Abstract
Non-small cell lung cancer (NSCLC) is an aggressive malignancy with close to half of all patients presenting with metastatic disease. A proportion of these patients with limited metastatic disease, termed oligometastatic disease, have been shown to benefit from a definitive treatment approach. Synchronous and metachronous presentation of oligometastatic disease have prognostic significance, with current belief that metachronous disease is more favorable. Surgical excision of intracranial and extracranial oligometastatic disease has been shown to improve survival, especially in patients with lymph node-negative disease, adenocarcinoma histology and smaller thoracic tumors. Definitive radiation to sites of oligometastatic disease and initial thoracic disease has also been shown to have a similar impact on survival for both intracranial and extracranial disease. Recent studies have reported on the use of targeted agents combined with ablative doses of radiation in the oligometastatic setting with promising outcomes. In this review, we present the historical and current literature describing surgical and radiation treatment options for patients with oligometastatic NSCLC.
Introduction
Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related death, and nearly half of all patients with NSCLC present with metastatic disease [Raz et al. 2007; Siegel et al. 2015]. Outcomes for these patients are poor, with survival rates up to 29% at 1 year and a median survival of 4–6 months [Detterbeck et al. 2009; Cetin et al. 2011]. Within this metastatic population, there is a subset of patients with a limited number of sites of metastatic disease, termed oligometastatic disease, that can benefit from definitive management of their disease and can achieve long-term survival that we describe in this review. The concept of oligometastases was coined by Hellman and Weichselbaum, who hypothesized an intermediate clinical state between locoregional and widely metastatic disease. In this state, it was theorized that local treatment of all known disease sites may improve progression-free survival and overall survival, and potentially be curative in a select minority of patients [Hellman and Weichselbaum, 1995]. Today, oligometastatic disease is generally defined as four to five or fewer sites of metastatic disease.
The states of oligometastases that are typically described include synchronous and metachronous disease. Synchronous oligometastatic malignancies are those that initially present with a limited number of metastatic sites at the time of diagnosis. Metachronous oligometastatic malignancies are defined as those with limited metastatic disease sites that present after the initial diagnosis, usually after the primary disease site has been treated with curative intent. In the setting of NSCLC, oligometastases also include parenchymal disease in the contralateral lung, ipsilateral pleura and pericardium in addition to disease in other organs. In this report, we review the management options, risk factors, supporting literature and ongoing studies for the treatment of oligometastatic NSCLC. The primary treatment options for oligometastatic disease that are discussed include surgery for both intracranial and extracranial disease, and radiation therapy, with a focus on stereotactic body radiation therapy (SBRT). It is important to note that many of the earlier reviews focused primarily on the treatment of intracranial oligometastases, whereas more recent studies have started to emphasize the treatment of extracranial oligometastatic disease.
Surgery for oligometastatic disease
Surgical resection of intracranial and extracranial oligometastatic disease has been shown to have a positive effect on survival rates in patients with oligometastatic NSCLC. This has been established based on randomized data for patients with limited brain metastases and is also supported in retrospective studies of resection for metastases to a limited number of extracranial disease sites, including the adrenal glands and contralateral lung parenchyma. It is prudent to note that the definitive nature of this treatment approach inherently selects for patients with a high performance status; therefore, this approach may lead to better outcomes than those oligometastatic patients who are managed with non-surgical approaches. Additionally, those patients who are selected for surgical resection may not have tumors that are in close proximity to critical organs, which may not be the case in patients who do not undergo surgical resection.
Intracranial oligometastases
The Lung Cancer Study Group reported that 6.4% of first recurrences after NSCLC resection were isolated to the brain alone, accounting for about 20% of all lung cancer recurrences [Figlin et al. 1988]. The standard approach for the treatment of solitary brain metastases should include consideration of surgical resection [Tsao et al. 2012]. Additionally, patients with neurologic symptoms from synchronous brain lesions should first undergo local therapy to treat their intracranial disease in order to relieve or prevent neurologic complications.
