Is radiofrequency ablation more effective than stereotactic ablative radiotherapy in patients with early stage medically inoperable non-small cell lung cancer?

Abstract

A best evidence topic in thoracic surgery was written according to a structured protocol. The question addressed was ‘is radiofrequency ablation more effective than stereotactic ablative radiotherapy in patients with early stage medically inoperable non-small cell lung cancer?’ Altogether, over 219 papers were found, of which 16 represented the best evidence to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. Radiofrequency ablation (RFA) and stereotactic ablative radiotherapy (SABR) offer a clear survival benefit compared with conventional radiotherapy in the treatment of early stage non-small cell lung cancer (NSCLC) in medically inoperable patients. Overall survival at 1 year (68.2–95% vs. 81–85.7%) and 3 years (36–87.5% vs. 42.7–56%) was similar between patients treated with RFA and SABR. However, 5-year survival was higher in SABR (47%) than RFA (20.1–27%). Local progression rates were lower in patients treated with SABR (3.5–14.5% vs. 23.7–43%). Both treatments were associated with complications. Pneumothorax (19.1–63%) was the most common complication following RFA. Fatigue (31–32.6%), pneumonitis (2.1–12.5%) and chest wall pain (3.1–12%) were common following SABR. Although tumours ≤5 cm in size can be effectively treated with RFA, results are better for tumours ≤3 cm. One study documented increased recurrence rates with larger tumours and advanced disease stage following RFA. Another study found increasing age, tumour size, previous systemic chemotherapy, previous external beam radiotherapy and emphysema increased the risk of toxicity following SABR and suggested that risk factors should be used to stratify patients. RFA can be performed in one session, whereas SABR is more effective if larger doses of radiation are given over two to three fractions. RFA is not recommended for centrally based tumours. Patients with small apical tumours, posteriorly positioned tumours, peripheral tumours and tumours close to the scapula where it may be difficult to position an active electrode are more optimally treated with SABR. Treatment for early stage inoperable NSCLC should be tailored to individual patients, and under certain circumstances, a combined approach may be beneficial.

INTRODUCTION

A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].

THREE-PART QUESTION

In [patients with early stage medically inoperable non-small cell lung cancer], is [radio-frequency ablation] superior to [stereotactic ablative radiotherapy] treatment?

CLINICAL SCENARIO

You are at a conference hearing about the effectiveness of radiofrequency ablation (RFA) and stereotactic ablative radiotherapy (SABR) in patients with early stage medically inoperable non-small cell lung cancer (NSCLC). You have an 85-year old patient who has been diagnosed with Stage IA NSCLC and is not fit for surgery due to his extensive comorbidities. You decide to do a literature search.

SEARCH STRATEGY

An English language literature review was performed on MEDLINE 1948 to July 2011 using the Ovid interface: [catheter ablation/OR radiofrequency ablation.mp.OR radiosurgery/OR stereotactic radiotherapy.mp. OR stereotactic ablative radiotherapy.mp. OR stereotactic body radiation therapy.mp. OR stereotactic Irradiation.mp. OR radiation therapy.mp] AND [non small cell lung carcinoma.mp. OR Carcinoma, Non-Small-Cell Lung] AND [disease-free survival/ OR survival/ OR survival.mp. OR toxicity.mp. OR control rate.mp].

SEARCH OUTCOME

The search returned 219 papers. In addition, the references of relevant papers were searched. Sixteen papers provided the best evidence to answer the question. These are tabulated in Table 1.

Table 1:

