RFA对于≤3cm的结肠癌肺转移瘤的作用

Annals of Surgical Oncology 14:1718-1726 (2007)
Treatment Failure After Percutaneous Radiofrequency Ablation for Nonsurgical Candidates With Pulmonary Metastases From Colorectal Carcinoma
Tristan D. Yan, BSc (Med), MBBS1, Julie King, MPH1, Adrian Sjarif, BSc (Med), MBBS1, Derek Glenn, MBBS2, Karin Steinke, MD1,2, Ahmed Al-Kindy, MD1 and David L. Morris, MD, PhD1
1 Department of Surgery, University of New South Wales, St. George Hospital, Sydney, New South Wales 2217, Australia
2 Department of Radiology, University of New South Wales, St. George Hospital, Sydney, New South Wales 2217, Australia

Approximately 10% of patients with colorectal cancer develop pulmonary metastases. Surgical resection is indicated in carefully selected patients, and this achieves 5-year survival rates of 20% to 40%.1–12 Resectability is dependent on the number of lesions present, their location, and the presence or absence of extrapulmonary metastases. Most patients are considered nonsurgical candidates and therefore are treated with systemic chemotherapy. The newer chemotherapeutic regimens for colorectal metastases have improved response rates, but the long-term survival remains limited.13–18 Since the early 1990s, an increasing number of minimally invasive techniques have been introduced into clinical practice in the treatment of lung tumors. Image-guided interventional and video-assisted thoracoscopic approaches have become attractive alternatives to open thoracic surgical resection. Other minimally invasive therapies using thermal energy sources to destroy tumors include radiofrequency ablation (RFA), cryoablation, focused ultrasonography, laser, and microwave. Percutaneous RFA is a relatively new treatment option for nonsurgical candidates with colorectal pulmonary metastases. RFA delivers a high frequency of 400 to 500 kHz through a needle electrode into the targeted tissue, causing ionic agitation, tissue heating, and cell death. Currently, four generator and electrode systems from four different manufacturers are available: the Boston Scientific RF-3000 generator with LeVeen and Concerto multitined expandable needle electrodes; the RITA system with a 1500X electrosurgical generator and StarBurst SDE, Semi-flex, XL, and XLi multitined expandable electrodes; the Valleylab Cool-tip RF Ablation System with internally cooled single and clustered needle electrodes; and the Berchtold Elektrotom 106 HiTT with open-perfused electrodes. Preliminary reports have shown that percutaneous RFA can be performed safely and that it may have a promising role in local disease control.19–29 Some studies have found that larger lesions are associated with higher local recurrence rates.25–29 However, to date, treatment failure rates after this interventional procedure have not been adequately addressed, and the criteria for patient selection are not well defined. The primary end point of this prospective study was progression-free survival (PFS), determined from the time of RFA intervention. Prognostic indicators for local and overall PFS were also statistically analyzed.

The inclusion criteria of this study consisted of (1) age between 18 and 85 years; (2) patients who were considered nonsurgical candidates as a result of previous metastases to the liver or lung, more than three lesions in either lung, multiple lobar metastases, bilateral disease involvement, or poor performance status; (3) complete resection of primary colorectal cancer and any hepatic metastases before entry onto the study; (4) patients refusing to accept surgery; and (5) signed informed consent. The exclusion criteria included (1) more than six lesions per hemithorax; (2) diameter of metastases >5 cm; (3) lesions immediately adjacent to major pulmonary vessels; (4) lesions immediately adjacent to major bronchi; (5) bleeding diatheses not responding to medical therapy (pro-thrombin time international normalized ratio >1.5 and platelets <100 x 109); (6) presence of extrapulmonary metastasis; (7) emphysematous bullae, previous surgery, or previous radiotherapy to the affected lung; and (8) significantly compromised lung function. The last criterion did not affect this reported patient group, because they were patients treated exclusively for colorectal lung metastases and did not have any significant compromise in lung function. However, any patients with a history of impaired respiratory function were assessed by a hospital respiratory physician to determine the suitability for RFA. The consensus on the treatment plans for patients with colorectal pulmonary metastases was obtained from a group of surgical oncologists, medical oncologists, and radiologists at weekly meetings.
Preprocedural Investigations
All patients had physical examinations; abdominal, pelvic, and chest computed tomography (CT; with contrast); and bone scans. Measurements of carcinoembryonic antigen (CEA) levels were also obtained. Positron emission tomography was not available.
Percutaneous RFA Procedure
All procedures were conducted by two experienced interventional radiologists. Two rectangular dispersive electrode pads were placed on the patient’s shaved thighs with the large edge facing the RFA site. For lesions located in the anterior part of the lung field, an anterior approach was used with the patient in the supine position; for lesions located in the posterior part of the lung field, a posterior approach was used with the patient in the prone position. Percutaneous RFA procedures were performed with patients under local anesthesia (lidocaine 1%) and conscious sedation (meperidine/midazolam), with CT-guided imaging (Xpress SX; Toshiba, Japan) using the RITA 1500 generator (RITA Medical, Mountain View), with real-time recording and display of temperature, power, and impedance. A RITA Starburst XL probe, either 10 or 15 cm, with a 14-gauge diameter and nine deployable tines, was used. The probe was available in three lengths (10, 15, and 25 cm) and was able to create a maximal lesion of 5 cm in diameter. Because of the space limitations caused by the CT gantry, only 10- and 15-cm probes were used.
The probe was inserted percutaneously into the lung and positioned so that the deployable tines surrounded the target lesion. The ablation algorithm consisted of a staged deployment in which the initial power setting was 35 W and then gradually increased to 150 W. Power was increased with incremental probe deployment to enhance the rate at which the temperature increased. The target temperature was 90°C, and when this temperature was reached, it was maintained for 15, 20, or 37 minutes to achieve a complete tumor ablation of 3, 4, or 5 cm, respectively. All patients with bilateral or multiple lesions were treated in a single session. For lesions >3 cm in diameter, overlapping ablations were performed to ensure a complete ablation. To minimize the incidence of developing pneumothorax in the cases in which additional cycles of ablation were required, the position of the electrode within the tumor was changed by withdrawing it into superficial lung tissue along its major axis, changing its angle, and then reinserting the electrode into the target without a complete withdrawal of the needle out of the pleura. For the purpose of this study, lesions that were 3 cm from the pulmonary hilum were considered as hilar lesions, and lesions located >3 cm away from the pulmonary hilum were considered to be peripheral lesions. Track ablation was routinely performed with cauterizing the access tract on the way out at completion of each lesion ablation. For a complete RFA procedure, all radiologically identified lesions were ablated according to the treatment protocol.

