Ablative Procedures and Additional Anticancer Therapy Prolong the Survival in Patients With Endoluminal or Mixed-type Malignant Central Airway Obstruction: A Single-institution Retrospective Study

Background In patients with malignant central airway obstruction (MCAO) receiving transbronchial interventions (TBIs), it remains unclear if the prognosis after the intervention might differ according to the bronchoscopic appearance of the airway obstruction. Methods TBIs were undertaken in MCAO patients with endoluminal obstruction (TM group, n = 19), extraluminal obstruction (EX group, n = 19) and mixed-type obstruction (MX group, n = 23), under moderate sedation and high fractions of inspired oxygen (FiO 2 ). We evaluated the differences in the overall survival period (OS) after the TBIs among the 3 groups.


Introduction
Malignant central airway obstruction (MCAO) is a complication sometimes in patients with advanced non-small cell lung cancer (NSCLC) or with lung metastases from primary malignant tumors at other sites; it is a serious and life-threatening complication, and can cause death from suffocation, if left untreated. MCAO is generally de ned as >50% obstruction of the trachea or mainstem bronchi and is classi ed as endoluminal obstruction, extraluminal obstruction or mixed-type obstruction [1] [2] [3].
Transbronchial interventions (TBIs) can relieve respiratory distress and additional anticancer therapy after TBIs has been reported to improve the quality of life in patients with MCAO [4] [5] [6] [7]. Laser ablation, electrocautery and argon plasma coagulation are commonly used ablative procedures in MCAO patients with endoluminal obstruction caused by ingrowing tumors into the lumen of the airway, and stent placement is employed in patients with extraluminal obstruction caused by tumors growing adjacent to the airway. In patients with mixed-type obstruction, an ablative procedure followed by stent placement is usually adopted.
Some authors have reported that the overall survival period (OS) after TBIs in MCAO patients is longer in patients with endoluminal obstruction than in those with extraluminal obstruction or mixed-type obstruction [6] [8] [9], presumably because patients with endoluminal obstruction have less severe airway obstruction or lesions that are more amenable to treatment by ablation (e.g., polypoid lesions). There are also some reports of the e cacy of ablative procedures in MCAO patients with endoluminal obstruction and respiratory failure (severe airway obstruction) [10] [11].
We speculated that there might be more reasons for the better prognosis after TBIs in MCAO patients with endoluminal obstruction than that these patients tend to have less severe air obstruction and that the lesions tend to be more amenable to treatment by ablative procedures. Therefore, we conducted this retrospective study to evaluate the difference in the OS after TBIs among the MCAO patients with endoluminal obstruction, extraluminal obstruction and mixed-type obstruction, and to investigate the prognostic factors in MCAO patients treated by TBIs.