Best-evidence papers

Author, date, journal and country, study type (level of evidence)	Patient group	Outcomes	Key results	Comments, study weaknesses
Huang et al., (2011), Eur J Cardiothorac Surg, China [2] Retrospective study	Three hundred and twenty-nine patients with 436 lung tumours treated with RFA from 1999 to 2006 Primary (n = 237) Metastatic (n = 92) Inclusion criteria: (1) Aged 18–80 (2) Poor pulmonary function with FEV1 < 1 l, FEV1% <50%; MVV <50% and/or high cardiac risk (3) Refusal of surgery	Median progression-free period Local progression 30-day mortality Overall survival: 1, 2 and 5 years Complications: Pneumothorax Haemoptysis Haemothorax Pneumonia Pericardial tamponade	21.6 months 78/329 (23.7%) 0.6% 68.2, 35.5, 20.1% 63/329 (19.1%) 14/329 (4.2%) 10/329 (3.0%) 15/329 (4.5%) 3/329 (0.9%)	RFA is a safe and well-tolerated procedure with confirmed efficacy in the treatment of malignant lung tumours Tumours >4 cm have a significantly increased risk of local progression Treatment-related complications were counted if within 30 days after RFA treatment
Beland et al., (2010), Radiology, USA [3] Retrospective study	Seventy-nine patients with 79 tumours underwent RFA for primary NSCLC from 1998 to 2008 Stage IA n = 35 (78%) Stage IB n = 7 (16%) Stage IIIB n = 3 (7%) Adjuvant external beam radiation n = 19 (24%) Concomitant brachytherapy n = 9 (11%) Mean follow-up 17 months	Residual tumour or recurrence Recurrence pattern: Local Intrapulmonary Nodal Mixed Distant metastases	34/79 (43%) 38% 18% 18% 6% 21% 23 months	RFA is a promising treatment option for primary lung cancer in non-surgical patients, however disease recurrence is common, occurring in 43% of patients At 2 years, local recurrence was the most common at 28%, suggesting that more aggressive initial RF ablation and adjuvant radiation may offer improvement in outcomes Potential for understaging of disease as patients did not undergo mediastinoscopy No standardized post-treatment follow-up imaging protocol
de Baère et al., (2006), Radiology, France [4] Prospective study	Sixty patients with 97 treatable lung tumours underwent 74 RFA using CT guidance Patients who received systematic chemotherapy during follow-up period n = 22 Primary n = 9 (15%) Metastatic n = 51 (85%)	Rate of incomplete local treatment per tumour at 18 months: Tumours <2 cm Tumours >2 cm Survival at 18 months: Overall Lung disease free Complications: Pneumothorax Alveolar haemorrhage Pleural effusion: Immediately after treatment 24–48 h after treatment FEV1 (L): Before treatment After treatment VC (L): Before treatment After treatment Post-procedure haemoptysis	7% 5% 13% (P = 0.66) 71% 34% 40/74 (54%) 8/74 (11%) 7/74 (9%) 45/74 (60%) 0.62–3.65 0.72–3.65 (P = 0.65) 0.80–8.0 0.83–7.98 (P = 0.93) 7/74 (10%)	RFA has high local success rates of complete ablation and curative treatment in inoperable primary and metastatic lung tumours and is well tolerated Ideal follow-up imaging to determine early treatment failure remains to be improved, probably including functional imaging and CT Small population Twenty-two patients received systemic chemotherapy for distant metastases during follow-up, hence difficult to evaluate the effect of such therapy on the rate of incomplete local treatment
Lencioni et al., (2008), The Lancet, Italy [5] Multicentre prospective trial	One hundred and six patients with 183 biopsy-confirmed lung tumours <3.5 cm underwent RFA NSCLC (n = 33) Median follow-up 24 months	Technical success Major complications: Pneumothorax Pleural effusion Complete tumour response lasting at least 1 year Overall survival (NSCLC): 1 year 2 years Cancer-specific survival (NSCLC): 1 year 2 years Stage I NSCLC: 2-year overall survival 2-year cancer specific survival	99% n = 27 n = 4 88% 70% 48% 92% 73% 75% 92%	Percutaneous CT-guided RFA yields high proportions of sustained complete ablation in patients with primary or secondary lung tumours, and is associated with acceptable morbidity Heterogeneous patient population Mean follow-up not long enough to detect late tumour recurrences PET scans not routinely used for the assessment of response
Hiraki et al., (2007), J Thorac Cardiovasc Surg, Japan [6] Retrospective study	Twenty patients with Stage I NSCLC underwent RFA Median follow-up 21.8 months	Local progression Local control rate: 1 year 2 years 3 years Mean survival Overall survival: 1 year 2 years 3 years Cancer-specific survival: 1 year 2 years 3 years Complications: Pneumothorax Pleural effusion	7/20 (35%) 72% 63% 63% 42 months 90% 84% 74% 100% 93% 83% 13/20 (57%) 4/20 (17%)	Treatment of Stage I NSCLC with one or more sessions of RFA offers promising outcomes in relation to survival. However, local progression rates are relatively high Short follow-up period Small study population
