Abstract
Background/Aim: Radiofrequency ablation (RFA) for colorectal cancer pulmonary metastases (CRCPulM) has been shown to be safe. Microwave ablation (MWA) has emerged in the treatment for CRCPulM. However, short to long-term efficacy of both modalities have yet to be understood. Patients and Methods: This is a retrospective study of 203 patients who received RFA and MWA from 2000-2018 at a major tertiary hospital in Australia. Results: A total of 161 patients underwent RFA and 42 MWA. Median ablation size and time was 4 (range=3-5 cm) vs. 3.5 cm (range=3-4 cm; p=0.0395) and 49 (range=26-65 min) vs. 8 min (5-13 min) in the RFA and MWA groups, respectively (p<0.001). The complication rate was 112 (55%) and 40 (74%) in the RFA and MWA group, respectively (p=0.011). Life-threatening pulmonary haemorrhage occurred in 1 (0.5%) and 4 (7.4%) patients in the RFA and MWA group, respectively (p=0.007). Local recurrences detected after discharge were similar in both groups [28% (p<0.001)]. However, the MWA group demonstrated higher survival rate and less recurrence rate than RFA in the first 24 months of follow up. Conclusion: RFA and MWA are competitive treatment methods for CRCPulM. Although MWA has significantly higher complication rate than RFA, it can be performed in a much shorter time and lead to a shorter length of hospital stay.
Colorectal cancer (CRC) is the third most common cancer and second leading cause of cancer death in the United States, Europe, and Australia (1-8). Pulmonary metastasis from CRC is the next most common site following liver metastasis (8). However pulmonary metastasectomy can only be offered to a small group of patients (1, 8, 9). Untreated metastatic CRC generally has a poor prognosis of 8 months. Palliative chemotherapy/immunotherapy improves survival (1).
Alternative treatments have been discussed in recent years to eradicate colorectal cancer pulmonary metastasis (CRCPulM) and to prolong survival in patients who are deemed inoperable. These include different types of systemic chemotherapy, radiation therapy, and minimally invasive image-guided techniques such as percutaneous radiofrequency ablation (RFA) and microwave ablation (MWA) (1, 8, 10-17). RFA has been the choice of percutaneous ablation due to its known promising results for loco-regional control of CRC pulmonary metastases (1-4, 8, 10, 18). With the advent of MWA and its theoretical advantage in terms of being able to generate greater ablation temperatures, larger heating radius in the lung which is a poor conductor of heat in a much shorter time, its use has been favoured over RFA (8, 15, 19, 20).
MWA has been compared to RFA in recent years and shown to be similarly effective in one study of Hepatocellular Carcinoma ablation (19-21). However, in several recently published retrospective series and systematic reviews, it is still unclear whether MWA is equivalent or possibly superior to RFA in the context of pulmonary tumours and metastases given a lack of quality evidence comparing the two modalities (8, 15, 22-25). There are few studies directly comparing the RFA and MWA in CRCPulM and it is difficult to discern if existing evidence concerning these treatments is comparable between primary and metastatic lung cancer, or between pulmonary metastases of colon and non-colon origin as a recent study showed heterogeneity between these groups (26). The purpose of this study was to evaluate and compare the effectiveness in treatment including short-term outcome as well as local tumour control, time to tumour progression, and survival rates among patients with pulmonary metastatic colorectal cancer who undergo RFA or MWA.
Patients and Methods
Patient selection and tumour criteria. The Human Research Ethics Committee of a major tertiary hospital in Australia approved the design of the study and waived the need for consent due to the retrospective nature of the study. All patients who had RFA and MWA and were followed up at the hospital were identified through our hospital database. A retrospective data analysis of a prospectively maintained database of 203 patients who received either RFA or MWA from November 2000 to February 2018 was performed (Table I). Where RFA was performed from the period of November 2000 to 2013 and MWA replaced it from the year 2013 onwards in our institution. The consensus on the treatment plans for patients with CRCPulM was achieved during weekly multi-disciplinary team meetings (MDT) by a group of oncological surgeons, medical oncologists, and radiologists.
