body PD-1 and PD-L1 positron emission tomography in patients with non–small cell lung cancer
.
Nat Commun
2018
;
9
:
4664
.
50.
De Michino
S
,
Aparnathi
M
,
Rostami
A
,
Lok
BH
,
Bratman
SV
. 
The utility of liquid biopsies in radiation oncology
.
Int J Radiat Oncol Biol Phys
2020
;
107
:
873
86
.
51.
Marine
JC
,
Dawson
SJ
,
Dawson
MA
. 
Non-genetic mechanisms of therapeutic resistance in cancer
.
Nat Rev Cancer
2020
;
20
:
743
56
.
52.
Young
LS
,
Dawson
CW
. 
Epstein-Barr virus and nasopharyngeal carcinoma
.
Chin J Cancer
2014
;
33
:
581
90
.
53.
Tay
RY
,
Fernández-Gutiérrez
F
,
Foy
V
,
Burns
K
,
Pierce
J
,
Morris
K
, et al
Prognostic value of circulating tumour cells in limited-stage small-cell lung cancer: analysis of the concurrent once-daily versus twice-daily radiotherapy (CONVERT) randomised controlled trial
.
Ann Oncol
2019
;
30
:
1114
20
.
54.
Praktiknjo
SD
,
Obermayer
B
,
Zhu
Q
,
Fang
L
,
Liu
H
,
Quinn
H
, et al
Tracing tumorigenesis in a solid tumor model at single-cell resolution
.
Nat Commun
2020
;
11
:
991
.
55.
Luke
JJ
,
Lemons
JM
,
Karrison
TG
,
Pitroda
SP
,
Melotek
JM
,
Zha
Y
, et al
Safety and clinical activity of pembrolizumab and multisite stereotactic body radiotherapy in patients with advanced solid tumors
.
J Clin Oncol
2018
;
36
:
1611
8
.
56.
Luke
JJ
,
Onderdonk
BE
,
Bhave
SR
,
Karrison
T
,
Lemons
JM
,
Chang
P
, et al
Improved survival associated with local tumor response following multi-site radiotherapy and pembrolizumab: secondary analysis of a phase I trial
.
Clin Cancer Res
2020
;
26
:
6437
44
.
57.
Theelen
W
,
Peulen
HMU
,
Lalezari
F
,
van der Noort
V
,
de Vries
JF
,
Aerts
J
, et al
Effect of pembrolizumab after stereotactic body radiotherapy vspembrolizumab alone on tumor response in patients with advanced non–small cell lung cancer: results of the PEMBRO-RT phase 2 randomized clinical trial
.
JAMA Oncol
2019
;
5
:
1276
82
.
58.
Bauml
JM
,
Mick
R
,
Ciunci
C
,
Aggarwal
C
,
Davis
C
,
Evans
T
, et al
Pembrolizumabafter completion of locally ablative therapy for oligometastatic non–small cell lung cancer: a phase 2 trial
.
JAMA Oncol
2019
;
5
:
1283
90
.
59.
Oshima
G
,
Poli
EC
,
Bolt
MJ
,
Chlenski
A
,
Forde
M
,
Jutzy
JMS
, et al
DNA methylation controls metastasis-suppressive 14q32-encoded miRNAs
.
Cancer Res
2019
;
79
:
650
62
.
60.
Morel
D
,
Jeffery
D
,
Aspeslagh
S
,
Almouzni
G
,
Postel-Vinay
S
. 
Combining epigenetic drugs with other therapies for solid tumours—past lessons and future promise
.
Nat Rev ClinOncol
2020
;
17
:
91
107
.
61.
Bryan
DS
,
Stack
M
,
Krysztofiak
K
,
Cichoń
U
,
Thomas
DG
,
Surcel
A
, et al
4-Hydroxyacetophenone modulates the actomyosin cytoskeleton to reduce metastasis
.
Proc Natl Acad Sci U S A
2020
;
117
:
22423
9
.
62.
Muller
S
,
Sindikubwabo
F
,
Caneque
T
,
Lafon
A
,
Versini
A
,
Lombard
B
, et al
CD44 regulates epigenetic plasticity by mediating iron endocytosis
.
Nat Chem
2020
;
12
:
929
38
.
63.
Lussier
YA
,
Xing
HR
,
Salama
JK
,
Khodarev
NN
,
Huang
Y
,
Zhang
Q
, et al
MicroRNA expression characterizes oligometastasis(es)
.
PLoS One
2011
;
6
:
e28650
.