Pennathur et al., (2007), J Thorac Cardiovasc Surg, USA [7] Retrospective study	Nineteen patients with Stage I NSCLC underwent RFA under CT-guidance Stage IA (n = 11) Stage IB (n = 8) Mean follow-up 29 months	Local progression Initial complete response Partial response Overall survival: 1 year 2 years Complications: Pneumothorax	8/19 (42%) 2/19 (10.5%) 10/19 (53%) 95% 68% 63%	RFA appears to be safe in high-risk patients with stage I NSCLC, with reasonable results in terms of survival in high-risk patients who are not fit for surgical intervention Small sample size
Simon et al., (2007), Radiology, USA [8] Retrospective study	One hundred and fifty-three patients with 189 tumours underwent 183 RFA sessions Primary (n = 116) Metastatic (n = 73) Stage I NSCLC (n = 75) Median follow-up 20.5 months	(≤3 cm) Local tumour progression–free rates: 1 year 2 years 3 years 4 years 5 years (>3 cm) Local tumour progression–free rates: 1 year 2 years 3 years 4 years 5 years Difference between the survival rates: (≤3 cm) and (>3 cm) Overall survival rates: (NSCLC) 1 year 2 years 3 years 4 years 5 years Complications: Pneumothorax Chest tube insertion	83% 64% 57% 47% 47% 45% 25% 25% 25% 25% (P < 0.002) 78% 57% 36% 27% 27% 28.4% 9.8%	Lung RFA appears to be a safe treatment for Stage I NSCLC and is linked with promising long-term survival and local tumour progression outcomes Biopsies were not routinely performed during follow-up A proportion of patients treated concomitantly with systemic chemotherapy and/or external beam radiation therapy
Kashima et al., (2011), Am J Roentgenol, Japan [9] Retrospective study	Four hundred and twenty patients with 1403 lung tumours underwent 1000 RFA sessions	Complications: Death Aseptic pleuritis Pneumonia Lung abscess Bleeding requiring transfusion Pneumothorax requiring pleural sclerosis Bronchopleural fistula Brachial nerve injury Tumour seeding Diaphragm injury	0.4% 2.3% 1.8% 1.6% 1.6% 1.6% 0.4% 0.3% 0.1% 0.1%	Lung RFA is a relatively safe procedure, but it can be fatal in few cases. Known risk factors such as age, tumour size, platelet count, previous systemic chemotherapy, previous external beam radiotherapy, emphysema should be used to stratify patients
Zemlyak et al., (2010), J Am Coll Surg, USA [10] Retrospective study	Sixty-four patients with Stage I NSCLC SLR (n = 25) RFA (n = 12) PCT (n = 27)	Overall 3-year survival: SLR RFA PCT 3-year cancer-specific survival: SLR RFA PCT 3-year cancer-free survival: SLR RFA PCT	87.1% 87.5% 77% (P > 0.05) 90.6% 87.5% 90.2% (P > 0.05) 60.8% 50% 45.6% (P > 0.05)	SLR, RFA and PCT are reasonable alternatives to lobectomy for patients who are poor candidates for major surgery. Survival at 3 years is comparable after sublobar resections and ablative therapies. Ablative therapies appear to be a reasonable alternative in high-risk patients not fit for surgery Small sample Selection bias SBRT not included in comparison
Lagerwaard et al., (2008), Int J Radiat Oncol Biol Phys, The Netherlands [11] Retrospective study	Two hundred and six patients with Stage I NSCLC underwent SRT Medically inoperable (n = 167) Refused surgery (n = 39) Inclusion criteria: (1) Tumour <6 cm (2) Confirmed malignancy (cytohistologic or CT) (3) Absence of metastases on PET scan Prescription dose 60 Gy in three, five or eight fractions	Median overall survival: 1-year survival rate 2-year survival rate Local recurrence: T1 T2 Overall recurrence DFS: 1 year 2 years Toxicity: Fatigue Chest wall pain Nausea Dyspnoea Cough Pneumonitis Rib fractures Chronic thoracic pain	34 months 81% 64% 3.5% 2 /129 (1.6%) 5/90 (5.6%) (P = 0.13) 21% (43 patients) 83% 68% 31% 12% 9% 6% 6% 3% 2% 1%	SRT is well tolerated in patients with extensive comorbidity presenting with inoperable Stage I NSCLC with high local control rates and minimal toxicity Median follow-up was only 12 months. Most recurrences occur within 2 years following treatment Only 88 patients attended for follow-up at 1 year Only a minority of the patients had pathologic confirmation of malignancy
Haasbeek et al., (2010), Cancer, The Netherlands [12] Retrospective study	One hundred and ninety-three patients ≥75 years with 203 tumours treated using SRT T1 (n = 118) T2 (n = 85) 80% medically inoperable 20% declined surgery Median follow-up 12.6 months	Survival rate: 1 year 3 years Median overall DFS: 1 year 3 years Complications: Fatigue Nausea Cough Dyspnoea Chest wall pain Rib fracture Grade ≥3 radiation pneumonitis Chronic chest wall pain Acute toxicity	85.7% 45.1% 32.5 months 89.2% 72.6% 32.6% 4.1% 5.7% 5.2% 3.1% 1.6% 2.1% 2.6% 1.6%	SRT achieved high local control rates with minimal toxicity in patients aged ≥75 years, suggesting it should be considered as a curative alternative in the treatment of NSCLC Median follow-up of 12.6 months not long enough
