A Retrospective Study on Using a Novel Single Needle Cone Puncture Approach for the Iodine-125 Seed Brachytherapy in Treating Patients With Thoracic Malignancy

Background Patients with progressive thoracic malignancy characterized by large irregular tumors with necrosis and life-threatening symptoms lack effective treatments. We set out to develop a single needle cone puncture method for the Iodine-125 seed (SNCP-125I) brachytherapy, and aim to report the initial results. Methods 294 patients with advanced thoracic malignancy were treated with local SNCP-125I brachytherapy between March 2009 and July 2020, followed by thorough evaluation of clinical outcome, overall survival (OS), progression-free survival (PFS) and procedure-related complications after treatment. Results The overall response rate (ORR) among the treated patients was 81.0% (238/294). Life-threatening symptoms due to tumor oppression, hemoptysis and large irregular tumor with necrosis were successfully alleviated after the SNCP-125I treatment with a remission rate at 91% to 94%. The median OS and PFS were 13.6 months and 5.8 months, respectively. Procedure-related side effects including pneumothorax (32/294), blood-stained sputum (8/294), subcutaneous emphysema (10/294), puncture site bleeding (16/294) and chest pain (6/294) were observed. Patients who were able to follow with chemotherapy or immunotherapy experienced extended OS and PFS, as compared with patients who opted to receive hospice care (16.5 months Vs. 11.2 months). Further pathological and immunological analysis showed that SNCP-125I induced tumor lymphocytes infiltration and long-term tumor necrosis. Conclusion SNCP-125I brachytherapy effectively eliminates life-threatening symptoms due to local tumor oppression, hemoptysis and large irregular and necrotic tumors in patients with unresectable chest malignancy and significantly induces local tumor regression. SNCP-125I brachytherapy combines with chemotherapy significantly prolong OS and PFS compare with SNCP-125I brachytherapy alone.


INTRODUCTION
Successful treatment of local incurable thoracic malignancy, including lung squamous cell carcinoma, metastatic esophageal cancer, and unresectable malignant thymoma, has been hampered by the lack of clinically effective regimens. This represents a particular obstacle for patients who experience rapid tumor progression as a result of large irregular and necrotic tumors, hemoptysis and lethal symptoms, but are not susceptible to local interventiondue to limited access to the site of tumors. As such, despite systemic therapeutic treatment, coupled with advanced imaging technology, local lesions still reoccur and develop. Further, tumor pathology is closely associated with disease progression and therapy responses (1)(2)(3). Patients with squamous cell carcinoma exhibit unfavorable outcomes when treated with chemotherapy, immunotherapy, and molecularly targeted therapy, especially for central lung cancer and mediastinal tumors that are frequently surrounded by large blood vessels and trachea. This elevates the potential risk for performing local puncture operations. It should be further noted that the tumors often progress aggressively, thereby constricting adjacent trachea, blood vessels, heart and esophagus and leading to lethal dyspnea, hemoptysis, dysphagia and superior vena cava syndrome, which primarily accounts for the mortality in cancer patients (4)(5)(6)(7). Even though there is a low risk resulting from puncture operations for large tumors, the irregular necrosis, ribs, blood vessels and trachea often compromise the obstacle of the conventional interventional treatment. However, there is no effective approach to overcome these clinical challenges for patients with advanced malignancy, and the outcome after conventional treatment remains poor for those with squamous cell carcinomas of lung cancer, invasive thymic carcinoma, and chest metastatic tumors (8)(9)(10). Therefore, instead of hospice care, effective clinical strategies for these patients is urgently needed.
