Bronchoscopic ablation for non-small cell lung cancer.

Introduction
Lung cancer is the leading cause of cancer-related death worldwide, with non-small cell lung cancer (NSCLC) accounting for 85% of cases (1). Long-term outcomes are best for patients with early-stage disease, where 5-year survival for patients with stage 1 NSCLC is approximately 75% (2). The prevalence of lung cancer has increased over time, with early-stage cancers accounting for the majority of these cases (2,3). Lung cancer screening programs for high-risk individuals and the increased utility of computed tomography (CT) imaging in medicine worldwide have increased the rates of lung nodule detection and the diagnosis of early-stage lung cancer (2,4).
Current evidence supports surgical resection as the best oncological therapy for early-stage NSCLC, with lobectomy being the standard surgical mode for the majority of patients. In most studies, 5-year survival following lobectomy for early-stage NSCLC is 64–77%, though recurrence is observed in over one-third (5,6). As the incidence of smaller tumours increases, approaches that preserve more lung have been sought. For patients with tumours less than 2 cm in size, segmentectomy has been shown to have similar outcomes to lobectomy with a 5-year survival of ~90% (7,8). However, some patients, due to either significant medical co-morbidities or a lack of physiological reserve, are unable to undergo surgical resection. In these patients, radiotherapy with stereotactic ablative radiotherapy (SABR) is standard of care and carries a 5-year overall survival of 84–87%, and a recurrence rate of 20–35% at 5 years (9,10), with some studies suggesting non-inferior outcomes compared to surgery (11). However, in a proportion of patients, the utility of SABR is limited by tumour size, location, and presence of underlying pulmonary disease such as pulmonary fibrosis.
Percutaneous thermal ablation techniques, including radiofrequency ablation (RFA), microwave ablation (MWA), and cryoablation, can be used in NSCLC when surgery and SABR are not an option. While there are no studies that have directly compared the outcomes of thermal ablation to SABR, large retrospective cohort studies have shown comparable long-term outcomes, especially if the lesion is under 2 cm in size (12-14). Percutaneous ablation has a significant side-effect profile, with pneumothorax rates of 30–40% (14-17), with about 13% needing chest tube insertion (14-16). Other complications include pleural effusion (5.2–9.6%), haemoptysis (3.9%), pneumonia (5.7%), respiratory failure (3.5%), and lung collapse (4%) (17). Bronchoscopic ablative techniques would theoretically have a lower risk profile, with some small feasibility studies demonstrating this previously (18-22), suggesting that a bronchoscopic approach could provide a suitable alternative for some patients. There has been a significant number of studies published over the last 5 years assessing various modalities of bronchoscopic ablation since our last review of the literature (23). Several studies have assessed the safety and feasibility of various modalities, including RFA, MWA, cryoablation, and pulsed electric field (PEF) ablation (Table 1).

Bronchoscopic RFA ablation
RFA delivers high-frequency alternating electrical current through a needle electrode, creating localized heat that leads to coagulative necrosis. Several studies have assessed the safety and feasibility of this approach, with several different catheters.
The largest prospective study to date was performed by Zhong et al. This was a prospective, single-arm, multicentre study assessing the safety and efficacy of a saline micro-perfusion RFA system after 12 months of follow-up. The saline micro-perfusion reduces local impedance and optimizes electrical conductance in the target zone. The study recruited 126 patients across 16 sites who declined or were not suitable for surgery, radiotherapy, or systemic therapy. Eighty-one percent of the patients had stage 1 lung cancer at the time of ablation, and six patients had lung metastasis from another solid organ tumour ablated. The primary endpoint assessed in this study was complete ablation, which at 12 months follow-up was observed in 90.48% of patients. At 12 months, progression-free survival was 88.98% and overall survival was 96.83%. Subgroup analysis demonstrated significantly higher rates of complete ablation and progression-free survival in patients with pure ground glass nodules, regardless of size. Rates of complete ablation and progression-free survival were better in patients with lesions less than 2 cm; however, this was not statistically significant (24).
A mortality rate of 1.59% was reported in this study, with two deaths observed. While one was due to tumour progression 9 months following ablation, the other death occurred due to massive haemoptysis at only 8 days after ablation of a 23 mm × 15.5 mm nodule (24). The lesion ablated was solid and located adjacent to a vessel and a cavity. This patient developed symptoms of pulmonary infection 24 hours after ablation, their condition deteriorated steadily despite anti-microbial therapy, and subsequently developed large volume haemoptysis 8 days post-ablation with no response to haemostatic therapy (24). CT imaging at day 7 post-ablation demonstrates a large cavitating lesion at the ablation site. Additionally, there were 5 (3.97%) patients with a pneumothorax, 8 (6.35%) with haemoptysis, 11 (8.73%) with pleural effusion, and 14 (11.11%) with a haemothorax post-ablation (24).
A second single-centre retrospective review of 46 patients with inoperable stage 1A NSCLC who underwent RFA with the same system has been published. This study had a median follow-up of 24 months and reported local progression-free survival rates at 1-, 2-, and 3-year as 87.5%, 73.4%, and 69.8% respectively. No deaths were reported in this cohort, and the safety profile was similar to the prospective study, with comparable rates of pneumothorax, haemoptysis, and pleural effusion (25).
Two smaller pilot studies have further demonstrated the feasibility of bronchoscopic RFA (26,27). One study utilised a flexible externally cooled RFA system (26), while the other used a needle-type RFA probe used with a convex probe endobronchial ultrasound (EBUS) bronchoscope (27). Both studies demonstrated a favourable safety signal overall (26,27). One patient required intensive care unit (ICU) admission post-procedure due to dispersal of heated saline into the endobronchial tree due to cough. Subsequent procedures were performed with neuromuscular blockade with no further complications reported (26). Patients in both studies underwent resection as standard of care, with ablation zones observed that were proportionate to the energy used (26,27).
Overall, these studies demonstrated the feasibility of bronchoscopic RFA in a technical sense, with a safety signal that appears favourable. However, the reported deaths in the largest prospective study warrant caution and further investigation, and we note with interest the proximity of the lesion in this patient to the large vasculature.