Le et al., (2006), J Thorac Oncol, USA [13] Retrospective study	Thirty-two patients with inoperable lung tumours treated with single-fraction SRT NSCLC (n = 20) Metastases (n = 12) Tumour diameter 20–62 mm 15–30 Gy in one fraction 15 Gy (n = 9) 20 Gy (n = 1) 25 Gy (n = 20) 30 Gy (n = 2) Median follow-up 18 months	Local progression: (NSCLC) >20 Gy <20 Gy CR rate: >20 Gy <20 Gy 1-year overall survival: (NSCLC) Toxicity: Fatigue Pneumothorax Grade 2–3 pneumonitis Pleural effusion	9% 54% (P = 0.03) 57% 10% (P = 0.21) 85% 10/32 6/32 (19%) 4/32 (12.5%) 1/32	Single-fraction SRT is feasible for selected patients with lung tumours and higher doses (>20 Gy) are associated with improved local control. For those who have had prior thoracic radiotherapy, doses ≥25 Gy may be too toxic Short follow-up period Variation in treatment techniques (breath-holding vs. tracking)
Onishi H et al., (2004), Cancer, Japan [14] Retrospective study	Two hundred and forty-five underwent hypofractionated high-dose STI between 1995 and 2003 BED ≥100 Gy (n = 173) BED 100 Gy (n = 72) Stage IA n = 155 Stage IB n = 9 Tumour diameter 7–58 mm (median, 28 mm) CT chest usually obtained 3-monthly for first year and repeated every 4–6 months thereafter Tumour response evaluated using previously published National Cancer Institute (NCI) criteria	Local progression overall: BED ≥100 Gy BED 100 Gy Local disease recurrence: BED ≥100 Gy BED 100 Gy 3-year survival rate: BED ≥100 Gy BED 100 Gy Overall survival rates: 3 years 5 years Local tumour response: CR (completely disappeared/replaced by fibrotic tissue) PR (≥30% reduction in the maximum cross-sectional diameter) Overall response rate BED ≥100 Gy BED 100 Gy Toxicity: Radiation-induced pulmonary complications	14.5% 8.1% 26.4% (P < 0.05) 13.5% 8.1% 26.4% (P < 0.01) 88.4% 69.4% (P 0.05) 56% 47% 57/245 (23.3%) 151/245 (61.6%) 84.8% 84.5% 83.3% 17/245 (6.9%)	Hypofractionated high-dose STI is a feasible and effective curative treatment of patients with Stage I NSCLC Local control and survival rates better in patients treated with BED ≥100 Gy than for BED 100 Gy Treatment parameters heterogeneous
Timmerman et al., (2010) JAMA, USA [15] Multicentre prospective study	Fifty-five patients with biopsy-proven peripheral T1-T2N0M0 NSCLC (measuring 5 cm in diameter) underwent SBRT T1 (n = 44) T2 (n = 11) Prescription dose 18 Gy per fraction ×3 fractions (54 Gy total) Median follow-up 34.4 months	Median overall survival Overall 3-year survival Disseminated recurrence at 3 years 3-year primary tumour control rate Local-regional control rate DFS Adverse events: Grade 3 Grade 4	48.1 months 55.8% 22.1% 97.6% 87.2% 48.3% 7/55 (12.7%) 2/55 (3.6%)	SBRT is an effective treatment in patients with inoperable NSCLC, with high rates of local tumour control and moderate treatment-related morbidity Rarely used invasive pathological staging and histological confirmation of recurrence, lowering accuracy
Fakiris et al., (2009), Radiat Oncol Biol Phys, USA [16] Retrospective study	Seventy patients with biopsy confirmed NSCLC underwent SBRT T1 (n = 34) T2 (n = 36) Prescription dose: 60–66 Gy in three fractions Median follow-up 50.2 months	Local control at 3 years Local recurrence Nodal recurrence Distant recurrence Median survival 3-year overall survival Median survival: T1 tumours T2 tumours	88.1% 4/70 (5.7%) 6/70 (8.6%) 9/70 (12.9%) 32.4 months 42.7 months 38.7 months 24.5 months (P = 0.194)	Treatment with SBRT results in high rates of local control in medically inoperable patients with Stage I NSCLC
Timmerman et al., (2006), J Clin Oncol [17] Prospective study	Seventy patients underwent SBRT for NSCLC Treatment dose: 60 to 66 Gy total in three fractions during 1–2 weeks Median follow-up 17.5 months	2-year overall survival 3-month major response rate Local control 2 years: Deaths due to: Cancer Treatment Comorbid illness Median overall survival Toxicity: Grade 1–2 Grade 3–4 Grade 5 (death)	54.7% 60% 95% n = 5 n = 6 n = 17 32.6 months n = 58 n = 8 n = 6	Local recurrence and toxicity occur late after this treatment; however, this regimen should not be used for patients with tumours near the central airways due to excessive toxicity Short follow-up period This trial did not define a limit on the period of observation of toxicity related to therapy-hospitalizations/deaths occurring more than a year after therapy might not have been attributed to SABR