Computerized tomography (CT)-guided local radioactive Idione-125 ( 125 I) seed brachytherapy has been widely used for various types of advanced cancers, including lung cancer, uveal melanoma, breast cancer, malignant gliomas and retroperitoneal malignant tumors (11)(12)(13)(14)(15)(16)(17). Treatment planning systems (TPS) are employed in CT-guided local radioactive 125 I seed brachytherapy to ensure that the tumor site receives the maximum therapeutic dose while sparing surrounding tumor tissue, which represents one of the most effective approaches for maximum clinical benefit to the patient (18,19). Conventional multiple-needle 125 I brachytherapy method is only suitable for patients who can make required postures and have multiple parallel sites accessible for the puncture operation. However, this is not practical for patients with lung hilar and mediastinal tumors, because multiple needle puncture poses a great risk of damaging large blood vessels, and patients can quickly develop respiratory failure. In addition, additional criteria must be satisfied before the treatment. Foremost, the tumor lesion is located adjacent to the main bronchus, blood vessels, esophagus and heart. Secondly, the patient experiences the typical symptoms including dyspnea, hemoptysis, dysphagia, arrhythmia and superior vena cava syndrome as a result of local tumor oppression. Thirdly, the tumor is larger than 7 cm in size with irregular necrosis (20). Additional technical difficulties also prevents utilization and effectiveness of the multiple-needle brachytherapy, including lack of multiple puncture sites, uneven distribution of radioactive doses for large tumors, inability to maintain a posture for the operation, and extra poor performance status of these patients. Unfortunately, these patients are ultimately only able to choose hospice care with unfavorable survival outcomes.
To tackle this challenge, herein we report a single needle cone puncture method for the 125 I seed brachytherapy, with which we implement a radioactive dose covering over 90% of the tumor volume using a single needle through one puncture site on the skin. We show greatly reduced risk associated with the operation and satisfied recovery rate upon treatment, especially in patients with lung hilar and mediastinum tumors. Utilizing this novel intervention method, we successfully and effectively treated thoracic malignancy patients with large irregular tumors and severe symptoms due to local tumor oppression and hemoptysis (ORR=81.0%). Detailed pathological and immunological analysis uncovers that 125 I seed brachytherapy causes tumor necrosis within 15 to 20 days, associated with increased infiltration of tumor lymphocytes. Taken together, these results provide informative and critical insights into a new treatment strategy which helps to prolong survival for patients with advanced thoracic malignancy.

Patients Information and Characteristics
Two hundred and ninety-four patients with unresectable thoracic cancer showing severe symptoms resulting from tumor oppression received local radioactive 125 I seed brachytherapy from March 2009 until July 2020 at Tianjin Beichen Hospital (Tianjin, China) were investigated in this study. This study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Tianjin Anti-Cancer Association and ethics committee of Tianjin Beichen Hospital. All cases were selected according to the following inclusion criteria: 1) diagnosis with stage III/IV thoracic cancer with unresectable tumor; 2) confirmed malignancy with biopsy; 3) recurrence after conventional treatment including surgery, chemotherapy, immunotherapy and systematic radiotherapy, with no active treatment options available; 4) at least one type of the symptoms including dyspnea, hemoptysis, dysphagia, and super vena cava syndrome due to local tumor oppression, or tumor larger than 7 cm causing cachexia; 5) no history of chronic lung disease including pneumonia and Chronic Obstructive Pulmonary Disease (COPD); 6) no liver and kidney dysfunction, severe heart disease, impaired hematopoietic function or systemic infection. Patients were excluded if they did not meet the criteria above. Patient clinical characteristics are shown in Table 1. Of the 294 patients, there were 290 with squamous cell carcinoma and 4 with malignant thymic carcinoma ( Table 1 and Supplemental Figure 1). Of the 294 patients received 125 I seeds, 117 patients received chemotherapy, two patients received immunotherapy 3 to 6 months after treatment, and 175 patients did not receive any other therapeutic treatments during follow-up. All patients' data were collected from the hospital medical records which were described precisely, and those with incomplete information were excluded. Informed consents of all patients were obtained for the study.