Results of radiofrequency ablation (RFA) vs. microwave ablation (MWA) for colorectal cancer pulmonary metastases (CRCPulM).
Radiofrequency ablation. Percutaneous RFA procedure is typically performed using computed tomography guidance, and the techniques for introducing the electrode to the tumour are simple and resemble those used in percutaneous lung biopsy. Its monopolar probe has a grounding pad usually attached to the patient’s leg that creates a sinusoidal electrical current of frequency 400-500 kHz between the electrode and grounding pad. Ionic agitation of the tissues, lung parenchymal and tumour in this case, leads to local frictional tissue heating to 60°C that results in instantaneous protein denaturation (1-4, 8, 9, 15, 27, 28). As air is a poor conductor of electricity and good heat insulator, it protects the surrounding lung parenchyma from the effects of the RFA. The surrounding pulmonary vessels also induce a sudden reduction in temperature, which minimizes the effect of RFA otherwise known as the “heat sink effect”. Consequently, due to air being protective of electrical currents and the “heat sink effect”, the resultant resection margin is significantly limited with the RFA technique (11, 22, 25, 29). Our center’s method of RFA treatment is consistent to methods used by Ferguson et al. 2015 and Yan et al. 2006 (1, 3). Percutaneous RFA procedures were performed under conscious sedation (meperidine-midazolam) and local anesthesia (Xylocaine 1%). Using fluoro-CT guided imaging (Xpress SX; Toshiba, Japan), the Rita 1500 generator (Rita Medical, Mountain View, CA, USA) was used, which provided real-time recording and display of power, temperature, and impedance. A Rita Starburst XL probe, with either lengths of 10 or 15 cm, a diameter of 14 gauge, and nine deployable tines, was used. The probe is able to ablate lesions of 5 cm in diameter.
After inserting the probe percutaneously into the lung, we positioned it so that the deployable lines surrounded the lesion. A staged deployment strategy according to the ablation algorithm consisted of an initial power setting at 35 W, which was gradually increased to 150 W. Power was increased to enhance the rate of temperature increments. The target temperature was 90°C, and when this temperature was reached, it was maintained for 15, 20, or 37 min to achieve a complete tumour ablation of 3, 4, or 5 cm in size, respectively.
Microwave ablation. MWA is a relatively new alternative percutaneous thermal technique that creates a much higher frequency electromagnetic radiation field (915 MHz – 2.5 GHz) around a monopolar electrode. The energy propagates directly through air and tissue whereby the lung, which is of high intrinsic impedance, is not a barrier. This then induces homogeneous heating and coagulation necrosis. Charring that usually leads to high impedance to RFA currents by causing high tissue resistance is no longer a barrier in MWA. MWA leads to higher temperatures much quicker than RFA and has the potential of simultaneously treating more lesions in a same or shorter time. Theoretically, MWA leads to a larger ablation area compared to monopolar RFA and has been used to treat lesions larger than 5 cm (10). Recurrence due to ‘heat sink’ effect is also proposed to be reduced in MWA (8, 21). In our center, MWA is performed under general anaesthesia. Under sterile technique, the duty interventional radiologist will introduce the Covidien Emprint™ Ablation System probe of either lengths of 15cm or 20cm and a diameter of 16 gauge percutaneously into the rib space parallel to the level of the lung lesion(s). This is done under real-time CT fluoroscopy to accurately position the electrode into the tumour. This was used in conjunction with a 2,450 MHz generator (AMICA, Hospital Services, Aprilia, Italy).
Patient selections in both groups were similar. They must first have inoperable CRCPulM. Their CT scans and PET scans will then be reviewed at the weekly MDT meetings to decide if they would benefit from systemic chemotherapy, radiotherapy, or ablation therapy. Percutaneous MWA has now been the preferred method in this institution since 2013 in place of RFA.