RFA is a controlled ablative technique that implements high-frequency alternating current to generate localized electromagnetic fields and heats targeted tissues to desiccation or thermal coagulation. Lung tumors may well be suited to RFA because of the so-called oven effect, whereby the high-resistance air-filled lung tissue surrounding a low-resistance intraparenchymal tumor affords an insulating effect and traps heat within the targeted tumor.30 RFA has mostly been performed as an outpatient procedure, usually under conscious sedation. Some preliminary reports have shown that percutaneous lung RFA is associated with relatively low morbidity and mortality rates and may have a role in local disease control for nonsurgical candidates with lung tumors.19–29
At our institution, 55 patients underwent percutaneous RFA for colorectal pulmonary metastases. There was no procedure-related mortality. The overall morbidity was 42%. Pneumothorax is the most commonly reported postprocedural complication.21 In this study, 16 patients were found to have pneumothoraxes on the follow-up chest radiographs. Nine symptomatic patients required chest drain insertion for resolution.
Early in the study period, we experienced five cases of intrapulmonary bleeding during the RFA procedure. In our experience, these adverse events did not cause these patients any symptoms, and the patients did not require any medical treatment for resolution. However, Dupuy et al.31 reported 1 death among 27 patients as a result of pulmonary hemorrhage after RFA, attributed to platelet dysfunction. Vaughn et al.32 also recently reported a severe hemorrhagic adverse event after lung RFA. In our series, intrapulmonary bleeding was largely due to probe placement, particularly when placed close to the hilar vessels or other major pulmonary vessels. With increased experience, we found that precise placement of the needle probe was a critical step to avoid intrapulmonary bleeding. In addition, all patients in our study group underwent an extensive preoperative workup, so patients with lesion(s) immediately adjacent to major pulmonary vessels or patients with bleeding diatheses were not considered for RFA. This careful patient selection process may, at least in part, explain why all five cases of intrapulmonary bleeding were self-limiting.
At the last time of contact, 21 patients (38%) were found to have disease progression at an original lung RFA site. In both univariate and multivariate analyses, a largest size of lung metastasis >3 cm was associated with a reduced PFS. In this study, all radiologically identified lesions were ablated according to the treatment protocol. For lesions >3 cm, two or three overlapping ablations were performed to try to achieve larger ablation coverage. However, during an ablation cycle, there is an immediate zone of pneumonitis surrounding the ablated area, and this may obscure the targeted lesion. Therefore, when repositioning the needle electrode for an overlapping ablation, it is sometimes difficult to determine the needle position. Akeboshi et al.29 achieved a lower rate of tumor necrosis in those targeted lesions >3 cm. Steinke et al.27 showed that it was difficult to achieve a complete ablation in large lung lesions. Lee et al.23 showed that lower rates of local disease control correlated with decreased mean survival rates: 19.7 vs. 8.7 months. However, from our current data, we are unable to analyze survival to determine whether any patients died of local disease alone, which may be influenced by local treatment options, such as RFA.
Persistent increases of post-RFA CEA levels may indicate an incomplete ablation of the lung lesion or undetected metastasis elsewhere. At our institution, we routinely used CT (contrast), compared with all previous scans, to follow up all patients. One of the major limitations of percutaneous lung RFA is the difficulty in monitoring the progression of disease. After RFA, the area of consolidation may appear larger than the original lung lesion, and the resolution of the consolidation may take months. During this period, disease progression may be difficult to assess. Positron emission tomography scans may continue to have positive results, thus making it difficult to assess whether viable tumor or scarring is present. Because the inflammatory changes after RFA often subside by 3 months, the 3-month scan is often more useful as the baseline measurement against which local progression can be judged.
Limiting resection to patients with single-organ metastases may be denying some patients a chance for long-term survival. There are strong, albeit retrospective, data that suggest that synchronous or metachronous liver and lung metastases may be treated surgically with as good an outcome as those achieved for liver alone or lung alone.33 Surgical resection is the gold standard treatment for colorectal pulmonary metastases, and we do not regard lung RFA as an alternative to surgery, but it may be indicated in non-surgical candidates. In this study, a number of selection factors influenced this patient population. In some cases, more than one factor influenced the decision to perform RFA rather than surgery. The results of percutaneous lung RFA versus systemic chemotherapy should be interpreted with the knowledge that these treatment strategies have not been compared directly. This study suggested that this interventional procedure might have a useful role in nonsurgical candidates with colorectal pulmonary metastases. However, its efficacy is limited for lesions that are >3 cm.

