Epidermal growth factor receptor (EGFR) mutations occur in almost 10–15% of patients affected by non-small cell lung cancer (NSCLC) in the western world and in 30–40% of the Asiatic population [1, 2]. These types of mutations are more common in non-smokers and females. At time of diagnosis, about 50% of patients present with advanced stage NSCLC, sometimes with an oligometastatic situation [3, 4]. The development of “targeted therapy” such as EGFR tyrosine kinase inhibitors (TKIs) dramatically improved the most important clinical outcomes when compared to historical doublet chemotherapy (CHT) [5]. EGFR TKIs are actually considered the standard first line therapy in patients with activating mutations of the EGFR exons 18–21. However, resistance to EGFR TKIs usually occurs after 9–13 months of first line with first-generations or second-generations agents. The most frequent tumor acquired alteration is represented by the T790M mutation that reduces the link of first-generation and second-generation TKIs to the ATP-binding pocket of EGFR with consequent progression of disease [6]. More recently, according to the results obtained in the phase III randomized FLAURA trial, osimertinib should be used as first-line systemic therapy regardless of T790M mutation status [7, 8] as it achieved a significant improvement in terms of both progression free survival (PFS) and even overall survival (OS).

In vitro and in vivo studies investigated the increasing sensitivity to radiotherapy (RT) of EGFR-TKIs. Ionizing radiation activates EGFR and mitogen-activated protein kinases (MAPK)-extracellular signal-regulated kinase (ERK) and phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) pathways causing the promotion of cellular proliferation and repair of radiation induced.

DNA damage through homologous and nonhomologous recombination [9]. The repeated exposition to radiation increased EGFR expression [9, 10]. EGFR-TKIs, such as gefitinib and erlotinib had a radiosensitizing effects at different levels: apoptosis, cell cycle arrest, repopulation and DNA damage [11]. EGFR-TKIs, combined with radiation, promote the reduction in S-phase fraction, the most radioresistant cell phase, increasing the cells number in G1 phase and G2 phase and enhance apoptosis induction [10].

Local treatment (LT), such as surgery, RT or thermal ablation, of primary or metastatic lesions in oligometastatic or oligoprogressive disease has been shown to obtain clinical benefits in EGFR mutated-NSCLC [12, 13]. The European Society for Medical Oncology (ESMO), suggest administering LT, such as high-dose RT or surgery, to no more than 3 synchronous metastatic sites in the oligometastatic setting after multidisciplinary discussion [13]. However, the management of oligoprogression is controversial due to the limited clinical data, and involves mainly radical LT, such as stereotactic body RT (SBRT) or surgery, to achieve disease control. In EGFR-mutant advanced NSCLC patients a common approach for treating oligoprogression is to continue the EGFR-TKI (that is controlling the greater proportion of the disease), while using local ablative therapy to eradicate the resistant clones in the area or areas of radiological progression. This therapeutic approach is supported by several, even if mainly retrospective, case series [12].

The concept of adding LT to systemic therapy comes from the idea that a metastasis directed LT approach may prolong the overall duration of the systemic treatment efficacy with a possible impact on clinical outcomes in select subgroups as already demonstrated in different solid tumors [14–16]. In the stereotactic ablative RT for the comprehensive treatment of oligometastases (SABR-COMET) basket trial patients with stage IV oligometastatic malignant neoplasms in whom controlled primary tumor and less than 5 metastases were diagnosed, SBRT to secondary lesions significantly improved median OS (28 months vs. 41 months). The treatment related toxicity was also satisfactory with grade 2 or superior side effects occurring in only 29% of patients; relevant grade 5 toxicity was reported only in 3% of the study population [17]. In a phase II trial, Iyengar et al. [18] evaluated the role of SBRT plus maintenance systemic therapy vs. maintenance CHT alone in patients with oligometastatic NCSLC (defined as six extracranial lesions, including primary site and less of 3 liver or lung lesions). The PFS was 9.7 months in the SBRT arm and 3.5 months after CHT alone without significative differences in terms of toxicities [18].

The aim of this literature overview is to analyze the role of the combination of metastasis directed therapy, focusing on RT, in combination with TKIs in patients with EGFR mutated oligometastatic and/or oligoprogressive advanced NSCLC.

For the first time in 1995, Hellman et al. [19] introduced the concept of a new “intermediate” clinical scenario between localized and advanced stage, defined as “the oligometastatic disease”: it is a clinical stage occurring during the natural clinical progression from a localized primary tumor to disseminated metastatic disease. For the authors, LTs focused on the few sites of oligometastatic localization were potentially able to improve clinical outcomes.

More recently, the European Society for Radiotherapy and Oncology (ESTRO) and OligoCare European Organisation for Research and Treatment of Cancer (EORTC) project identified several different scenarios [20]:

1)

“De novo” oligometastatic disease: patients without other prior diagnosis of oligometastatic disease which is considered “synchronous” if the time interval between diagnosis of primary tumor and metastases is less than six months or “metachronous” if it is superior than six months.