Patients And Methods
This retrospective, single-institution study was conducted with the approval of the Institutional Review Board of Kumamoto Regional Medical Center (approval date, June 16, 2021; approval number, 21-003).
The data of 65 patients with advanced NSCLC or lung metastases from other primaries with symptomatic MCAO were retrieved from our medical records database for the period between October 1, 2013, and December 31, 2020. Some of the patients who underwent ablative procedures have been included in previously published data [11].
We performed transbronchial microwave ablation (TMA) as the ablative procedure in MCAO patients. Microwave induces polar molecules (mainly H 2 O) in the tissue to rotate and produce dielectric heat in an electrical magnetic eld at a frequency of 2450 MHz around the monopolar antenna (electrode). TMA therapy can be performed more safely in MCAO patients with respiratory failure with high fractions of inspired oxygen (FiO 2 ) without increased risk of airway re or accidental penetrating injury, as compared to other ablative therapies, because the tissue temperature rarely exceeds 100℃ (boiling point of water) during TMA [12].
In the patients enrolled in our study, TMA therapy was performed using a Microwave Tissue Coagulator (Microtaze® OT-110M or AFM-712, Alfresa Pharma Co., Osaka, Japan) and stent placement was performed using an Ultra ex™ Tracheobronchial Stent System (Boston Scienti c Co., Tokyo, Japan), under moderate sedation (intravenous administration with midazolam at the starting dose of 3-4 mg) and high FiO 2 values (supplemental oxygen via an oxygen mask at a ow rate of 5-10 L/min). In patients with MCAO with life-threatening respiratory failure, TBIs performed under moderate sedation could save time for general anesthesia in the operating room [11]. TMA therapy and stent placement are considered successful if at least 50% airway patency is restored [1] [2] [3] within an hour of initiation of the rst session of the TBI.
We strati ed the total of 61 MCAO patients who underwent TBIs into 3 groups according to bronchoscopic appearance of the MCAO, as follows: patients with endoluminal obstruction (TM group, n = 19), patients with extraluminal obstruction (EX group, n = 19), and patients with mixed-type obstruction (MX group, n = 23) (Figure 1). The endpoint of the study was the differences in the median OS among the TM group, EX group, and MX group. OS was de ned as the period from the initiation of the TBIs until patient death. Comorbidities at the time of the initiation of TBIs were evaluated according to the Charlson comorbidity index [13]. Presence of respiratory failure was de ned as a ratio of the partial pressure of oxygen in the arterial blood (PaO 2 ) to the FiO 2 (PaO 2 /FiO 2 ) of <300.
Statistical analysis was performed using the Stat View J 5.0 statistical program (SAS, Institute Inc., Berkeley, CA, USA). One-way factorial analysis of variance and multiple comparison tests were used to compare three independent parameters. Analysis of categorical data was performed with the χ2 test. The log-rank test was performed to identify the clinical variables potentially associated with the OS. Variables identi ed as being signi cant with p values of < 0.10 were entered into a Cox proportional-hazards model for multivariate analysis. The hazard ratios (HRs) and corresponding 95% con dence intervals (CIs) were determined. OS was estimated by the Kaplan-Meier method. A two-tailed p-value of less than 0.05 was considered as being indicative of a statistically signi cant differences.

Patient and disease characteristics
There were no signi cant differences in the median age of the patients, percentages of patients with an Eastern Cooperative Oncology Group performance status (PS) score of 3-4, proportions of patients with respiratory failure, distributions of patients with different pathological types of lung cancer or different sites of origin of the metastatic lung tumors, or proportions of patients with a previous history of anticancer therapy among the 3 groups. The symptoms of dyspnea on exertion, cough/bloody sputum, and stridor were the main indications for TBI (Table 1).

E cacy and complications of TBIs
The most common site of MCAO was the trachea/tracheal bifurcation in all the 3 groups. The initial TBI procedure was TMA therapy in the TM and MX groups, and stent placement in the EX group. The successful airway patency restoration rate (95% vs. 74% or 89%) as well as the percentage of patients in whom the symptoms/PS improved after the TBIs (84% vs. 65%/74%) was higher in the TM group than in the MX group/EX group. Seven (32%) of the patients in the MX group received stent placement as second-line therapy, because of failure of TMA to successfully restore the airway patency (Table 2).
There were no differences in the frequencies of complications during the TBIs among the 3 groups. Three patients required tracheal intubation during the TBIs due to hypoxemia, and 2 patients developed airway re-occlusion because of bronchial epithelial detachment within 24 hours of the TMA therapy, as a result of the normal bronchial mucosa having been accidentally coagulated during the TMA procedure. The frequency of complications after the TBI (>24 hours) was statistically signi cantly higher in the EX group than in the TM group/MX group (37% vs. 0%/13%, p = 0.008) ( Table 2). Regarding the frequency of complications after stent placement as the TBIs, complications occurred in 9 (35%) of the 26 patients who received stent placement and 1 (3%) of the 35 patients who did not receive stent placement.

Post-TBI anticancer therapy
The percentage of patients who received post-TBI anticancer therapy (79% vs. 30%/58%) and the median number of anticancer therapy regimens (2 regimens vs. 1 regimen/1 regimen) was higher in the TM group than in the MX group/EX group. The reasons for not administering post-TBI anticancer therapy were that the patients refused to receive post-TBI anticancer therapy or their PS was poor, precluding post-TBI anticancer therapy (Table 3). In the TM group, 2 patients (11%) with lesions that were amenable to treatment by ablation alone did not need post-TBI anticancer therapy; a 72-year-old woman with a single endoluminal metastatic tumor from postoperative colon cancer, and a 56-year-old man with an endobronchial carcinoid in the left main bronchus.