NSCLC: non-small-cell lung cancer; RFA: radiofrequency ablation; FEV₁: forced expiratory volume in 1 s; VC: vital capacity; DFS: disease-free survival; MVV: maximum voluntary ventilation; SLR: sublobar resection; PCT: percutaneous cryoablation therapy; SRT: stereotactic radiotherapy; SBRT: stereotactic body radiation therapy; STI: stereotactic irradiation; BED: biologic effective dose; CR: complete resolution; PR: partial resolution; MTD: maximum tolerated dose.

RESULTS

The effectiveness of RFA and SABR in the treatment of inoperable NSCLC is well documented. Huang et al. [2] reported a median progression-free-interval of 21.6 months following RFA. Overall survival at 1, 2 and 5 years was 68.2, 35.3 and 20.1%, respectively, and 23.7% patients developed local progression during follow-up. There was no significant difference in outcome for tumours <3 cm, while there was a significant difference in the risk of local progression in tumours >4 cm (P = 0.01). In another study [3], 57% of primary lung tumours treated with RFA had no recurrence. The local recurrence rate was 38%, with increasing tumour size (P = 0.02) and disease stage (P = 0.007) significantly increasing its likelihood.

de Baère et al. [4] documented an 18-month survival rate of 71%, and a trend towards better efficacy for tumours <2 cm in diameter (P = 0.066). The respiratory function was not adversely affected when measured within 2 months of RFA treatment (P = 0.51); however, the long-term effects are unknown and the FDA have received reports of patient deaths associated with lung tumour ablation using RFA.

