Introduction

Lung cancer is the most common cancer and is the leading cause of death from cancer worldwide, with 2.1 million new cases in 2018 [1]. The lung is also one of the most common sites for metastatic malignant disease, often with a poor prognosis [1, 2].

The standard of treatment for early-stage (stage I or II) non-small-cell lung cancer (NSCLC) and metastatic disease is surgical resection, typically by lobectomy, although anatomical segmentectomy and wedge resection could be considered for smaller lesions [3]. With the surgical option, the 5-year survival rate is 50–75% [4]. Unfortunately, only one-third of patients may meet the criteria for lobar or sub-lobar resection [5], due to severe medical comorbidities or poor cardiorespiratory function that preclude surgical intervention or the presence advanced-stage malignancy at diagnosis. Notably, only 17% of patients present with stage I disease [6].

In these patients, image-guided ablation and radiation therapy are increasingly offered as alternative therapies in non-surgical candidates [7,8,9,10,11,12]: conventional external beam radio-therapy (RT) may be considered as a first line treatment, but this modality poses a considerable risk for radiation pneumonitis [5]. Image-guided ablation, such as RFA, Microwave ablation (MWA) or cryo-ablation, is increasingly offered as alternative therapy in non-surgical candidates [13,14,15,16,17,18] for primary and secondary lung tumors treatment. RFA has been used extensively in treatment of lung lesions in patients who are not candidates for resection and there is growing scientific evidence supporting its use for primary and secondary lung tumors treatment [19,20,21,22,23,24,25,26,27,28].

Several studies have demonstrated that RFA can completely destroy an area of healthy lung or malignant lung tumors in an animal tumor model [29, 30] and a clinical study of RFA before resection showed 100% necrosis for 9 of 9 lung metastases [31].

A review of 17 of the most recent publications of lung RFA including NSCLC and lung metastases demonstrated a median rate of complete ablation of 90%, with a variability from 38 to 97% [32].

The objectives of our retrospective study are to evaluate overall survival (OS), Progression-Free Survival (PFS), Local Progression-Free Survival (LPFS), Cancer-Specific Survival (CSS) and complication rates of RFA for primary and metastatic lung tumors in patients that cannot be considered surgical candidates.

Material and methods

Patients

Our retrospective study was done in accordance with the Declaration of Helsinki and with the approval of the Ethics Committee of Azienda Ospedaliera Universitaria "Luigi Vanvitelli"—AORN "Ospedali dei Colli" of Naples (No. 311/18). Informed consent was obtained before any of the procedures.

In 10 years (2007–2017), 174 patients affected by primary (n = 78; 45%) or secondary (n = 96; 55%) lung cancer were selected to RFA treatment. There was a total of 112 males and 62 females with an age range of 36–83 years (mean [± SD] age, 67.7 ± 8.7 years) (Tables 1, 2).

Table 1 Kaplan–Meier analysis for local progression-free survival (LPFS); progression-free survival (PFS); overall survival (OS), and cancer-specific survival (CSS)
Full size table
Table 2 Local progression-free survival (LPFS), progression-free survival (PFS), overall survival (OS), and cancer-specific survival (CSS) rates at 1, 3, and 5 years
Full size table

Each patient was visited by our tumor board composed by pulmonologists, oncologists, thoracic surgeons, interventional radiologists and diagnostic radiologists. After the tumor board decision, all patients provided written informed consent to the RFA procedure after a careful explanation of the technique and associated risks.

Specific inclusion criteria consisted of (1) age between 18 and 85 years; (2) evidence of up to three metastases per lung; (3) patients rejected or considered unsuitable for surgery (patients with impaired lung function, poor cardiopulmonary reserve, significant medical comorbidities, anatomical restraints, advanced disease, elderly patients and poor performance status of two or greater with Eastern Cooperative Oncology Group performance) or patients refusing to undergo surgery. Exclusion criteria included the following: (1) more than three lesions per lung; (2) maximum diameter of metastases > 5 cm; (3) lesions immediately adjacent to major pulmonary vessels or major bronchi or the pulmonary hilum of the lung; (4) bleeding disorders (international normalized ratio (INR) more than 1.5; platelet count less than 100 × 109/L) not responding to medical treatment; (5) presence of pace-maker.

