Introduction

The prognosis of metastatic esophageal cancer (EC) remains poor with an average life expectancy of around 9–12 months with standard systemic chemotherapy [1, 2]. The concept of oligometastatic disease (OMD) was first introduced in 1995 by Hellman and Weichselbaum and described as a stage of transition between localized and widespread metastatic disease [3]. It represents a probable stage in natural evolution of cancer in which tumor cells may proliferate in ‘sanctuaries’ hidden from systemic circulation before wider systemic dissemination. It remains unclear if there is a generic intermediate state of limited metastatic disease present in all cancer types or represents a more specific state driven by a particular molecular signature.

The concept of OMD in EC is controversial with no universally accepted definition. In a systematic review aimed at determining consensus definition OMD was commonly defined as involvement of 1 organ with less than 3 metastatic lesions or 1 extra-regional lymph node (LN) station [4].

We present small case series (n = 4) with advanced metastatic EC who developed an unusual and rather intriguing treatment response to standard therapy with long-term survival and probable cure. In retrospect, all the cases would have been classified as OM with striking similarities in terms of their clinical presentation and long-term outcomes. The primary intention of the manuscript is to highlight the probable presence of OMD state in EC and importance of the combined approach of local and metastasis-directed therapy.

Case report

RT technique

All patients received a palliative dose of 30 Gy in 10 fractions and were set up using virtual simulation in supine position with arms by side and placement of two lateral and central anterior tattoos. The target definition included placement of anterior and posterior fields with incorporation of primary tumor and any adjacent areas of residual metastatic disease with 2–3 cm margin to account for any microscopic spread and organ motion/patient movement. Multi-leaf collimation was to shield any adjacent areas as clinically appropriate. Patients were prescribed 30 Gy in 10 fractions to the isocentre (parallel opposed pair of fields) and delivered 5 days a week over 2 weeks. Image verification was performed daily with kilovoltage (kV) imaging prior to each treatment. Patient staging and corresponding RT portals are illustrated in Figure 1 (case 1 and case 2) and Figure 2 (case 3 and case 4).

Display full size

Figure 1. 

PET imaging showing the primary tumor and area of oligometastasis and corresponding radiotherapy portals (case 1 and case 2). (A) Coronal and axial PET scans of case 1 and case 2; (B) coronal PET scans of case 1 and case 2; (C) corresponding radiotherapy portals in case 1 and case 2. PET images with blue arrow indicating primary tumor and red arrow indicates site of oligometastasis [case 1: para-aortic node (patient also had lung metastasis)—both resolved completely after initial chemotherapy; case 2: paraaortic node]; radiotherapy portals were defined to cover the primary tumor and areas of adjacent metastatic disease/visible residual disease (metastasis-directed therapy)

Display full size

Figure 2. 

PET imaging showing the primary tumor and area of oligometastasis and corresponding radiotherapy portals (case 3 and case 4). (A) Coronal PET scans of cases 3 and 4; (B) axial and coronal PET scans of cases 3 and 4; (C) radiotherapy portals in cases 3 and 4. PET images with blue arrow indicating primary tumor and red arrow indicates site of oligometastasis (case 3: T9 vertebral metastasis; case 4: high paraesophageal node); radiotherapy portals were defined to cover the primary tumor and adjacent areas of metastatic disease/visible residual disease (metastasis-directed therapy)

Discussion

The European Society for RT and Oncology (ESTRO) and European Organisation for Research and Treatment of Cancer (EORTC) have developed a consensus classification and nomenclature of OMD as part of the Oligo-Care project. The classification introduced the concept of ‘induced OMD’ after systemic therapy in patients with previous poly-metastatic disease which was distinguished from ‘genuine OMD’ that was further subdivided into new onset ‘de novo’ OMD or ‘repeat’ OMD developing after previous treatment for OMD [6]. De-novo OMD was divided into metachronous (> 6 months after treatment of primary) and synchronous OMD (< 6 months after treatment of primary).