Lencioni et al. [5] achieved a technical success rate of 99% in performing RFA. 12.5% of patients with NSCLC showed incomplete response or progression of disease. The overall 2-year survival of patients with NSCLC was 48%. Hiraki et al. [6] observed a high local progression rate (35%) within a median of 9.0 months after the first session.

Pennathur et al. [7] documented a local progression in 42%, and the median time to progression was 27 months. The overall 2-year survival rate was 49% for primary lung cancers. Simon et al. [8] documented a significant difference in survival between patients with large (>3 cm) and small (≤3 cm) tumours (P < 0.002). Local recurrence was most common, suggesting that more aggressive RFA and adjuvant radiation may improve outcomes [9, 10]. Kashima et al. [9] concluded that puncture number (P < 0.02) and previous systemic chemotherapy (P < 0.05) were significant risk factors for aseptic pleuritis. Increasing age (P < 0.02) and previous external beam radiotherapy (P < 0.001) were significant risk factors for pneumonia, as were emphysema (P < 0.02) for lung abscess and pneumothorax requiring pleural sclerosis (P < 0.02), and serum platelet count (P < 0.002) and tumour size (P < 0.02) for bleeding. Zemlyak et al. [10] reported comparable survival rates following sublobar resections (87.1%) and ablative therapies (87.5%).

SABR, a non-invasive technique, precisely delivers very high radiation doses in a short period of time. It is well tolerated in patients with extensive comorbidity with high local control rates and minimal toxicity, and results have been so promising that there are ongoing trials comparing SABR with surgery in operable patients. Lagerwaard et al. [11] observed local recurrences in only 3.5% of patients, which is much less than previously reported when using conventional radiotherapy in Stage I NSCLC. Haasbeek et al. [12] found no significant difference in overall survival between the older and younger patient cohorts (P = 0.18) following SABR; however, disease-free survival was slightly better in older patients (P = 0.04). Le et al. [13] reported an association between prior thoracic radiotherapy or chemotherapy and treatment-related toxicity.

On comparing outcomes between patients treated with biologic effective doses (BED) of ≥100 Gy and <100 Gy, Onishi et al. [14] found improved local control and survival rates with BED ≥100 Gy. They reported the most benefit in those with medically operable tumours, treated with BED ≥100 Gy.

Timmerman et al. [15] reported a 3-year local control rate of 97.6%, and an overall 3-year survival rate of 55.8%. Fakiris et al. [16] reported lower rates of toxicity after SABR in patients with peripheral tumours. However, there was no significant difference in survival between patients with peripheral and central tumours (P = 0.69). The 3-year local control (88.1%) was comparable to that following lobectomy. Another study [17] reported mainly grade 1–2 toxicities (83%), consisting of fatigue, musculoskeletal discomfort and radiation pneumonitis. Most cases resolved within 3–4 months of SABR. The authors concluded that patients with perihilar/central tumours had an 11-fold increased risk of experiencing severe toxicity compared with more peripheral locations. Tumours close to the left hemidiaphragm may also be very dangerous to treat with SABR due to their proximity to the stomach.

CLINICAL BOTTOM LINE

SABR is associated with higher 5-year survival rates compared with RFA and conventional radical radiotherapy (40–47% vs. 20.1–27 vs. 19%) [18] and local control rates up to 80–90% [19] are two to three times greater than conventional fractionated radiotherapy. This modality has a favourable toxicity profile in peripheral tumours measuring ≤5 cm. RFA can be performed in one session, whereas SABR is more effective if larger doses of radiation are split over two to three fractions. RFA is more difficult in central tumours but is being increasingly performed with increased operator experience and confidence. Both treatment modalities are associated with side-effects, and risk factors should be used to stratify patients. Overlapping ablations in the same sitting also improve the outcomes for larger tumours. In certain circumstances, a combined approach may be beneficial.

Conflict of interest: none declared.