Bronchoscopic MWA ablation
Several studies have assessed a variety of flexible MWA probes and systems, demonstrating the technical feasibility of this technique (28-31). The largest prospective study was by Lau et al., who performed a single-arm, multicentre study where patients with confirmed NSCLC of less than or equal to 30 mm underwent ablation. Four patients (13.3%) experienced severe adverse events [Common Terminology Criteria for Adverse Events (CTCAE) grade 3], which included pleural effusion, chest pain, pleural effusion, haemoptysis, and post-ablation syndrome. No other serious adverse events were reported at 1-month follow-up, and no pneumothoraxes or deaths were observed. At 1-month post-ablation, all patients had CT evidence of complete ablation (28). This safety profile was similar to most other studies, with relatively low rates of serious adverse events. In these studies, the most prevalent adverse outcome was pneumothorax requiring chest tube insertion (30,31).
In terms of the long-term efficacy of bronchoscopic MWA, data are limited as the objective of most studies was to demonstrate safety and feasibility. In one retrospective study of 25 patients, at the time of publication, radiologic outcomes among 15 patients at 6 months included partial response in 12, stable disease in two, and one showing complete response as per modified Response Evaluation Criteria in Solid Tumours (mRECIST) criteria (31). Similarly, a study of 14 inoperable patients demonstrated a 2-year local control rate of 71.4% (30).
It should be noted that one study of an MWA catheter for bronchoscopic tumour ablation was terminated early after a severe adverse event resulting in patient death. This study was again assessing the safety and feasibility of a novel flexible MWA catheter in patients with inoperable peripheral lung cancer. This patient had an out-of-hospital sudden death at day 15 post-ablation. The ablated lesion was <10 mm in size and located in the middle third of the lung. It was reported that the patient had experienced chest pain post-procedure, and CT imaging at day 14 demonstrated a large cavitating lesion at the ablation site (29). This case highlights the potential risks of ablative therapy and the need for rigorous safety assessment.
Small retrospective series have explored specific scenarios in which MWA may be applied, including treatment of multiple lesions, ablation in combination with surgical resection, and for lung nodules in patients with a history of previous lung resection (32). The safety profile in these specific scenarios appears similar to that of other MWA studies, with the most prevalent adverse event being pneumothorax (32-34).
Overall, there has been rapid development in bronchoscopic MWA techniques over the last 5 years. However, larger prospective studies are needed to further assess the safety profile and efficacy of this technique.