Local Single Needle Cone Puncture-125 I seed (SNCP-125 I) Brachytherapy Procedure To achieve an accurate and dosimetric distribution of 125 I-seed implantation, treatment-planning system was applied for the 125 I treatment (TPS; standard version; Beijing ASTRO Technology Development Co., Ltd.) based on the American Association of Physicists in Medicine TG43 brachytherapy formalism (21)(22)(23)(24)(25). Dose, seed distribution, and depth of needles of each patient were determined preoperatively via TPS ( Table 2 and Supplemental Figure 2). Dosimetric evaluation parameters were: 1) D90, dose covering 90% tumor volume; 2) V90, tumor volume covered by 90% dose; 3) V100, tumor volume covered by 100% dose; and 4) V150, tumor volume covered by 150% dose. V90 ≥90% was considered as adequate dosage and distribution. Under general anesthesia, patients lay down in appropriate operational positions and an intraoperative stereotactic CT was performed. Thereafter, puncture paths were selected according to the preoperative planning under the CT scanning with a slice thickness of 5mm and seed brachytherapy of each needle channel was completed using 18G needle (15-20 cm\18Gneedle, Zhuhai Hejia Inc., China). Enhanced CT scan was generally applied during operation due to the complicated structure of mediastinum and lung hilar. Seeds distribution and dose was verified through TPS after operation, and there is no significant difference between TPS predicted and implanted dose ( Table 2). 125 I seeds used in this study was 4.50 ± 0.3 mm long, sealed and covered with an envelope of nickel titanium alloy (Atomic High-Tech Co., Ltd., Beijing). It is with an outer diameter of 0.80 ± 0.03mm, half-life of 59.6 days and an activity of 0.8 mCi. Summary of implanted 125 I seed parameters was shown in Table 3.
Particularly, we developed the Single Needle Cone Puncture method for the 125 I seed implantation (SNCP-125 I). We first selected one puncture site on the skin, withdrew the needle upon completion of 125 I seed implantation in the first needle channel until the needle tip was 1 to 2 cm from the tumor, then adjusted puncture direction avoiding blood vessels and performed the second puncture with the needle still inside the thoracic cavity. Similarly, the third or fourth puncture path was conducted inside the thoracic cavity according to the tumor shape, size and TPS plan, the needle direction was adjusted inside the thoracic cavity. Three needle channels form a three-dimensional cone shape covering over 90% of tumor volume. The end of each needle channel was close to be parallel with a minimum distance of 1.5cm between each two adjacent needle tracks to ensure the distribution of radioactive seeds covering most of the tumor area. Further, tumor with a diameter of 3 to 4 cm usually can be covered with one three-needle channel formed cone structure. It can be completed with several separate cones or superimposed  cones when tumors are larger than 4cm according to the shape of tumors. The ultimate goal is to make sure the radioactive sources as evenly distributed in the tumor as possible under the premise of patient safety. It is worth to note that, we always try to keep a minimal distance of 1.5 to 1.7 cm of the seeds and normal tissues/ organs to limit radiation dose outside the tumor and avoid radioactive complications.

Tumor Measurement and Observation of Therapeutic-Related Side Affects
CT scans were performed to measure tumor size pre-and 3 to 6 months post treatment. Tumor response evaluations were conducted according to the Response Evaluation Criteria in Solid Tumors (RECIST version 1.1) guidelines. Tumor size was measured by the sum of biggest diameter of all target lesions. Objective responses were defined as follows: CR, complete response; PR, partial response, described as a 30% decrease in the biggest diameters of all targeted tumors; PD, progressive disease, defined as new tumor appearance or a minimum 20% increase in the biggest diameters of all targeted tumors; and SD, stable disease, determined as tumor change between PR and PD. Procedure related side effects were recorded during treatment and follow-up according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 (26). Disease-related symptoms before treatment were not disclosed unless they worsened after 125 I brachytherapy.
To count CD8+ and Ki67 positive cells, the richest positive cell areas were identified at low magnification (×10) were selected, images were took at a magnification of ×20, and then CD8+ and Ki67 positive cells were evaluated quantitatively by two independent observers who analyzed five fields from these areas under a high-power (×40) field using an Olympus confocal microscope (Center Valley, PA, USA).

Follow-Up Assessment
All patients were followed from the date of the treatment of SNCP-125 I brachytherapy up to September 2020 or up to the time of death. Treatment-related adverse events were recorded during the treatment and follow-up.

Statistical Analysis
All statistical analyses were performed using the GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA). Survival curves and rates were calculated using the Log-rank (Mantel-Cox) Test or Gehan-Breslow-Wilcoxon Test, and survival was measured from the date of the treatment of local radioactive Iodine-125 seed brachytherapy up to September 2020 or the time of death. A student t test was used to analyze the statistical significance between groups. Spearman test was used to analysis correlations between groups. A p value less than or equal to 0.05 was the threshold used to determine statistical significance.