Patient follow-up. All patients were admitted for a night for observation and a chest X-ray post-procedure. If uncomplicated, they will usually be discharged next day. Following discharge, patients will be followed-up in rooms with a repeat CT-chest 1 month after followed by intervals of 3 months for 3-5years. Carcinoembryonic antigen (CEA) levels were also checked at every follow up. Recurrence was defined as CT-proven new distant lung metastasis with increased size over time or growing lesion at the ablation site.
Statistical analysis was performed using RStudio (RStudio, Boston, MA, USA). Categorical variables are presented as frequency (proportions), while continuous variables are presented as median (interquartile range).
Results
A total of 161 (79%) patients underwent 204 (79%) RFA sessions to 442 (76%) lesions; and 42 (21%) patients underwent 54 (21%) MWA sessions to 139 (24%) lesions. There were 95 (46%) females in the RFA group and 15 (28%) females in the MWA group (p=0.013). The median age was 65 (57-73) years in the RFA group and 64.5 (59-69) years in the MWA group (p=0.512). General anaesthesia was administered to 136 (66%) patients in the RFA group and all patients receiving MWA (Table I).
Stratifying the two groups by primary site, 58 (43%) of RFA and 4 (12%) of MWA had primary cancer in their rectum; 47 (34%) of RFA and 12 (36%) of MWA had sigmoid primary; 10 (7%) RFA patients and 2 (6%) MWA patients had right-sided colorectal cancer. In terms of lung metastasis, 66 (73%) of RFA and 24 (27%) of MWA had bilateral lesions. One hundred and two (77%) of RFA patients and 30 (43%) of MWA patients had lesions on the right side.
The overall follow up time was 28 months (range=16-44 months); RFA group was followed for 32 months (range=18-51 months) and MWA group was followed up for 17 months (range=7-25 months). The median number of lesions ablated per session and ablation size calculated on CT were similar in both groups; 2 (range=1-3) lesions ablated per session in both groups and 4 cm (range=3-5 cm) in the RFA compared to 3.5 cm (range=3-4 cm; p=0.0395) in the MWA group. The median ablation time was significantly longer in the RFA group taking 49 min (range=26-65 min) compared to 8 min (range=5-13 min) in the MWA group (p<0.001). The median follow-up time was 32 months (range=18-51 months) and 17 months (range=7-25 months) in the RFA and MWA group, respectively.
The complication rate was 112 (55%) and 40 (74%) in the RFA and MWA group, respectively (p=0.011). Pneumothoraxes occurred in 57% of patients but rates were similar between groups (54% RFA vs. 66% MWA, p=0.106). Chest drains were inserted in 32 (15%) patients in the RFA group vs. 19 (35%) in the MWA groups (p=0.001). Median number of days the chest drain was inserted for was 3 days (range=1-8 days) and 1 day (range=1-2 days) in the RFA and MWA groups, respectively (p=0.026). Lung abscess occurred in 1 (0.5%) and 1 (2%) patient in the RFA and MWA groups, respectively (p=0.37). Similarly, rates of pneumonia were 4 (2%) and 2 (3.7%) patients in the RFA and MWA groups, respectively (p=0.6). Of note intrapulmonary haemorrhage occurred in 1 (0.5%) and 4 (7.4%) patients in the RFA and MWA group respectively (p=0.007). All patients were followed up with a CT scan after 1 month, 3 months, 6 months and then annually.
Median length of stay (LOS) was 2 days (range=1-3 days) for the RFA and 1 day (range=1-3 days; p=0.140) for the MWA group. Overall patient survival of the whole cohort was 37 months (Figure 1), where no significant survival difference at 1 or 3 years was found between both ablation techniques (92% for RFA vs. 95% for MWA; p=0.4 at 1 year and 51% for RFA and 28% for MWA; p=0.4 at 3 years; Figure 2). There were no differences in local recurrences (after discharge from initial ablation and found on follow-up CT) in the RFA (28%) group, where recurrence occurred 15 months (range=8-25 months) after RFA; compared to the MWA (28%) group at a time of 7 months (range=2-19 months) after MWA. No significant recurrence free survival differences were found between both ablation techniques (71% vs. 55% in RFA vs. MWA at 1 year, 32% vs. 27% in RFA vs. MWA at 3 years; p=0.1; Figure 3 and Figure 4). However, the MWA group demonstrated a higher survival rate and less recurrence rate than RFA in the first 24 months of follow up.