Magilligan and colleagues reported the first series of patients who underwent resection of both a primary NSCLC with a synchronous solitary brain metastasis. These patients received whole brain radiation therapy (WBRT) with or without chemotherapy, in addition to their surgery and had an overall survival rate of 21% at 5 years and 15% at 10 years [Magilligan et al. 1976, 1986]. Patchell and colleagues conducted a randomized control trial comparing surgery and radiation with radiation alone for oligometastatic (single) brain metastases. In this study, 37 out of 48 patients with single brain metastases had NSCLC primaries. The surgical excision of single brain metastases followed by whole brain radiation resulted in a significantly prolonged median overall survival compared with whole brain radiation alone (40 weeks versus 15 weeks, p < 0.01), establishing the role of surgery for single brain metastases [Patchell et al. 1990].
An institutional review by Hu and colleagues reported comparable outcomes for NSCLC patients with solitary brain metastases with synchronous Stage I thoracic disease, managed with thoracic surgery to those patients with Stage I thoracic disease without brain metastases, with a 3-year overall survival rate of 50%. They suggested that this definitive management may not be justified in locally advanced disease with solitary brain metastases given the significantly shorter survival times for patients with brain metastases and otherwise stage II–III thoracic disease compared with otherwise stage I disease (p = 0.006) [Hu et al. 2006]. Similarly, Billing and colleagues suggested that craniotomy for synchronous brain metastases followed by resection of the primary lung cancer was beneficial in those patients with lymph node-negative disease. Using this approach, the authors reported a 5-year survival rate for patients with N0 disease of 35%, compared with no long-term survivors among patients with N1 or N2 lymph node disease [Billing et al. 2001]. Similarly, other series have reported a better prognosis for patients with N0 disease, adenocarcinoma histology, smaller thoracic tumor size (T1 versus T3), lobectomy and a complete pulmonary resection [Mussi et al. 1996; Bonnette et al. 2001; Granone et al. 2001; Collaud et al. 2012].
In addition to examining nodal status and histology as prognosticators for survival, Iwasaki and colleagues analyzed oncologic systemic markers in patients who underwent resection for synchronous brain metastases. They found that no patient with an elevated level (>4.0 ng/ml) of carcinoembryonic antigen (CEA) was alive at 3 years, compared with a 3-year survival rate of 40% in those with a normal CEA level [Iwasaki et al. 2004].
Collectively, the data suggest that surgical resection of oligometastatic brain metastases results in long-term survival rates of up to 37% (Table 1). Certain patient factors may offer prognostic information when deciding to pursue a surgical approach for individual patients. Additionally, patient performance status must be carefully considered for this definitive approach in order to minimize the morbidity of treatment.
Summary of surgical resection for intracranial oligometastases from non-small cell lung cancer.
WBRT, whole brain radiation therapy; OS, overall survival; S, synchronous; M, metachronous; NR, not reported.
Included extracranial oligometastatic sites.
Extracranial oligometastases
Although the brain is the most common site of metastatic disease in NSCLC, extracranial sites are also common areas of oligometastatic spread, including the adrenal glands, extrathoracic lymph nodes, contralateral lung and bone. Several reported series demonstrate that definitive surgical management of patients with isolated extracranial metastatic sites can result in long-term survival for oligometastatic NSCLC (Table 2).
Summary of surgical resection for extracranial oligometastases from non-small cell lung cancer.
WBRT, whole brain radiation therapy; OS, overall survival; S, synchronous; M, metachronous; NR, not reported.
Mean survival reported, not median survival.
Also included intracranial oligometastatic disease.
Includes 15 patients with T4 (ipsilateral lobe nodule according to AJCC 7th ed.), which was considered metastatic disease at time of individual study publication based on AJCC 6th ed.
The difference in outcomes between synchronous and metachronous is conflicting when examining individual case series; however, the current sense is that metachronous disease is more favorable than synchronous disease. Porte and colleagues reported no difference in median survival between patients who presented with synchronous and metachronous disease [Porte et al. 2001]. However, Mercier and colleagues reported that all patients with a disease-free interval of less than 6 months (synchronous metastasis) were not alive within 2 years of surgical resection [Mercier et al. 2005]. A similar trend of lower survival in synchronous oligometastatic patients was shown by Pfannschmidt and colleagues [Pfannschmidt et al. 2005]. Tanvetyanon and colleagues compiled a review of 10 retrospective series with a total of 114 patients with oligometastatic NSCLC to the adrenal gland. When comparing synchronous (n = 48) with metachronous (n = 66) disease, they found that median overall survival was significantly shorter for patients with synchronous metastases compared with those with metachronous metastases (12 months versus 31 months, p = 0.02). However, the 5-year survival estimates were equivalent at 26% and 25%, respectively [Tanvetyanon et al. 2008].