All the patients in the present case series presented with locally advanced EC and synchronous small-volume metastatic disease. Three patients (Case 2–4) had only 1–2 sites of distant metastasis (bone and/or non-regional LN) and one patient (case 1) had lung and non-regional LN involvement. In retrospect, all the above patients could have been classified as ‘genuine de novo’ OMD. They were treated with standard protocol of initial chemotherapy followed by RT to primary tumor which also included adjacent areas of visible/residual disease that could be safely included in the radiation field (metastasis-directed therapy) (Figures 1 and 2) [7].

In the recent past, several retrospective studies have reported on improved outcomes in EC patients with OMD. These studies have included patients presenting with synchronous or metachronous OMD and have reported 5-year survival rates of 25–50 percent. These studies have emphasised the use of appropriate local (plus systemic) therapy for treating the OMD lesions and adequate control of the primary tumor. These studies have employed different local treatment modalities including surgery with metastasectomy or RT with stereotactic ablative RT (SABR) or conventionally fractionated RT [8–14]. The importance of optimisation of local therapy was highlighted in a recent meta-analysis in which local therapy improved OS compared with systemic anti-cancer therapy alone (HR: 0.47, 95% CI: 0.30–0.74) including liver oligo-metastases (HR: 0.39; 95% CI: 0.22–0.59) [4].

All cases were treated with fractionated RT with palliative dose of 30 Gy in 10 fractions. As it was palliative dose patients were set up and treated with simple parallel opposed pair field arrangement as the dose was within cord tolerance. However, advances in RT technology now easily enable dose escalation with delivery of radical dose (50–60 Gy) to primary tumor with conventional (dose per fraction of 1.8–2 Gy) or hypofractionation (dose per fraction of > 2 Gy).

The advancement of imaging techniques enables better localization of the tumor volume. There are expert panel consensus guidelines available for contouring of target volumes in EC which recommend definition of gross tumor volume (GTV) using a combination of pre-treatment PET scan, CT scan, and endoscopy appearances (EUS preferable) [15]. The continuous advancements in RT technology over the past decades have allowed for EC to be treated with 3-dimensional (3D) treatment planning, including intensity-modulated RT (IMRT), volumetric-modulated arc therapy (VMAT) and helical tomotherapy. These modern techniques alongside advancements in image-guided RT allow the oncologist to deliver higher doses of radiation with more precision to the tumor and with less toxicity to the surrounding normal tissue, which has dramatically reduced morbidity. An additional consideration when defining target volumes is the effect of respiratory motion, which is most important for distal tumors and those involving the GOJ. Multiple techniques that control or account for respiratory motion are available and one of the most straightforward is the measurement of respiratory motion using 4D-CT at the time of simulation, in which case the magnitude of respiratory excursion can be incorporated into the choice of planning target volume (PTV) margin [16]. MRI-guided respiratory gated IMRT is another exciting technology that can be used to incorporate and adjust the organ and tumor motion within the contouring algorithm enabling improved sparing of adjacent organs at risk [17]. Proton therapy is particulate irradiation that deposits maximum dose in immediate vicinity of tumor with no ‘exit’ dose which can be potentially associated with an improvement in the therapeutic index by reducing dose to surrounding organs at risk. In phase 2 study comparing protons with IMRT the use of proton therapy was associated with significant reduction in toxicity burden and post-operative complications in patients with EC [18].

Liu et al. (2020) [19] reported on prospective, single-arm, phase 2 trial to assess the safety and efficacy of SABR for patients with OM SqCC of esophagus. The main inclusion criteria were 3 or fewer metastases and a controlled primary malignancy after radical treatment, with all metastatic lesions amenable to SABR. The 1- and 2-year overall survival rates were 76.2% and 58.0%, respectively [19]. Al-Batran et al. (2017) [20] reported on a prospective phase 2 randomized study (AIO-FLOT3) in patients with OM AC of the stomach and GOJ who received neoadjuvant chemotherapy (FLOT) prior to surgical resection. The median overall survival (OS) was 31.3 months (95% CI: 18.9–NA) for patients who proceeded to surgery [20].