The number of metastases needed to identify the “oligometastatic” stage is still debated. Lievens et al. [21] identified a number of 5 different metastatic sites as the upper boundary for defining a clinical stage as “oligometastatic”. For other authors, the definition of oligometastatic disease is not only a matter of number (of lesions), but also of limited regions [22] or organs [19] involved.

Identifying a clinical tumor stage as “oligometastatic” requires optimal imaging represented by 18-fluoro-deossi-glucose positron emission tomography (18-FDG PET) together with brain imaging (contrast enhancement magnetic resonance or computed tomography). This kind of imaging seems to detect those patients with more favorable prognosis, defined as patients with less than 5 site of metastases [3]. Regarding the localization of systemic lesions, brain, lung, adrenal gland, bone, liver and lymph nodes represent the most frequent sites of metastases [23].

Moving to the definition of the “oligoprogressive” scenario, the type and timing of detection of new or progressive lesions is the characterizing feature. It may be represented by single or few sites of progression, while the primary tumor is stable during treatment with systemic therapy (CHT, chemo-immunotherapy or immunotherapy alone) [24]. According to the ESMO clinical practice guidelines, oligoprogression in case of EGFR mutated disease can be defined as the progression of a small number of metastatic sites with stability of all the other systemic lesions during target therapy treatments [1]. In this scenario, it is supposed that a population of tumor cells acquired a sort of biological resistance to systemic therapy that can be classified as: receptor alteration, acquisition of bypass signaling pathway and phenotypic alterations [25].

Yang et al. [26] defined different patterns of progression: dramatic symptomatic progression with an increase of lesion size of more than 20% in less than 6 months; gradual progression if the volume of the metastatic lesions increases more than 20% in more than 6 months; local progression in case of a single extracranial progression. Depending on the type of progression, some observational studies seem to suggest the use of SBRT to eradicate resistant tumoral cell clones without discontinuation of TKIs with the aim of extending the same systemic therapy duration [1, 27].

EGFR-TKIs are the major first line therapy for patients with advanced NSCLC harboring activating EGFR mutation based on several phase III clinical trials exploring the efficacy of first, second and third generation EGFR-TKIs.

Gefinitib, erlotinib and icotinib are low molecular weight, reversible, oral first-generation EGFR inhibitors [28]. While gefinitib and erlotinib are Food and Drug Administration (FDA) and European Medicines Agency (EMA)-approved, icotinib is approved only in China. Both gefinitib and erlotinib have anilinoquinazoline-based structures and serve as ATP-competitors [29]. The second-generation EGFR-TKIs, including afatinib and dacomitinib, are irreversible pan-human EGFR/HER-1, HER-2 and HER-4 (ErbB) inhibitors which covalently bind to EGFR [30]. A broader inhibitory profile on the ErbB receptor family and a more robust inhibition of downstream signaling distinguish second generation TKIs from first generation EGFR-TKIs [31].

Several phase III trials have explored the efficacy of gefitinib (NEJ002, WJTOG-3405, IFUM, first-SIGNAL), erlotinib (EURTAC, OPTIMAL, ENSURE) and afatinib (LUX-Lung 3, LUX-Lung 6) [32]. A significant clinical benefit was observed for all the compounds in comparison with standard platinum-based CHT, both in terms of PFS, objective response rate (ORR) and disease control rate (DCR). In this regard, EGFR-TKIs granted a median PFS of 8.0–13.1 months and ORR of 62.1–75%, which was significantly better compared to standard platinum-based CHT [33].

Dacomitinib was compared to erlotinib and gefitinib in the “Advance Research for Cancer for targeted pan- HER therapy (ARCHER)” 1009 and ARCHER 1050 trials, respectively. Although dacomitinib showed more efficacy, the problematic safety profile limited the use in clinical practice [34, 35].

Outcomes of EGFR mutant metastatic NSCLC improved significantly with the introduction of third generation TKIs. Osimertinib, an oral irreversible EGFR-TKI, was compared to standard first-generation and second-generation EGFR-TKIs as a first-line treatment in the FLAURA trial. Remarkably, osimertinib was superior to both erlotinib and gefitinib in terms of PFS [18.9 months vs. 10.2 months, hazard ratio (HR, 0.46; P < 0.001)] [7], and OS [38.6 months vs. 31.8 months, HR 0.799; 95.05% confidence interval (CI), 0.641–0.997; P = 0.0462] [8]. Compared to early generation TKIs, osimertinib demonstrated a considerable survival benefit and a higher capacity to penetrate the blood brain barrier resulting in further central nervous system (CNS) response rates > 60%. Hence, as a result, osimertinib was approved in 2018 by both the FDA and EMA for the first-line treatment of NSCLC with activating EGFR mutations and it constitutes the standard treatment in this setting to date.