The extent of resection varies depending on the site of metastatic disease. For adrenal metastases, two case series have reported both the safety and efficacy of laparoscopic surgery compared with open surgery [Lucchi et al. 2005; Strong et al. 2007]. There was no difference in local recurrence, disease-free interval or 5-year overall survival between the surgery types. However, laparoscopic surgery resulted in lower estimated blood loss (106 cc versus 749 cc, p < 0.0001), shorter hospital stays (2.8 days versus 8.0 days, p < 0.0001) and fewer total complications (4% versus 34%, p < 0.0001) [Strong et al. 2007]. In contralateral lung metastases, two separate series have reported no difference in survival but potentially more toxicity with a lobectomy or pneumonectomy compared with a sublobar resection [De Leyn et al. 2008; Voltolini et al. 2010]. In addition to showing no difference in survival based on the type of surgery, Voltolini and colleagues also demonstrated that nodal status impacts survival. In their series, patients with node-negative cancer had a 5-year survival rate of 57% compared with 0% for node-positive patients [Voltolini et al. 2010].
The benefit of PET/CT in the staging and subsequent treatment planning of NSCLC has been shown in a number of studies [Nestle et al. 2006; Kligerman and Digumarthy, 2009; MacManus et al. 2009]. The concept of stage migration has become more evident with the use of PET/CT. Disease progression between scans or between the initial scan and the start of treatment has been documented to occur in 21–48% of patients, frequently resulting in major treatment modifications [O’Rourke and Edwards, 2000; Everitt et al. 2010; Mohammed et al. 2011]. In a recent study by Geiger and colleagues, 14 patients out of 47 (30%) were found to have evidence of extrathoracic metastatic disease and 10 patients (21%) had new nodal disease. At a scan interval of 20 days, the rate of upstaging was 17% [Geiger et al. 2014]. The routine use of PET/CT will inevitably lead to the upstaging of some newly-diagnosed NSCLC patients and will declare a portion of these patients to be oligometastatic.
Numerous surgical series have demonstrated long-term survival rates ranging from 23–86% with surgical resection of extracranial oligometastatic NSCLC. Resection of synchronous metastatic disease may achieve lower survival rates than for metachronous disease that can help to decide whom to treat, definitively. Additionally, the efficacy of less invasive surgery in the resection of adrenal and contralateral lung oligometastatic disease does not appear to affect outcomes and may allow for fewer complications. Similar to intracranial oligometastatic disease, patient performance status should be carefully considered in order to maximize the benefit of this definitive treatment approach.
Radiotherapy for oligometastases
Advances in radiotherapy have allowed for more conformal treatments and dose escalation, which in turn have been shown to reduce toxicities while improving local control. SBRT, also termed stereotactic ablative radiotherapy (SABR), allows for the delivery of highly conformal and ablative doses, and it can be a curative treatment option for many disease sites, including early stage NSCLC in inoperable patients [Baumann et al. 2009; Timmerman et al. 2010]. Kavanagh and colleagues hypothesized that the ablative effect of SBRT would eliminate large volumes of tumor that may delay disease progression beyond the threshold of lethality [Kavanagh et al. 2007]. Data that extend this radioablative technique to the oligometastatic setting continue to emerge and are summarized here.
Definitive radiotherapy
The radiation dose and fractionation schemes utilized have varied considerably in the series summarized in this review that may account for the wide range in local control achieved with this approach. Salama and colleagues reported the first dose-escalation study using SBRT for multiple sites of extracranial oligometastases. A total of 61 patients received treatment to one to five metastatic lesions from 24 Gy to 48 Gy. All doses were delivered in three fractions separated by at least 48 hours each. NSCLC was the most common primary histology (18%) and the lung was the most common site of treatment (36%). A dose of 42 Gy was achieved without dose-limiting toxicities in the lung, abdomen and liver. Additionally, two patients with lung lesions and one patient with an abdominal lesion were treated to 48 Gy without dose-limiting toxicity. Overall survival at 1 and 2 years was 81.5% and 56.7%, respectively [Salama et al. 2012].