Treatment

Before any RFA procedure, a thorough clinical assessment was carried out, recent laboratory tests evaluated and imaging studies reviewed. Antiplatelet and anticoagulant medication was previously discontinued.

All RFA procedures were performed percutaneously with CT-guidance (Brilliance CT 16-slice Philips, Healthcare Philips Medical Systems, Best, The Netherlands) by one senior interventional radiologist (F.L., whom has more than 10 years of experience with RFA of lung lesions). Scanning parameters were as following: (120 kV, 200 mA; collimation: 5 mm; pitch: 1.25).

After localization of the nodules on the preview, each patient was specifically positioned over the CT table according to the predetermined needle tract, to provide the shortest and most direct path to the tumor, without crossing major thoracic structure (great blood vessels or major bronchi, bronco-vascular bundles, bullae and/or inter-lobar fissures was chosen).

All patients were administered local anesthesia with Lidocaine and analgo-sedation, using Remifentanil and Midazolam. All patients were submitted to a pre-anesthetic consultation and, during the procedure, were continuously monitored by an anesthesiologist monitoring vital parameters (ECG, blood saturation, breath frequency).

The successful central placement of the needle electrode into the center of the lesion was confirmed.

In these ten years, we used three different devices: we started with a multiple tines device, the Rita Starburst CL probe with a RITA 1500 generator (Rita Medical, Mountain View, CA, USA), a single tip disposal, the Cool Tip probe (Covidien, Medtronic, Minneapolis, MN, USA), and a single tip probe, AMICA (HS, Italy) were used (Fig. 1). The electrode is linked to an RF generator and grounding pad (placed on the patient’s skin). The electrode is linked to an RF generator and grounding pad (placed on the patient’s skin). An alternating high-frequency electrical current is created within the tissue of the patient between the RF applicator and the grounding pad [33, 34] resulting in coagulative necrosis and tissue destruction nearby the probe.

Fig. 1
figure 1

Three different radiofrequency ablation (RFA) devices used. a Rita Starburst CL probe with a RITA 1500 generator (Rita Medical, Mountain View, CA, USA), before and after tines opening. b Cool Tip probe (Covidien, Medtronic, Minneapolis, MN, USA). c AMICA (HS, Italy)

Full size image

Time of RFA procedures was calculated to obtain an ablation area was large enough to cover the entire focus with ablative safety margins around the nodules of at least 0.5 cm. After ablation the electrode was retracted with cauterization of the electrode path from the tumors to the subcutaneous tissues to prevent bleeding and tumor seeding. Spiral CT was performed immediately after the ablation for a preliminary assessment of results and detection of complications. A high-density area at the treatment site with a diameter equal to or greater than the initial tumor surrounded by a rim of parenchymal ground-glass was considered an early indicator of treatment success [18].

Technical effectiveness was defined as the rate of complete ablation observed on enhanced CT imaging performed 1 month after the first RFA procedure (Figs. 2, 3).

Fig. 2
figure 2

Metastatic (colorectal) lung lesion in the 1st month follow-up. In these CT-scans there is no evidence of contrast enhancement, also shown in the ROIs

Full size image
Fig. 3
figure 3

Three-month follow-up of a primary lung cancer (adenocarcinoma). There is no enhancement of the lesion after contrast injection

Full size image

Follow-up

The follow-up period was defined as the duration from the date of enrollment in the RFA procedure until death or the last visit; patients that started treatments from February 2007 to September 2017 were considered, the follow-up was conducted until December 2019.