PEF ablation
PEF ablation, where high-intensity electric pulses are used to induce target tissue injury, has gained prominence in the last few years. PEF appears to preserve adjacent connective tissue, pleura, and vasculature (41-43) and might provide a better safety profile to thermal ablative.
The Aliya PEF system (Aliya; Galvanize Therapeutics Inc., Redwood City, CA, USA) has been assessed in patients with both resectable and unresectable lung cancer. 36 patients with resectable NSCLC underwent PEF as part of the INCITE-ES study with no serious adverse events observed. Two patients had pneumothorax, one in the setting of percutaneous biopsy prior to ablation and the second noted post-ablation. Both were graded as CTCAE grade 1. All patients underwent uneventful surgical resection (37). A similar safety profile was seen in 30 patients with metastatic NSCLC who underwent PEF without resection. In this cohort, one patient had a pneumothorax needing intercostal catheter insertion (36). An additional retrospective study reported 4 pneumothoraxes in 41 consecutive patients that underwent PEF treatment for stage 4 NSCLC resistant to standard therapy (38).
The above studies did not aim for complete ablation, as safety was the primary endpoint. However, characterization of the post-ablation tissues on resected patients demonstrated a treatment zone characterized by a depletion of tumour cells and a variable degree of inflammation (37). Similarly, in patients who did not undergo resection, a CT chest 30 days post-resection showed minimal to no change (36). Progression-free survival and overall survival at 12 months post-PEF for 41 patients with stage 4 NSCLC were 63.2% and 74.3%, respectively, which was significantly better than a propensity-matched control group (38). Additionally, this study showed that the more tumours that were ablated, the better the overall survival (38).
The above studies demonstrate a favourable safety profile for PEF; however, further studies are needed to assess dose requirements and overall efficacy.

Bronchoscopic cryoablation
Cryoablation utilises the Joule-Thomson effect to achieve low temperatures. This is where the rapid expansion of a gas results in temperature changes. The degree of temperature change depends on the gas used, with argon, nitrogen, and carbon dioxide used commonly in medical cryoprobes (44). Percutaneous cryoablation, which primarily utilises argon, is widely used for tumour ablation (44), and flexible cryoprobes are used for bronchoscopic cryotherapy and cryo-recanalization (45) of large airway tumours and for bronchoscopic cryobiopsy (46,47). There have been very few studies assessing the utility of flexible cryoprobes for cryoablation of peripheral lung cancers.
Gu et al. published the only prospective study to assess the safety and feasibility of a bronchoscopic cryoablation (35). This was a single-arm, prospective study of patients with stage IA NSCLC or lung metastasis who were ineligible or refused resection. Cryoablation was performed using a novel flexible cryoprobe that used nitrogen. The system was initially assessed in an animal model, followed by cryoablation of 10 lesions in nine patients (35). Eight patients had confirmed lung adenocarcinoma, and one patient had lung metastasis from colon cancer. Technical success of the procedure, which was defined as an appropriate probe position within tumour and successfully completed ablation was achieved in all patients. At 6 months follow-up, seven patients demonstrated complete ablation with a reduction in tumour size, with two patients showing local progression. No deaths or major adverse events were reported in this study (35). Similar to the other modalities described, this study demonstrates the feasibility of bronchoscopic cryoablation as a technique. This study also shows a more favourable side effect profile, but the small sample size limits the ability to make this conclusion. Further studies with larger numbers are needed to assess efficacy and safety further.

Clinical utility
Most studies in the last 5 years have demonstrated the feasibility of bronchoscopic RFA, MWA, PEF ablation, and cryoablation. Several smaller first-in-human studies have demonstrated the favourable safety and feasibility of bronchoscopic thermal vapour ablation (BTVA) (39) and photodynamic therapy (PDT) (40) as well, with further data needed to assess its efficacy. While this represents a significant development in bronchoscopic techniques, many questions remain regarding the efficacy, optimal modality, and clinical utility in the current lung cancer treatment landscape.
The objective of most studies published to date has been to determine the safety profile and feasibility of each ablative technique and device. Unfortunately, the small patient numbers, short follow-up time points, and heterogeneous endpoints for efficacy make it difficult to comment on overall efficacy. Large, prospective trials are needed to assess the efficacy of each technique in comparison to standard-of-care treatment options. No literature to our knowledge directly compares different ablation modalities, and consequently, data at present does not allow for assessment of optimal ablative modality; and it should be considered that each ablative technique may represent a unique tool to use depending on the nuance of the case and patient in question. This will need to be further elucidated in future studies.
The utility of these minimally invasive ablative therapies is likely to be strongest for the treatment of small or multifocal tumours, where lung-sparing is of benefit. To further determine its role, robust data comparing this approach with surgical resection and radiotherapy are needed. The SABR experience demonstrates that recruitment to such studies may be challenging, and it may be some time before clear clinical evidence is obtained (48,49). Early efficacy data are likely, as was the case for SABR, to come from ablative therapy offered to patients with contraindications to established therapies (e.g., cardiorespiratory disease precluding resection, anatomic position, or pulmonary fibrosis precluding SABR). From this experience, we hope a clearer understanding of the patient and lesion characteristics suggestive of benefit might arise, which will inform future well-designed prospective trials.