Multiple series have shown a survival advantage in patients presenting with metachronous oligometastatic disease who are treated with definitive radiation therapy. Flannery and colleagues reported patients with solitary brain metastases treated with stereotactic radiosurgery had a significant decrease in median survival for synchronous metastatic disease compared with metachronous disease presentation (8.6 months versus 33.3 months, p = 0.001) [Flannery et al. 2003]. This has subsequently been corroborated in a larger series of patients with solitary brain oligometastases [Mariya et al. 2010]. Patients with extracranial sites of synchronous oligometastases also fared worse. The University of Rochester reported a 5-year overall survival rate of 14% for metachronous patients but did not observe any 5-year survivors among synchronous patients (p = 0.003). All metastatic disease sites in that study were treated with SBRT [Cheruvu et al. 2011]. A recent meta-analysis of 757 patients with oligometastatic NSCLC showed a survival advantage in those with metachronous versus synchronous disease, with a hazard ratio (HR) of 1.96 (p < 0.001). In this study, patients were treated with ablative therapy to all sites of oligometastatic disease [Ashworth et al. 2014].
Although synchronous metastatic presentation generally portends a lower survival, several retrospective series suggest that definitive management of thoracic disease in these patients may provide a survival benefit. One of the first series to report definitive radiation therapy in the treatment of oligometastatic NSCLC was a study from Rush University, which described a cohort of patients who received either definitive concurrent chemotherapy and radiation or surgery for their thoracic disease, in addition to definitive radiation to their sites of metastatic disease. They reported a median progression-free survival of 12 months and median survival of 20 months. Additionally, seven patients were disease free at last follow up [Khan et al. 2006]. More recent series have also demonstrated prolonged survival for synchronously metastatic patients who have received definitive thoracic treatment (DTT) with chemotherapy for primary lung cancers with mediastinal nodal involvement [Jabbour et al. 2011; De Ruysscher et al. 2012; Collen et al. 2014; Xanthopoulos et al. 2015]. Comparing outcomes for patients with DTT with those who did not receive such therapy has revealed drastic survival differences. Gray and colleagues found that patients with DTT (surgical resection or radiation dose > 45 Gy) had a greater median survival compared with those who did not receive DTT (26.4 months versus 10.5 months) [Gray et al. 2014]. Similarly, a review by Parikh and colleagues of 186 patients with oligometastatic NSCLC found that those patients who received definitive therapy to their primary disease had prolonged survival compared with those who did not receive treatment to their primary tumor (HR 0.65, p = 0.043). The median overall survival for each subset was 19 months and 16 months, respectively [Parikh et al. 2014].
The noninvasive nature of SBRT allows for the potential to treat multiple oligometastatic disease sites more safely compared with surgical resection. The number of oligometastases and organs involved has prognostic significance, even within the commonly accepted definition of one to five lesions. An analysis of patients enrolled on the Southwest Oncology Group metastatic NSCLC protocols revealed an improved median survival for solitary single-organ metastases compared with multiple metastases in a single organ, and with multiorgan involvement (8.7 months versus 6.2 months versus 5.1 months, respectively, p ⩽ 0.01) [Albain et al. 1991]. In a separate analysis of baseline prognostic factors in the phase III FLEX trial, longer median survival times were seen in those patients with metastases to one site compared with two sites and compared with three or more sites (12.4 months versus 9.8 months versus 6.4 months, respectively, p < 0.001) [Pirker et al. 2012]. The case series presented in this review have limited the treatment of oligometastatic disease to five sites and have median survival periods ranging from 9–26.4 months (Table 3).
Summary of non-palliative radiation for oligometastases from non-small cell lung cancer.
CNS, central nervous system; RT, radiotherapy; PFS, progression-free survival; NR, not reported; DTT, definitive thoracic therapy.
Similar to surgical series, nodal status impacts survival in oligometastatic NSCLC patients treated with radiotherapy. De Rose and colleagues reported a drastically improved median survival in patients with node-negative disease (55 months) compared with patients with node-positive disease (32.1 months, p = 0.01). The 5-year survival rates for these subsets were 45.5% and 8.2%, respectively (p = 0.01) [De Rose et al. 2016]. Similarly, Ashworth and colleagues reported in their analysis a survival benefit in node-negative patients compared with patients with N1 or N2 disease, with HRs of 1.59 (p = 0.01) and 1.72 (p < 0.001), respectively [Ashworth et al. 2014].