Clinical Outcomes of SNCP-125 I Brachytherapy-Treated Patients With Incurable Thoracic Malignancy
We investigated 294 patients with advanced thoracic malignancy who were diagnosed as squamous cell carcinoma and progressed on multiple types of conventional treatments ( Figure 1A).  Table 4). Patients with large irregular and necrotic tumors also show clinical improvements ( Figure 1D). Of the 294 patients, 117 patients underwent chemotherapy, 2 patients followed with immunotherapy and 175 patients opted to receive hospice cares 3 to 6 months post SNCP-125 I treatment ( Table 4). The 1-, 2-, 3-and 5-year overall survivals of these patients were 60.2%, 18.5%, 7.7% and 2.7%, respectively. The median OS and PFS were 13.6 months and 5.8 months ( Figures 1F, G). Procedure-related side effects of pneumothorax (32/294), bloodstained sputum (8/294), subcutaneous emphysema (10/294), puncture site bleeding (16/294) and chest pain (6/294) were observed ( Table 4). These results suggest that, as a further local treatment opportunity, SNCP-125 I brachytherapy can effectively and safely treat incurable thoracic malignancy, particularly for patients with inaccessible puncture sites, complications from tumor oppression and tumor bleeding. Importantly, further survival analysis showed that patients who were able to follow with chemotherapy or immunotherapy experienced extended overall and progression free survival, as compared with patients who opted to receive hospice care (16.5 months Vs. 11.2 months, p<0.0001; 6.8 months Vs. 5.2 Months, p<0.001, respectively, Figure 2). Basic clinical characteristic factor comparisons of these two groups showed no significantly difference ( Table 5). We also show greatly reduced risk associated with the operation and satisfied recovery rate upon treatment, especially in patients with lung hilar and mediastinum tumors. To further explain how the SNCP-125 I approach provides an optimal clinical benefit for patients, we show the detailed treatment process for representative patients below by different clinical symptom categories.

SNCP-125 I Brachytherapy Induces Rapid Tumor Regression of Large Recurring Solid Tumor After Conventional Treatment Failure
It is common that patients develop large lung squamous cell carcinoma and metastasis esophageal squamous cell carcinoma, and some even grow tumors on the skin surface with rupture and infection. These patients are often accompanied by serious systematic symptoms due to large tumor consumption including anemia, hypoalbuminemia, high fever and pain, with a poor performance status of Performance Score (PS) >3. Few traditional treatment strategies are currently available for these patients except hospice care. Particularly, Patient 53 was diagnosed as advanced mediastinal lung squamous cell carcinoma which compressed the heart and esophagus and caused arrhythmia and dysphagia ( Figure 3A). Patient 11 progressed on chemotherapy and radiotherapy, developed lung squamous cell carcinoma in the left lung with no druggable targets as revealed by DNA sequencing, and exhibited a poor performance status (PS=4) ( Figure 3A). Patient 282 was diagnosed as esophageal cancer with a large and ruptured  metastatic neck tumor, and experienced severe subclavian vein oppression, edema in the right upper limb, persisting fever, anemia and low platelets levels ( Figure 3B). We successfully administered local SNCP-125 I treatment for these and other 84 similar patients, and found that the tumors regressed in 3 weeks (Pt.53), 4 weeks (Pt.11) and 2 months (Pt.282) respectively ( Figures 3A, B, data not shown). The symptoms due to tumor oppression were subsequently alleviated, and the performance status greatly improved ( Figure 1E).