Overall survival rate of patients with colorectal cancer pulmonary metastases undergoing ablation.
Stratified survival rate of patients with colorectal cancer (CRC) pulmonary metastases undergoing radiofrequency ablation (RFA) and microwave ablation (MWA).
Recurrence-free survival (RFS) rate of patients with colorectal cancer (CRC) pulmonary metastases undergoing ablation.
Stratified recurrence-free survival (RFS) rate of patients with colorectal cancer (CRC) pulmonary metastases undergoing radiofrequency ablation (RFA) and microwave ablation (MWA).
Discussion
Pulmonary metastasis from colorectal cancer (CRCPulM) is the next most common site following liver metastasis. There has been an increase in detection since 1985 according to a study of colorectal cancer (CRC) statistics in the United States (8). It has been reported that 50% of patients with CRC will die from metastatic disease, 10-20% of metastasis occurs in the lung, and pulmonary metastasectomy provides a 20-40% survival (1-4, 8). However, surgery can only be offered to a small group of patients, as contraindications to surgery include more than 3 lung metastases, bilateral or deep lung metastasis, patients with significant comorbidities, other distant metastasis, short disease-free interval between primary resection and progression, discovery of pulmonary metastases whilst having chemotherapy and if the patient refuses to undergo surgery (1, 8, 9, 27, 30-32).
Radiofrequency ablation (RFA) has been the mainstay of ablation for pulmonary metastatic nodules and previous studies have shown that it is a safe and effective method of managing CRCPulM (1, 15, 18). RFA is especially effective as an alternative treatment for patients with CRCPulM that were inoperable because of number, distribution, location (multiple lobes or bilateral disease), poor performance status, or the patient’s refusal to accept surgery. Since its advent, microwave ablation (MWA) has been the treatment of choice for ablation of non-surgical lung and liver lesions. MWA has been purported to have theoretical advantages over RFA in that it is able to generate higher ablation temperatures, larger heating radius in the lung which is a poor conductor of heat in a much shorter time (8, 15, 21). A 2018 meta-analysis by Yuan et al. demonstrated similar safety profile in treatment of pulmonary metastases on unspecified primary malignancies, though suggested an improved 1, 2, and 3 year overall survival rate in RFA treated patients compared to MWA but was limited by small sample sizes (33). To date, there have been very few studies comparing RFA and MWA specifically in the treatment of CRCPulM. However, the available and emerging studies comparing both methods for CRC liver metastasis have shown that the superiority of MWA to RFA is still debatable (17, 34-37).
Our study has compared a total of 203 patients undergoing 258 ablation sessions (RFA and MWA) for 581 lung lesions. It is thus one of the largest studies to date to compare outcomes of CRCPulM ablation between RFA and MWA with excellent follow up of maximum of 15 years. With that said, it is prudent to note that MWA had only come into use and replaced RFA in 2015 in our center. Therefore, due to marked differences in size of the MWA population, consideration in regard to statistical analysis and comparison of length of follow up and rate of recurrence when comparing with the RFA group needs to be taken.