Radiotherapy concurrent with targeted agents
The effectiveness of targeted agents for locally advanced and metastatic NSCLC and their ability to improve clinical outcomes has been reported in recent studies [Shepherd et al. 2005; Mok et al. 2009; Zhou et al. 2011; Rosell et al. 2012]. The combination of these agents with ablative radiotherapy was first examined by Weickhardt and colleagues. They reviewed metastatic NSCLC patients with either anaplastic lymphoma kinase gene arrangements (ALK+) or epidermal growth factor receptor mutations (EGFRm) treated with crizotinib or erlotinib, respectively, who received local ablative therapy to sites of oligometastases. Most patients were treated with SBRT to a median dose of 40 Gy to extracranial disease sites. Median progression-free survival for the entire cohort was 9.8 months (9 months for ALK+ patients, 12 months for EGFRm patients) [Weickhardt et al. 2012]. In the setting of NSCLC, patients with targetable mutations or translocations being treated with a targeted therapy and progressing in one or a limited number of disease sites, termed oligoprogression, SBRT may allow for the eradication of these progressing sites that may have potentially lost their targetable mutation or translocation. Doing so can allow these patients to remain on their existing targeted therapy, as opposed to switching to a second-line target therapy or starting cytotoxic chemotherapy [Camidge et al. 2014].
Iyengar and colleagues conducted the only reported phase II trial of a targeted agent combined with SBRT to treat extracranial oligometastatic NSCLC patients who progressed through first-line systemic therapy. A total of 24 patients were included on this trial and a sum of 52 metastatic sites were treated, with no more than five sites treated per patient. Erlotinib was administered, despite not knowing the EGFR mutation status of all patients, during SBRT, and continued until disease progression. Median progression-free survival was 14.7 months and median overall survival was 20.4 months. Interestingly, none of the 13 patients tested had a targetable EGFR mutation, so it was thought that the erlotinib provided little systemic benefit [Iyengar et al. 2014].
Future considerations
Current trials are focusing on comparing localized ablative therapy with conventional palliative radiation therapy, as well as the combination of localized ablative therapy and targeted agents for the treatment of oligometastatic NSCLC. The ATOM study [ClinicalTrials.gov identifier: NCT01941654] is a single-arm phase II study that is investigating if local ablative therapy will improve 1-year progression-free survival rates in patients with a good partial response to an EGFR tyrosine kinase inhibitor and up to four positron emission tomography (PET)-avid sites representing residual disease. The SABR-COMET study [ClinicalTrials.gov identifier: NCT01446744] is a randomized phase II trial comparing SABR to standard of care (palliation) in patients with up to five metastatic lesions with a primary outcome of overall survival. Specific dose fractionations have been described in the trial protocol based on tumor location and size. An ongoing NRG (National Surgical Adjuvant Breast and Bowel Project, RTOG, and Gynecologic Oncology Group) study (NRG-BR001) is looking to determine the recommended SBRT dose for up to four metastatic locations being treated for oligometastatic NSCLC, prostate cancer and breast cancer. Finally, the SARON study [ClinicalTrials.gov identifier: NCT02417662] is a randomized phase III trial comparing chemotherapy alone with chemotherapy combined with radiation therapy to the primary site (conventional or SBRT) and oligometastatic sites (SBRT) for patients with oligometastatic NSCLC. The primary endpoint for this trial is overall survival.
Conclusion
The definitive management of a select group of metastatic NSCLC patients has been shown to prolong survival in many studies. These oligometastatic patients benefit from definitive management of both their primary tumors and sites of metastatic disease, both surgically and with definitive radiation therapy. The use of SBRT to treat sites of oligometastatic disease has been shown in recent reviews to provide high rates of local control. The incorporation of upfront systemic therapy may help to select patients who do not progress and may be more likely to benefit from local therapies. Furthermore, synchronous versus metachronous presentation, nodal status, and number and location of oligometastatic sites have all been shown to impact survival in this subset of patients. Additionally, the use of targeted agents concurrently with SBRT has shown promising results in both retrospective and prospective studies. It is prudent to note that the current treatment paradigm is based mainly on retrospective data, which has its inherent limitations. Ongoing trials investigating the use of targeted agents with or without chemotherapy combined with SBRT are currently accruing and will hopefully provide more definitive evidence for the treatment of oligometastatic NSCLC.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Conflict of interest statement
The authors declare that there are no conflicts of interest.