Severe Dyspnea Due to Thoracic Tumor Oppression was Completely Remedied by SNCP-125 I Brachytherapy
Patients with squamous cell carcinoma of the lung often develop dyspnea (~60%) and acute respiratory failure, when the tumor is located adjacent to trachea, especially to the large main bronchus (28) Among the 294 patients in the present study, 238 developed dyspnea, and showed a significantly correlation with the degree of tracheal stenosis ( Figures 1B and 4A) (27,29). The patients usually die in a short time if the airway cannot be recovered rapidly. However, they are unlikely to undergo conventional radiotherapy and chemotherapy as an optimal treatment strategy due to insensitivity to targeted drugs. Furthermore, the tumors are adjacent to trachea and no multiple puncture sites are readily accessible for conventional 125 I seed brachytherapy. Consequently, rapid control and elimination of local tumor lesions represents an effective means to completely relieve the respiratory distress and severe hypoxia. As shown in Figures 3C, D, as representative individuals, Patients 1, 7, 35 and 95 with progressive squamous cell carcinoma of the lung all experienced dyspnea due to airway obstruction, and continued to exhibit dyspnea, orthopnea, tracheal inhalation wheeze, and declined blood oxygen saturation. Remarkably, the symptoms due to airway obstruction gradually improved in two weeks after SNCP-125 I treatment and disappeared 1month after. The tumor significantly regressed and the main airway became significantly reconstructed in 1 to 3 months with a greatly improved cancer dyspnea score ( Figures 1C and 4B) (27,29). It is worth to note that it was changeling for the patients to lie  down and maintain a posture for more than 5 min due to breathing difficulties, thus limiting the operation time and requiring quick determination of the puncture site and completion of the seeds implantation within 5 to 10 min. Bleeding and severe pneumothorax during the operation will aggravate the breathing difficulty for the patients, and sufficient preparation for secondary complication and clinical emergency is needed before the operation.

SNCP-125I Brachytherapy Successfully Alleviated Fatal Hemoptysis in Advanced Lung Cancer Patients
Hemoptysis due to lung tumors is one of the most common causes for patient death, with a mortality rate of about 59% and We also effectively stopped bleeding using local SNCP-125 I treatment which primarily reduced local tumor load and restored the trachea (Figures 5B-E).

Dysphagia, Arrhythmia, and Superior Vena Cava Syndrome Caused by Mediastinal and Lung Tumors Oppressions Is Recovered by SNCP-125 I Brachytherapy
Malignant aggressive thymoma is a rare epithelial tumor that often occurs in the anterior superior mediastinum. Thymoma is categorized into four stages according to the  oppressed the heart and caused arrhythmia which was recovered after SNCP-125 I treatment ( Figure 6A). Patient 97 was also diagnosed as malignant aggressive thymoma who was insensitive to radiotherapy and chemotherapy, experienced compression of the superior vena cava causing swollen head, face, and upper limbs and superficial venous dilation of the chest wall ( Figure 6B). Following SNCP-125 I brachytherapy, the tumor significantly regressed and superior vena cava syndrome was completely alleviated ( Figure 6B). Dysphagia as a result of tumor compression is another difficult-to-cure symptom. The lung tumor in Patient 37 compressed the laryngeal nerve and caused dysphagia, and experienced choking after eating. The tumor in this patient also invaded both pulmonary artery and trachea causing dyspnea ( Figure 6C). Patient 34 was diagnosed as mediastinal lung cancer with the tumor compressing the esophagus and causing dysphagia ( Figure 6D). Dysphagia was completely remedied in these two patients within a month after SNCP-125 I treatment, and this was followed by tumor regression ( Figures  6C, D). Herein, we successfully treated 54 patients who developed severe dysphagia. Taken together, the findings demonstrated that, instead of supportive care, SNCP-125 I brachytherapy is an effective approach to reduce tumor burden when treating the life-threatening symptoms due to thoracic tumor compression.