The use of first-generation and second-generation EGFR-TKIs is now limited in first line therapy, except afatinib which demonstrated a great activity against uncommon EGFR point mutations or duplications in exons 18–21, as showed in the post hoc analysis of LUX-Lung 02, LUX-Lung 03 and LUX-Lung 06 trials [36].

Despite EGFR-TKIs having led to a dramatic improvement in clinical outcomes in EGFR mutated NSCLC, the majority of patients eventually experience resistance to EGFR-TKIs, regardless of the lines of treatment.

The optimal second-line therapy for patients who progressed to first-line EGFR inhibitors depends mainly on the molecular mechanism of resistance. To offer the optimal therapy, a molecular recharacterization by tissue or liquid biopsy is strongly recommended in patients in which EGFR-TKIs failed as first-line treatment [37].

The most common resistance mechanism to first and second-generation EGFR-TKIs is the T790M mutation, occurring in up to 50–60% of cases. Osimertinib is the best treatment choice in this setting [38]. AURA 3 was a phase III clinical trial exploring osimertinib in patients having experienced progression disease after first-generation or second-generation EGFR-TKIs due to the onset of T790M resistance mutation. Compared to platinum plus pemetrexed CHT, osimertinib achieved a PFS of 10.4 months vs. 4.4 months (HR 0.30, P < 0.001) and an ORR of 71% vs. 31% [38]. In addition, osimertinib improved survival at 24 months and 36 months vs. CHT (55% vs. 43% and 37% vs. 30%, respectively), with a longer median OS (26.8 months vs. 22.5 months, respectively) [39]. Consequently, FDA and EMA approved osimertinib for the treatment of patients with metastatic EGFR-mutated NSCLC with an acquired T790M mutation after progression on previous EGFR-TKI therapy.

The mechanisms of resistance to osimertinib, regardless of the line of treatment, are various and are still under investigation. They can be classified in two different groups: EGFR dependent mechanisms and EGFR independent mechanism [37]. Among the first group, the C797S mutation is the most common occurring in 10–26% and 7% of patients treated with osimertinib as second-line and first-line treatment, respectively [38, 40]. In absence of a T790M mutation and in presence of a C797S mutation, patients may be treated with early EGFR inhibitors [41, 42]. When C797S appears in presence of a T790M mutation, the allelic context in which C797S is acquired with respect to T790M has potential implication for treatment. If C797S is found concomitantly (in trans) with T790M mutation, a combination of first-generation and third-generation EGFR-TKIs could be a valid choice. On the contrary, if the mutations are in cis, a resistance to all available EGFR-TKIs was reported and CHT is the best choice [43].

Moreover, other uncommon tertiary EGFR mutations (e.g., G796S, L792H, G724S, and L718X) that confer resistance to osimertinib, could be treated with early-generation EGFR-TKIs in the absence of T790M mutation [44].

The EGFR independent mechanisms of resistance to osimertinib include alternative signaling pathway activation or histological transformation.

Mesenchymal-epithelial transition (MET) amplification has been identified as the most common EGFR-independent mechanism of osimertinib resistance, accounting for 5–24% of cases [45]. A feasible choice to overcome the resistance driven by MET amplification is the combination of osimertinib with a MET inhibitor. Indeed, crizotinib in association with osimertinib showed clinical and radiological benefit [46, 47]. Other promising MET inhibitors in association with osimertinib are now under investigation [48]. Alterations in rat sarcoma (RAS)/MAPK and PIK3/AKT pathways, cell cycle gene alteration and oncogenic fusions have been identified as other mechanism of resistance [45].

After the failure of the treatment with osimertinib, novel agents such as human EGFR-3 (HER-3) or trophoblast cell surface antigen-2 (TROP-2) directed antibody-drug conjugates (ADC) exhibited promising results. Patritumab deruxtecan, an HER-3 ADC, or datopotamab deruxtecan, a TROP-2 targeting ADC, are being investigated for patients who were previously treated with osimertinib [48].

The transformation from adenocarcinoma to small cell lung cancer (SCLC) and squamous cell carcinoma (SCC) has been described after treatment with third-generation EGFR-TKIs like osimertinib as first-line or second-line [49–52]. Identification of histological transformation is imperative, due to the therapeutic and prognostic implications.

In absence of druggable mechanisms of resistance, CHT remains the standard of care. The addition of immune-check point inhibitors (ICIs) and anti-angiogenetic drugs was investigated in a subgroup of patients included in the IMpower150 trial, showing survival benefits in terms of both PFS and OS in the atezolizumab, bevacizumab, paclitaxel and carboplatin groups compared with the bevacizumab, paclitaxel and carboplatin ones [53, 54].