It has been reported that complication rates may be higher for patients who are having re-do ablations, who have had previous thoracic external beam radiotherapy or surgery due to disordered vasculature and poor tissue healing (15, 38). Complications of MWA are not uncommon, with the most common adverse effects reported by studies being pneumothorax and pleural effusion occurring at a rate similar to that of RFA (1, 33, 38). However, our findings show the complication rate in the MWA group (74%) was significantly higher than that of the RFA group (55%, p=0.01). Of note the rate of peri-lesion haemorrhage was significantly higher in the MWA group (7.4%) compared to that of the RFA group (0.5%, p<0.05). This was quite different from Vogl et al. who reported peri-lesion haemorrhage in 8% of the MWA cohort and 9.3% in RFA cohort, and from Aufranc et al. who reported pulmonary haemorrhage in 2.5% of the MWA cohort and 7.6% in the RFA cohort (p=0.16) (8, 22). Healey et al. found that not only were the odds of primary technical success 11 times less for tumours greater than 3cm, but the odds of complication also increased by 3% for every millimeter increase in the maximal tumour diameter (20). This explains the findings in Vogl et al. 2016, where the lesions complicated with peri-lesion haemorrhage were 3 cm or greater (8).
The MWA group had a significantly shorter length of stay compared to the RFA group of 1 day. This contradicts its higher rate of complications; however, our team has had more than a decade’s worth of experience with managing these complications and therefore would have managed them better in the MWA era. Also, the smaller absolute number of complications in the MWA group must be taken into consideration. Therefore, the overall length of stay can still be shorter than the RFA group even with more complications.
The Kaplan-Meier curves for stratified survival and recurrence rate have shown that the MWA group has a positive trend to statistical significance if given a longer follow up time. It has been suggested that the cooling effect of adjacent vessels or the ‘heat sink’ effect is reduced in MWA and thus recurrence rate is reduced (15, 17). Whilst this is an advantage of MWA, this may be the reason for the significantly higher amount of peri-lesion haemorrhage in MWA than RFA. In addition, MWA can propagate through lung tissues without having much impedance unlike RFA and can reach higher temperatures and ablate larger tissue volumes in a shorter time (17, 21). With that, it also increases the risk of haemorrhage compared to RFA.
Microwave antenna design has been purported to affect the size and shape of ablation zones. The different types of antennae have been described to control the energy delivery, thus allowing for more accurate ablation therapies and less complications (17). However, there has been consideration for more controlled microwave ablation by way of the use of bipolar rather than monopolar probes (39, 40).
Limitations. The main limitations of this study are the number and follow-up time of the MWA population, which has also been impacted by the recent global pandemic – COVID-19. Given a second review aimed at 5 years down the track, we will be able to better compare complications and provide statistically significant data on recurrence free survival and overall survival rate for the MWA cohort and more accurately compare it to the RFA group. There is also a time series bias introduced by the nature of the analysis. However, there is generally a trend over time for improved survival so when you compare the MWA period to the RFA period you would expect changes in long-term outcome due to emergent therapies. The learning curve is also steep with MWA and the issues with pulmonary haemorrhage, longer term follow up, and research and development need to be addressed to prevent this from occurring in the future.
Conclusion
RFA and MWA are two competitive alternative treatment methods for CRCPulM. MWA can be performed in a much shorter time with seemingly better short-term results in terms of overall survival rates and recurrence free rates than RFA. However, MWA has shown to have statistically significant higher complication rates than RFA. This may be due to the reduced ‘heat sink’ effect and larger probe size, although it may be prudent to factor in improved statistical comparison with time and increased numbers. More long-term studies, randomized trials and animal studies would be required to help further understand the benefits of MWA over RFA.
Footnotes
Authors’ Contributions
Charis Tan: Methodology, Investigation, resources, data curation, writing - original draft, writing - review & editing, visualization, project administration. Oliver M. Fisher: Formal analysis, writing - review & editing. Linna Huang: Writing - review & editing. Nayef Alzahrani, Winston Liauw, Derek Glenn: Supervision. David L. Morris: Conceptualization, supervision.
Conflicts of Interest
The Authors declare no financial or ethical conflicts of interest. No funding has been provided for the publication of this study.
- Received March 14, 2022.
- Revision received July 11, 2022.
- Accepted July 12, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.