SNCP-125 I Brachytherapy Is an Effective Approach in Treatment of Thoracic Malignancies With Limited Anatomic Puncture Site
As described above, SNCP-125 I was effective in treating incurable thoracic malignancies. We herein highlight multiple rationales underlying the design of the SNCP-125 I method. Foremost, tumors located in the mediastinum or hilus of the lung are normally blocked by blood vessels and trachea and thus become inaccessible to puncture. Particularly, only one accessible CT scan layer can be the puncture site with an available insertion gap of about 0.5 to 1 cm, for which typical multiple-needle puncture-125 I brachytherapy is difficult to administer. Representative CT images of Patients 1, 12, 37, 42 and 95 were showing in Figures 7A-E. Secondly, patients with poor performance status require the operation to be promptly carried out, but conventional multiple needles puncture-125 I brachytherapy usually takes over 1 h, longer than what the patients can withstand. In addition, patients cannot maintain a proper posture required for the operation because of dyspnea To avoid damage to large blood vessels, only one CT layer could be the gap for needle insertion, which was less than 1 cm (c). The puncture had to pass through the whole left lung (10 cm from the tumor) that only allowed a single needle insertion (c). The first puncture was done under the aortic arch (d), and the second and third puncture through the right pulmonary artery by adjusting the needle in the aortic window without exiting the vascular space (e). (B) A lung tumor in patient 1 in the aortic window compressed the main trachea. The patient was only able to lie on the left side due to breathing difficulty, and this greatly limited the operation time (a-c). The first puncture was done through the intercostal space and the aortic window (d), and the second and third punctures by adjusting the direction of the needle from the edge of the diaphragm (e). due to tumor oppression. Given these practical challenges, SNCP-125 I becomes a feasible and effective approach which works though only one puncture site and can be completed in 5 to 10 min. Furthermore, for patients who were treated with bronchial stent but re-developed tracheal stenosis, SNCP-125 I brachytherapy still restored the tracheal stenosis (Supplemental Figure 5). An example CT scan showing how two punctures were carried out using a single needle is shown in Figure 8A. Patient 37 is shown as an example to explain how the threedimensional cone shape formed using a single needle. As illustrated in Figure 8B, three seed paths covered 90% of the tumor volume by forming a three-dimensional (3D) tapered path as verified through TPS plan ( Figure 8C). Afterward, we selected one puncture site, implanted seeds into three channels by withdrawing, adjusted the direction of the needle and performed the other two punctures inside the tumor, and a 3D tapered path then formed as shown by the 3D reconstructed seeds ( Figure 8D). As described in the method, It is important to make sure the distal ends of each channel as parallel as practical to ensure an evenly distribution of radioactive sources in the tumor. Three more representative cases of Patient 11, 256 and 284 treated with SNCP-125 I were shown in Figure 8E. It was worth to note that the numbers of total puncture channels and formation of the cones were associated with the size of treated tumor (Supplemental Figures 6A, B). Taken together, SNCP-125 I brachytherapy is an effective method for incurable thoracic cancer patients with mediastinum or lung hilum invasion and inaccessible anatomical puncture site.

125
I brachytherapy induces rapid tumor regression ( Figure 1D) (11)(12)(13)(14)(15)(16)(17), but the pathological feature of irradiated tumor cells remains poorly defined. We thus assessed pathological alterations in tumor tissues from eight patients before and after SNCP-125 I brachytherapy. We found that the majority of tumor cells underwent necrosis in 15 to 20 days after treatment ( Figure 9A). After 2 or 3 months, no or little live cells were observed in the treated tumor site, which is usually difficult to confirm via CT scans ( Figure 9B). Surprisingly, after 9 months, the remaining tumor cells became fibrotic ( Figure 9C), although CT scans still show a high density shadow in the treated tumor area in most cases. All pathological alterations were further confirmed by staining with tumor specific markers p40 and p63 for squamous cell carcinoma that treated tumor fails to identify any residual cancer cells 2 to 3 months after treatment ( Figures 9A-C). proliferation was significantly suppressed after SNCP-125 I brachytherapy (p=0.012, Figures 10A, B and Supplemental Figure 7A). This result suggests that 125 I brachytherapy induces tumor necrosis and fibrosis to eliminate tumor cells in a long-term manner, in line with the rapid tumor regression as observed through CT images ( Figure 1D).
In vitro studies have revealed that systematic radiotherapy increases T-cell infiltration (35,36). We therefore asked whether SNCP-125 I brachytherapy plays a role in tumor microenvironment alteration in the patients. We found that tumor infiltrated CD8+T cells were significantly increased 4 weeks after SNCP-125 I brachytherapy, while CD4+T cell tumor infiltration showed no change (p=0.043, Figures 10C, D and Supplemental Figure 7B). These results provide an additional mechanistic rationale for combining 125 I brachytherapy with immunotherapy that modulates tumor infiltrated CD8+T cell in the clinic. However, only a small population of patients was evaluated, and increasing numbers of patients are needed to confirm this result. In addition, the dynamics of CD8+T cells infiltration remains another key question that needs to addressed.