Follow-up visits were scheduled at 1-, 3-, 6-, 12 months and after every 6 months. Each follow-up appointment included clinical examination at the Oncology and Pulmonology Unit and tumor recurrence monitoring by contrast-enhanced chest CT examination. Frequency of local recurrence, disease progression and mortality were the primary end-points. Increased enhancement after the intravenous administration of contrast material on chest CT at 1–3 months of follow-up compared to the pre-treatment baseline enhancement was recognized as residual or recurrent malignant disease. When enhanced CT scanning indicated that abnormal enhancement still could be detected tumor residue or local recurrence was considered. On the other hand, necrosis, identified as a non-enhancing area larger than the ablated tumor, nodular zone, cavitation zone, fibrosis, atelectasis or cysts were considered as complete response to treatment (Figs. 4, 5) [34].

Fig. 4
figure 4

Metastatic lesion before, the day after the treatment and 6 months later

Full size image
Fig. 5
figure 5

Primary pulmonary lesion before and 6 months later. The lesion after the treatment is larger than before, but there is no enhancement

Full size image

All the lesions were assessed by three radiologists, each of whom has more than 10 years of clinical and radiological experience. Potential complications, such as pleural effusion, pneumothorax, subcutaneous emphysema, lung abscess, bleeding or worsening of pulmonary function, were monitored and reported if present. The definition of minor and major complications was assigned according to the Society of Interventional Radiology Classification System for Complications by Outcome [35].

Statistical analysis

Kaplan–Meier analysis for OS, PFS, LPFS and CSS was performed searching differences between metastatic and primitive lung cancer patients.

SPSS v25.0 (IBM, Armonk, New York) was used for all statistical analyses. p values were considered significant when < 0.05.

Results

From February 2007 to September 2017, 264 CT-guided ablation sessions were performed in 174 patients affected by primary (n = 78; 45%) or secondary (n = 96; 55%) lung cancer were selected to RFA treatment. There was a total of 112 (64.37%) males and 62 (35.63%) females; median age was 68 years (range 36–83 years).

The median number of lesions per patient was of 1.43 lesions (range 1–3). The median size of the largest treated lesions was 20.06 mm (range 5–50). Median length of stay for patients treated with RFA by the percutaneous approach was 2.5 days (range 1–5).

Median temporal duration of RFA procedures was 9 min (range 5–22 min). During a median follow-up time of 30.57 months (range 4.2–97.5), 47/174 (27.01%) recurrences, 102/174 (58.62%) deceases, and 94/174 (54.02%) cancer-related deceases were reported. Only 14/47 (29.79%) relapses were reprocessed; the other 33 lesions were not retreated because of the worsening of general clinical conditions of the patients, appearance of metastasis in organs other than the lung or for the death of the patients due to non-oncologic causes.

Kaplan–Meier analysis showed a statistically significant difference between mean survival for metastatic and primitive disease for OS (p value < 0.00003). No statistically significant difference was found for LPFS (p value 0.154), PFF (p value 0.674) and CSS (p value 0.743). More detailed data are shown in Table 1 and in Fig. 6. LPFS, PFS, OS, and CSS rates at 1, 3 and 5 years are showed in Table 2.

Fig. 6
figure 6

Kaplan–Meier analysis for progression-free survival (PFS); overall survival (OS), and cancer-specific survival (CSS). Time is expressed in months. Censored cases are represented by a vertical line (+)

Full size image

Immediate radiofrequency ablation-related complications (pneumothorax, pleural effusion and subcutaneous emphysema) were observed respectively in 42, 53 and 13 of 264 procedures (15.9%, 20% e 5%). There also occurred one major complication (lung abscess, 0.36%). No significant worsening of pulmonary function was noted.

Discussion

In our study, percutaneous RFA proved to be an effective minimally invasive treatment in both primary and secondary lung neoplastic lesions.