Finally, ICI monotherapy showed little clinical benefit in EGFR mutated NSCLC, thus its use in clinical practice is not recommended.

Oligometastatic disease is often approachable with local ablative treatments (LATs), such as SBRT or surgery. A retrospective study by Ashworth et al. [23] found a median 5-years OS of 29.4%, in patients treated on both the primary tumor and the metastatic lesions with radical intent, despite of type of treatment (surgery or SBRT).

Different characteristics should be considered in patients that are candidates for LATs: age, performance status, comorbidities, time of appearance of metastases (synchronous vs. metachronous), number of metastases, site of lesions (a better prognosis seems to be achievable in case of brain, lung or adrenal gland metastases), extension of primary tumor, mediastinal lymph node involvement (negative lymph nodes have a better prognosis) [55].

Park et al. [56] demonstrated that asymptomatic patients may obtain clinical benefits from the continuation of EGFR TKI treatment even if restaging had shown a radiological progression. In this clinical scenario, adding LATs to progressing metastatic sites should be considered, particularly if oligoprogression occurred [1]. For these reasons, aggressive LT with SBRT and/or surgery may help to extend the disease control interval [23].

Recently, Giraud et al. [57] tried to summarize all the possible choices of combined treatments and proposed a treatment scheme to be used in EGFR mutated oligometastatic/oligoprogressive patients treated with early generation TKIs:

1)

Oligometastatic scenario (less than 5 lesions), synchronous:

a.

Extracranial metastases: LATs to all metastatic sites.

Similarly, Passaro et al. [58] recently completed the ESMO consensus statements on the management of EGFR mutant NSCLC patients confirming the clinical need of having data that are more consistent to be used in the daily clinic in the absence of randomized prospective trials. Even if in this consensus specific information on the use of LATs in this setting is not provided, several authors have suggested that SBRT may be used in case of oligoprogression during first or later TKI therapies with the aim of continuing the same treatments without increasing toxicities [58]. Similarly, in case of oligopersistence after a good response to systemic TKI therapy consolidating these results using metastasis directed SBRT with the aim of killing clonogen and TKI resistant cells (to primary tumor or systemic lesions) may be considered [58]. Different studies investigated the safety of the association between RT, on thoracic or metastatic sites, and EGFR TKIs. In a Chinese study, the combination of thoracic RT and TKIs for stage IV EGFR mutated NSCLC, found that grade 3 or superior adverse events were radiation pneumonitis and rash, frequently very well controlled by corticosteroids administration [59]. Jia et al. [60] found a high rate of grade 3 or superior pneumonitis (54%), including a Common Terminology Criteria for Adverse Events (CTCAE) grade 5, when palliative RT was associated to osimertinib. Zhou et al. [61] evaluated the safety of whole brain RT with or without EGFR TKIs. The combination of the treatments improves the ORR and 1-year survival rate. The treatment related rash was statistically higher in whole brain RT (WBRT) + TKI, but the incidence of nausea and vomiting, fatigue and myelosuppression were not different between two groups [61]. Guimond et al. [62] retrospectively analyzed the association between RT and EGFR TKI association. Erlotinib and gefitinib were reported safe with RT, but data with metastasis-directed or palliative RT are absent. The authors proposed to stop these systemic therapies 1–2 days before starting RT. They found a lack of data also for crizotinib and osimertinib, so authors recommended at least two days of suspension and also stressed the importance of lung dosimetry during the RT planning, especially for patients with interstitial lung disease [62].

In patients with EGFR mutated oligometastatic NSCLC, systemic therapy with third line EGFR inhibitor such as osimertinib represents the “gold standard” achieving very favorable results in terms of clinic outcomes when compared with CHT. However, many patients still develop loco-regional or systemic failure, often with a “slow” progression limited to few metastatic sites (oligoprogression or oligorecurrence). In this scenario, in well select patients, LT, such as SBRT or radiosurgery, should be considered. Indeed, LAT may be crucial to kill tumoral clones resistant to a given systemic therapy, to avoid systemic multiple sites dissemination and to stabilize the clinical state, permitting to continue the same systemic therapy. The limitations of this review are represented by its intrinsic not-systematic analysis of the literature, by the retrospective nature of most of the studies and by the non homogeneus population investigated. Furthermore, most of the presented studies, particularly those that are retrospective, do not clarify the timing of combination between RT and TKIs. Several studies had a satisfactory number of patients analyzed achieving very interesting results in terms of clinical outcomes. Finally, more prospective data are needed to better select candidates for consolidative LT without significant toxicity risk. This review may be useful to tailor combined therapy in patients clinical routine as it has been reported the current state of art in safely applying sophisticated personalized LT such as SBRT to metastatic oligoprogressive disease sites in the clinical practice. However, more robust data from clinical trials are needed to better clarify the role, the timing and the safety of metastasis directed local therapies and their impact in the daily routine.