结直肠癌肺转移：肺段 vs 楔形切除

研究背景
虽然楔形切除是结直肠癌肺转移最常见的术式，但是关于肺段切除治疗结肠转移瘤的报道很少。鉴于此，来自日本的教授等开展了一项研究，研究结果发表在近期的European Journal Cardio-Thoracic Surgery杂志上。

研究方法
该研究为日本结直肠癌肺转移瘤切除的全国性回顾性研究的亚组分析。
研究者纳入2004年1月－2008年12月，接受肺段切除（n=98）或楔形切除（n=455），术前未接受化疗的结直肠癌转移瘤，共553例患者。
研究者评估这两种手术方式对复发模式，无复发生存期和总生存期的影响。


研究结果
1：两组患者的一般资料中只有转移瘤的中位直径存在统计学差异，肺段切除组（中位数为18mm，5-50mm）和楔形切除（中位数为14mm，5-51mm）。
2：肺段切除组相比楔形切除组患者漏气时间延长率较高（5.1％ vs 1.8％，P=0.048）。
切缘复发率楔形切除组高于肺段切除组（7.3％ vs 2.0％，P=0.035）。
3：5年无复发生存率肺段切除组为48.8％，楔形切除组为36.0％（图2）。
4：5年总体生存率肺段切除组为80.1％，楔形切除组为68.5％（图1）。
5：多变量分析表明，肺段切除术后复发的保护因素（风险比：0.63，95％置信区间：0.44-0.87，P=0.005），但与总生存率无显著相关性（风险比：0.65，95％置信区间：0.38-1.05，P = 0.080）。









研究结论
肺段切除治疗结直肠癌局限性肺转移可降低切缘的局部复发率。

结直肠癌复发性肺转移反复肺切除术对长期生存的影响

日本国家癌症中心附属医院东
区胸外科Hishida 等报告，对可切
除且复发局限于肺部复发的结直肠
癌肺转移（PM-CRC）术后患者，
反复肺切除术可带来理想的生存
率，尤其是对进行肺转移灶切除术
前没有肝转移的结肠癌患者。（Ann
Thorac Surg. 2016 年10 月25 日在
线版）
为了阐释PM-CRC 患者反
复肺切除术（RLR） 后的长期
生存情况，该项回顾性研究入组
2004~2008 年日本46 家临床机构中
经过R0 切除的898 例PM-CRC
患者， 分析了216 例PM-CRC
切除术后复发仅限于肺部复发的
患者。研究分析了反复肺切除术
（RLR）后的总生存（OS）率，并
利用多变量Cox 分析阐释了预后因
素。
结果显示：在216 例患者中，
132 例（61%）患者经受了反复肺
切除术（RLR）， 其5 年OS 率
为75.3%。22 例患者经受了两次
RLR，2 例患者经受了三次RLR。
分析发现，即使经受了两次RLR，
5 年OS 率仍较理想（55.1%）。同
步的肝转移（进行肺转移灶切除术
前已完全切除或消融）提示RLR
术后预后不佳（HR=4.84，95%CI
1.48~14.8）。原发直肠肿瘤也提
示预后不佳（HR=3.16，95%CI
1.17~9.35）。无这两项预后不良因
素的58 例患者RLR 术后5 年OS
率为82.6%。

结直肠癌肺转移瘤切除术是否可使患者获益（前瞻性研究结果）

转移瘤不可切除的无症状mCRC，原发瘤切除的意义？