Kroese et al. (2022) [12] reported on multicentre study of 205 patients with de-novo OMD (first-time diagnosis of ≤ 5 distant metastases on FDG-PET/CT). The majority of included patients had EC (73%) with AC histology (79%) and metachronous OMD (52%). The primary tumor was controlled in 68 percent. Improved OS was independently associated with combined use of local plus systemic therapy. The median OS after local plus systemic therapy was 35 months (95% CI: 22–NA) as compared with 13 months (95% CI: 9–21, P < 0.001) after systemic therapy alone [12].

The addition of systemic anti-cancer therapy (SACT) including chemotherapy and immunotherapy agents alongside local therapy has been shown to improve outcomes. However, the optimal timing and scheduling of SACT remains unknown. In the phase 2 study by Liu et al. (2020) [19] patients received chemotherapy after completion of SABR and in AIO-FLOT3 patients received neoadjuvant FLOT regime before surgery [20]. In one of the studies evaluating the use of PD1 inhibitors both sequential and concurrent approaches were used and found to be safe and effective.

In patients presenting with synchronous OMD neoadjuvant chemotherapy would seem to be the most appropriate initial option as it may help to select patients appropriate for subsequent local treatment. All the four patients in present case series with synchronous OMD were treated with upfront chemotherapy followed by RT to primary tumor and inclusion of adjacent areas of metastatic disease particularly all areas of visible residual disease. Therefore, the patients had both primary and metastasis-directed therapy.

These cases were identified retrospectively in view of the unexpected long-term survival and were rather peculiar as areas of residual metastatic disease after initial chemotherapy were adjacent to the site of primary tumor and could be included in the same radiation portal. However, the same may not be true for many patients with more distant sites of OMD where an alternative strategy for metastasis-directed therapy may be warranted possibly in the form of separate RT portals or SABR.

The importance of controlling the primary tumor in the context of OMD has been reported previously [11] and was highlighted in the recent phase III STAMPEDE and HORRAD trials in patients with de-novo hormone sensitive OM prostate cancer. A comprehensive meta-analysis of these studies revealed an OS benefit for prostate-directed RT in patients with fewer than 5 bone metastases (HR: 0.73; 95% CI: 0.58–0.92; P = 0.007) [21].

The dose of RT used in all patients was standard hypo-fractionated palliative regime of 30 Gy in 10 fractions which is less than the biological equivalent dose (BED) used in other studies. In a retrospective study of metachronous OMD in OG cancer 55 patients received RT to areas of metastasis. Patients receiving radical dose of RT (BED₁₀ ≥ 60 Gy) had significant survival benefit compared to those receiving less than radical dose of RT (median OS of 16 months cf. 10 months, P = 0.033) [10]. Similarly, in another study of synchronous OMD there was an indication of better outcomes in patients with limited extra-regional LN involvement who could receive radical dosage of RT. All patients received more dose-intense schedules of chemotherapy compared to above studies which may have influenced the eventual complete response despite palliative dosage of RT. It also remains unclear whether the above responses were due to increased radio-sensitivity determined by particular molecular signature sub-types.

The combined approach using systemic and local therapy (including primary and metastasis-directed therapy) appears to be the favoured approach associated with the best outcomes and based on limited available data [12, 14, 19].

The unusual outcome of these patients renders further support to the existence of probable OM state in EC which may represent a completely different and prognostically favourable sub-group. It adds to the growing pool of evidence indicating the management strategy of OMD patients should be different from those with conventional metastatic disease. A combined strategy of local therapy (surgery or radiation) with appropriate control of primary and metastasis-directed therapy and systemic therapy may achieve best long-term outcomes. The performance of focused prospective research with large phase III RCTs to determine the optimal management of OMD in EC is absolutely essential. The OligoMetastatic Esophagogastric Cancer (OMEC) consortium was designed to develop a multidisciplinary consensus statement for the definition and treatment for OM esophagogastric cancer and drive the direction of future research [22, 23].