Survival
During the median follow-up period of 5.6 months (range, 0.4 to 95.6 months) after the TBI, 63% of patients in the TM group, 83% in the MX group, and 89% in the EX group died ( Table 3). The OS was longer in the TM group than in the MX group/EX group (18.5 months vs. 3.8 months/4.5 months, p = 0.001) (Figure 2-a). Multivariate analysis identi ed adoption of TMA as the initial procedure (HR 0.294, 95%CI 0.113-0.770, p = 0.013), successful airway patency restoration after the TBI (HR 0.239, 95%CI 0.106-0.538, p = 0.005) and post-TBI anticancer therapy (HR 0.194, 95% CI 0.087-0.431, p < 0.001) as signi cant factors associated with a reduced risk of death (Table 4).
Among the patients who received post-TBI anticancer therapy, the OS was longer in the TM group/MX group than in the EX group (27.2 months/32.9 months vs. 6.0 months, p = 0.011) (Figure 3-b). Among the patients who received best supportive care (BSC) alone, the OS tended to be longer in the TM group/MX group than in the EX group (3.2 months/3.1 months vs. 1.4 months, p = 0.072) (Figure 3-c).

Discussion
Our results showed that adoption of the TMA procedure as the initial procedure, successful airway patency restoration following the TBI and post-TBI anticancer therapy were factors that were independently associated with a reduced risk of death in MCAO patients, and that the OS tended to be longer in the TM group and MX group than that in the EX group both in the subgroups of patients that received post-TBI anticancer therapy and in the subgroups of patients that received BSC alone.
Many studies have already reported that additional anticancer therapy after TBI such as chemotherapy or radiotherapy, improves the quality of life in patients with MCAO [4] [5] [6] [7]. In the present study, the percentage of patients who received post-TBI anticancer therapy was lower in the MX group and EX group than that in the TM group; the reason was the patient refusal or poor PS. This could explain the high frequency of complications after TBI in the EX group (Table 2). Complications after the TBI occurred in 35% of the patients who received stent placement and 3% of the patients who did not receive stent placement. Although stent placement can be performed more quickly and safely as compared to ablative procedures, long-term complications associated with an airway foreign body, including bacterial pneumonia, mucus plugging of the stent, stent migration, or massive hemoptysis caused by a tracheoarterial stula have also been reported after stent placement [14] [15].
In our study, the median OS time was 18.5 months in the TM group, 3.8 months in the MX group and 4.5 months in the EX group, consistent with previous reports [8] [9] [14] [16]. In addition, in the TM group, 30% of the patients had respiratory failure (severe airway obstruction), and 89% of the patients needed post-TBI anticancer therapy. We suggest that TMA therapy itself decreases the risk of death in the TM group, because ablative procedures can directly decrease the volume of the ingrowing tumors into the lumen of the airway without placement of an airway foreign body. Stent placement is the best procedure for MCAO patients with extraluminal obstruction. However, as Ost et al [17] points out, stent placement for its barrier effect should be undertaken after airway patency restoration by ablative procedures in those with endoluminal or mixed-type obstruction.
Our study had certain limitations. Firstly, the sample size was small because this was a retrospective study conducted at a single institution. Secondly, we performed TMA therapy as the initial ablative procedure in all the patients of the TM group and MX group, even though laser ablation, electrocautery and argon plasma coagulation are also commonly used ablative procedures [1]. As we have previously reported that TMA therapy is well-tolerated and effective in patients with MCAO [11], we do not believe that adoption of this procedure would have affected our results. Thirdly, we could not advise all the patients about the bene ts of post-TBI anticancer therapy, because some patients were admitted to our hospital for the purpose of TBIs alone. If a greater number of patients in the MX group had received post-TBI anticancer therapy, a better prognosis might have been achieved in all the patients of the MX group.
In conclusion, it is bene cial for MCAO patients with endoluminal obstruction or mixed-type obstruction to receive post-TBI anticancer therapy following ablative procedures.

Declarations
Compliance with ethical standards

Funding
No funding was received from any source for this work.