Ablative immunotherapy—a novel paradigm
Thermal ablation techniques have demonstrated the ability to stimulate anti-tumour immune responses in several solid organ malignancies, with the potential to improve outcomes when combined with immunotherapy. Evidence from both preclinical models and clinical studies indicates that tissue necrosis from thermal ablation triggers an immunological cascade. The dying tumour cells release tumour-specific antigens (neoantigens) along with damage-associated molecular patterns (DAMPs)—cellular distress signals that activate the immune system. This leads to improved antigen presentation by dendritic cells, which then prime cytotoxic T-cells and recruit natural killer cells to the tumour site. The local inflammatory environment is further augmented by increased chemokine and cytokine production, creating conditions that may overcome the immunosuppressive tumour microenvironment and potentially improve the response to immune checkpoint inhibitors (ICIs) (50). While this “in situ vaccination” effect has been demonstrated in other tumour types, its applicability and efficacy in lung cancer remain inadequately characterised in clinical studies.
Harnessing the immune activation observed following thermal ablation may provide therapeutic benefit for patients with NSCLC. Response rates to ICI therapy are 30–45%, and 10–27% develop resistance after an initial response (51-54). Ablation could disrupt the tumour microenvironment, enhance T cell priming, and increase pro-inflammatory immune response, which may assist with improving the efficacy of ICI therapy (50). This could potentially increase pathologic complete response (pCR) rates in resectable patients if delivered prior to neoadjuvant chemo-immunotherapy, where higher pCR rates are associated with reduced long-term recurrence (55). Alternatively, it may be used to re-sensitise patients with metastatic NSCLC who have demonstrated progressive disease on ICI therapy. While this represents a novel potential treatment paradigm, further clinical trials are needed to determine its efficacy.
Studies following bronchoscopic thermal ablation of NSCLC have shown some evidence to suggest both local and systemic enhancement of anti-cancer immune responses, and that further studies exploring the potential for bronchoscopic ablation in NSCLC to augment ICI efficacy are warranted. Increased CD8+ T cells on day 1 post-cryoablation and increased antigen presentation and T-cell activation 7 days post-ablation (56) have been observed in patients undergoing thermal ablation. Additionally, an increase in PD-L1 expression within post-ablation tumour was reported in patients undergoing BTVA (57). Similarly, tissue post-PEF ablation has shown an increase in tertiary lymphoid structures (38), which has been associated with an improved response to immunotherapy (58).
Liu et al. demonstrated the safety of MWA with neoadjuvant chemotherapy and immunotherapy prior to resection (59). In this retrospective cohort of eight patients, higher rates of pCR were observed (59). Additionally, there are case reports demonstrating the safety of combining PEF ablation with ICI therapy (60).
Most recently, Tsay et al. performed cryoablation with a clinically available carbon dioxide-based cryosystem in 21 patients with stage 3 and stage 4 NSCLC. This was a dose-escalation study that again demonstrated the safety and feasibility of cryoablation as a technique. In addition, this study performed cytokine and flow cytometry analyses of blood obtained 14 days post-ablation, demonstrating an increase in pro-inflammatory cytokines and memory T cells (61). Desilets et al. further assessed the safety and feasibility of using this cryosystem in conjunction with immunotherapy in patients with advanced NSCLC (62). In this study, eight patients underwent cryoablation followed by ICI therapy. This study demonstrated the safety of combining ICI therapy with cryoablation; however, only two patients demonstrated a partial response to this technique (62). The safety and feasibility of combination ablation + ICI are encouraging, and methodological limitations related to the temperatures achieved with carbon dioxide as the cryogen may have been insufficient to trigger tissue necrosis; as a result, the degree of tissue injury may not have been sufficient in this study to achieve adequate synergy with ICI therapy. Heat-based ablation (e.g., RFA, MWA) or the use of cryogen gases with lower freezing temperatures (such as nitrogen or argon) should be evaluated in future studies in combination with ICI.

Conclusions
Current studies have shown the feasibility of bronchoscopic ablation in a technical sense, and the ongoing development of flexible ablation probes will only enhance this further. However, it is hard to draw significant conclusions regarding efficacy due to the small sample sizes, short follow-up periods, and heterogeneous efficacy endpoints used in most studies. Additionally, while the safety profile of bronchoscopic ablation appears favourable when compared to percutaneous ablation, the deaths observed in the RFA and MWA studies is a reminder that further work is needed to fully understand the safety profile of these modalities, a preferred modality is uncertain and requires further investigation, and should be assessed rigorously alongside trials examining the clinical efficacy of bronchoscopic ablation in the treatment of lung cancer. Large prospective trials, powered to assess efficacy, are the next steps needed to determine the utility of bronchoscopic ablative techniques. Furthermore, early evidence of its potential to enhance anti-cancer immunity and augment immunotherapy is encouraging and warrants further study, and may represent another role in the treatment paradigm of lung cancer.

Supplementary
The article’s supplementary files as