DISCUSSION
125 I brachytherapy is known to be safe and effective in advanced cancer patients (11)(12)(13)(14)(15)(16)(17), and the multiple-needle parallel puncture method is the most common and widely used treatment regimen for 125 I brachytherapy (11)(12)(13)(14)(15)(16)(17)37). It requires multiple accessible puncture sites without obstruction, and is suitable for tumors with relatively regular shapes. The treatment can be carried out with technical ease, and radioactive 125 I seeds can be readily distributed evenly through multiple paralleled needles with an equal distance in between. However, critical challenges remain as it may cause severe lung injury with bleeding and pneumothorax, especially for lung hilar and mediastinal tumors (38)(39)(40). Furthermore, this approach is not applicable to larger tumors with severe irregular ulcers ( Figure  3B), and the overall survival benefit of this treatment needs to be further determined in future controlled studies involving large populations of patients.
Herein, we present the clinical and pathological outcomes of 294 patients treated with SNCP-125 I brachytherapy that we developed for incurable thoracic cancers. All the patients were stage III/IV thoracic malignancies, with 289 developing lung hilar and mediastinal invasion, 240 experiencing vascular and tracheal invasion, and 283 showing a performance score of 3 or 4 indicative of a high death risk. To design effective treatment strategies urgently needed to for these patients, we developed SNCP-125 I brachytherapy which represents a highly localized treatment option and rapidly reduces local tumor burden (ORR of 81.0%), thereby significantly improving quality of life and extending treatment window for the patients ( Figure 1E and Table 4). In this retrospective clinical study, we report that SNCP-125 I brachytherapy produced a median overall survival of 13.6 months. Moreover, we found that SNCP-125 I brachytherapy combines with chemotherapy is more efficacious than SNCP-125 I brachytherapy alone. Importantly, we show our procedures are safe, easy to deploy, and improved quality of life for this group of patients. SNCP-125 I brachytherapy thus represents an effective and unique treatment strategy that offers a major clinical benefit for patients with incurable mediastinal and lung hilum malignancies accompanied with life-threatening symptoms. This SNCP-125 I method is promising and advantageous for the following reasons: 1) multiple puncture routes can be completed with only one puncture site and one needle, and this is carried out by withdrawing and adjusting the needle direction inside the thoracic cavity; 2) it is effective treating incurable mediastinal and lung hilum tumors and large tumors with irregular ulcers; 3) it enables rapid control of lifethreatening symptoms as a result of local tumor oppressions including dyspnea, hemoptysis, dysphagia and super vena cava syndrome; 4) it leads to minimal collateral tissue damage and operation-related complications ( Table 4); 5) The procedure can be completed promptly within 5 to 10 min for patients who developed severe dyspnea; and 6) it can restore tracheal stenosis even after ineffective bronchial stent treatment (Supplemental Figure 5). In addition, there are several alternative approaches can be used to control the life-threatening symptoms of these patients: 1) tracheal stent placement can be used to relieve dyspnea. However, the tumor burden is not resolved, as the tumor progresses, the airway will restenosis, and the stent can also cause expectoration difficulty by continuously stimulation of the endotracheal lining. Therefore, patient's life of quality and overall survival will not be much improved; 2) interventional vascular embolization technology can be used to control hemoptysis. Blood supply of lung tumors is usually supported by both pulmonary artery and vein. Therefore, interventional embolization is not effective as the tumor changes following treatment that will likely cause hemoptysis occurs again; 3) esophageal stent implantation is usually used to treat dysphagia. Dysphagia of these patients usually caused by mediastinal tumor compression, however, esophageal stent implantation does not treat solid tumors, but temporarily solves the eating difficult. Re-stenosis usually occurs in a short time as tumor grows, and the stent implantation will bring a lot of pain to the patients. By comparing with these approaches, the SNCP-125 I brachytherapy treatment solves the problem of tumor compression by shrinking solid tumors in a relatively long time; thus, the curative effect, overall survival and patient quality of life are much better improved.