Previous reports of large series showed excellent results in terms of OS, especially in patients with small lesions with a diameter of less than 2 cm [36, 37]. In particular Kodama et al. obtained very good results with one-, three- and five-year OS of 97.7%, 72.9% and 55.7% respectively [38]. More recently, Gao et al. reported results in 108 patients affected by NSCLC treated with CT-guided RFA including lesions larger than 3 cm, with a 1-, 2- and 3-year OS rate of 80%, 59% and 24% respectively [39]. Mu et al. reported results in 79 oligometastatic HCC patients with well-controlled intrahepatic disease, reporting OS rates of 91%, 70% and 48% at 1-, 2- and 3-year respectively [40]. In our series, comprehensive of large lesions, up to 5 cm in diameter, we obtained less good results in OS, this may be due to selection of severely ill patients, in fact our PFS, LPFS and CSS compare well with these other studies. There is a similar tendency towards greater responsiveness of metastasis, though that the difference is not significant.

In our experience, RFA confirmed to be a is a feasible alternative either when surgery is not possible or is refused, with prolonged life expectative, a better quality of life and very low complication rate, and as a complementary role in treatment for lung cancer.

Currently, no standard imaging follow-up protocol has been established or uniformly accepted [41]. We decided to use contrast-enhanced CT scan in our patients because CT appeared to be a reliable way to immediately evaluate treatment results and certify complete tumor necrosis by measuring the post-treatment size increase at the tumor site, recognizing ground-glass opacity halo and discarding enhancement after contrast administration [18].

We obtained rate an OS rate at 1, 3 and 5 years of 66.73%, 23.13%, 16.19% respectively in patients affected by primary lung tumors and 85.11%, 48.86%, 43.33% in patients affected by metastatic lung tumors.

The goal of loco-regional therapies is mainly to chronicalize the illness and according with this objective PFS at 1, 3 and 5 years were 79.8%, 60.42%, 15.4% in primary lung tumors and 78.59%, 51.8% and 6.07% in metastatic patients.

The LPFS at 1, 3 and 5 years were 79.8%, 64.69%, 18.87% in primary lung tumors and 86.29%, 69.15% and 44.45% in metastatic patients.

CSS at 1, 3 and 5 years was 95.56%, 71.84%, 56.72% in primary lung tumors and 94.07%, 71% and 71% in metastatic patients.

Cumulative median OS was 49.11 months, with a statistical significant difference between primitive and secondary pulmonary lesions (p < 0.00003): 53.65 months in metastatic patients and 36.87 months in primitive lung cancer patients.

We conducted the procedure in analgo-sedation and local anesthesia in free breathing, our patients never required intubation. Recovery was always immediate as soon as the procedure terminated. In our study there was a total of 47 local recurrences, of which only 14 have been treated. The other 33 lesions have not been treated because of the worsening of general clinical conditions of the patients or, appearance of metastasis in organs other than the lung.

In performing RFA, it is of importance to avoid overheating during the thermal ablation to avoid tissue charring that prevents electrical conduction and consequently a complete tumor necrosis.

Another possible limitation of RFA compared with other ablative techniques such as MWA is heat loss via the “heat sink” effect. In this case blood circulating in vessels or air in bronchi have a cooling effect limiting the area of tissue necrosis [35], for this reason it is critical to take in account the proximity of major vessels or bronchi in the proximity of the area to be treated. Anyway, we considered the heat sink effect a protection against iatrogenic damages of vessels and bronchial walls.

Immediate RFA-related complications (subtle pneumothorax, pleural effusion and subcutaneous emphysema) were observed respectively in 42, 53 and 13 of 264 procedures (15.9%, 20% and 5%).

The overall procedure-related complications incidence was in good agreement with that reported in previous studies [42, 43]. There also occurred one major complication, lung abscess, in the 0.36% of cases. We didn't observe other significant complications related to the procedure such as hemorrhage, infections, seeding of lesions and no significant worsening of pulmonary function was noted.

In conclusion we consider RFA a valid alternative to surgery alone, prolonging OS and ameliorating quality of life in primary and metastatic lung tumors, without relevant side effects or complications.

A major limitation of our study is that it is retrospective, so to accurately put the place thermal ablation among the different therapeutic choice it is necessary to initiate to prospectively compare this technique with other, already consolidated therapies in lung cancer, alone or in combination.