It is worth to note that multiple technical details must be paid attention to for SNCP-125 I brachytherapy to succeed: 1) the puncture paths need to be design to avoid thick blood vessels and trachea, as otherwise blood vessel injury causes hemoptysis, and trachea injury causes cough and pneumothorax during and after the operation; 2) the distance between the puncture sites on the skin and the tumor should be as short as possible, which will greatly minimize the possibility of tissue injury and puncture deviation; 3) keep the needle and the predicted puncture paths (D) Quantification data showed a significant increase in tumor infiltrated CD8+T cells after treatment (p=0.043, n=4). A student t test was used to analyze the statistical significance before and after treatment. All images were taken in a ×20 magnification and zoomed out. Data were collected under a high-power (×40) field. HPF, High power field. *, p<0.05; **, p<0.01.
relatively static following the breath floating of patients to avoid puncture deviation; 4) for new operators, can insert the needle in stages and correct in real time the direction of the needle to avoid blood vessels and trachea as found appropriate. For example, for a predicted puncture channel of 10 cm, insert the needle for 3cm to 4cm first, stop and check if the puncture direction is correct, adjust needle direction as necessary and insert another 3cm to 4cm until 10cm. It is also important to note that for lung cancer patients, the lung tissue is elastic and eligible for sufficient safe angles for needle adjustment once the tip of the needs nearly reaches to the tumor; 5) the end of each needle channel is designed to be close to be parallel with a minimum distance of 1.5cm between each two adjacent needle tracks to ensure the distribution of radioactive seeds covering most of the tumor area; and 6) the aortic window is a narrow path (normally 1 cm wide) surrounded by the aortic arch and pulmonary artery which can be used as an important path to insert into the mediastinum to avoid blood vessels. Moreover, we assessed the impact of SNCP-1 2 5 I brachytherapy on tumor pathological and tumor immune microenvironment, and found long-term tumor cell necrosis and fibrosis, and increase tumor CD8+T cell infiltration after the treatment. These results further provide a mechanistic rationale for further exploring combination strategies manipulating tumor-infiltrated T cells. In conclusion, SNCP-1 2 5 I brachytherapy is a unique, feasible and effective minimally invasive therapy for patients with incurable thoracic malignancies, especially for those who develop life-threatening symptoms due to tumor oppression and exhibit limited numbers of accessible anatomical puncture sites for the conventional multiple-needle 125 I brachytherapy.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Tianjin Beichen Hospital Ethics Committee. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
FL and XD designed the study. FL wrote and prepared the manuscript with assistance and feedback of XD. FL, LW, and YZ conducted statistical plan and analyzed the data. TJ and WF collected the data. XD performed the imaging, treatments, and biopsies. ZW provided reagents or performed the histological experiments. ZL determined pathological diagnosis. XD read and measured the patient CT scans. XD provided project oversight.
All authors contributed to the article and approved the submitted version. Puncture paths were designed following several criteria: 1) avoid thick blood vessels and trachea as practical as possible; 2) the distance between the puncture site on the skin and the tumor should be as short as possible, so as to decrease the risk of tissue injury and puncture deviation; 3) try to keep the end of each needle channel parallel with a minimum distance of 1.5cm between each two adjacent needle tracks to ensure that the distribution of radioactive seeds covers most of the tumor area; 4) the aortic window is a narrow path (normally 1cm wide) surrounded by the aortic arch and pulmonary artery that can be used as a path to insert into the mediastinum to avoid blood vessels. Magenta line-predicted puncture needle path, Red dots-125 I seeds. CT images showed that the right trachea (yellow arrow) and super vena cava (red arrow) were severely compressed by the tumor, and this was alleviated 3 months after SNCP-125 I brachytherapy. (B). Patient 211 with lung squamous cell carcinoma experienced both hemoptysis and dysphagia. CT scans showed that esophagus was compressed and the right lung hilum was invaded by the tumor, and this was alleviated in 3 months and until 12 months.
Supplementary Figure 5 | An example of tracheal stenosis relapse after tracheal stent implantation in a lung cancer patient. (A). Relapse tracheal stenosis in Patient 250 was successfully controlled by SNCP-125 I brachytherapy in 3 months and tracheal reconstruction lasted until 12 months (yellow arrow). Thrombosis was found in the vena cava stent (red arrow), which was not curable with collateral circulation established. (B). 3D reconstruction of tracheal stent (white arrow) and vena cava stent (red arrow), and tracheal stent (white arrow) was seen under bronchoscopy 12 months after treatment.
Supplementary Figure 6 | The tumor size in treated patients is significantly correlated with the number of puncture channels (A) and the number of 3D cones formed (B). Spearman correlation test was used for the analysis. Figure 7 | Raw microscopy images on tumor core biopsy tissues staining with Ki67 (A) and CD8 (B) for Figure 10. All images were taken in a ×20 magnification.