Clinical outcomes of percutaneous cryoablation for pulmonary metastases of soft tissue and bone sarcomas under local anesthesia: a single-institution retrospective cohort study

Abstract

Background

The management of pulmonary metastases of soft tissue and bone sarcomas is challenging due to their resistance to chemotherapy and radiotherapy. This study aimed to evaluate local tumor progression, overall survival (OS), and procedural safety after percutaneous cryoablation for pulmonary metastases of soft tissue and bone sarcomas.

Methods

This retrospective cohort study included 28 patients who underwent percutaneous cryoablation for 84 metastatic lung tumors of soft tissue and bone sarcomas during 53 sessions between July 2002 and December 2016. Percutaneous cryoablation was performed under local anesthesia using the CRYOcare system and computed tomography fluoroscopy. Procedural complications within 30 days were assessed using prospectively recorded clinical data and retrospectively graded according to the Common Terminology Criteria for Adverse Events version 5.0. The time to local progression of the treated tumors and overall survival were estimated using the Kaplan–Meier method.

Results

All cryoablation procedures were performed under local anesthesia. The median tumor diameter was 1.1 cm (range: 0.4–6.5 cm). Twelve tumors (14.3%) showed local tumor progression after initial cryoablation during the median follow-up duration of 16 months (range: 0–216 months). The primary efficacy rates were 88.8%, 79.1%, and 79.1% at 1, 3, and 5 years, respectively. After repeat cryoablation for four tumors with local progression, the assisted efficacy rates were 95.6%, 82.8%, and 77.9% at 1, 3, and 5 years, respectively. Three tumors in two patients had no local progression for more than 10 years. A tumor diameter of ≤ 2 cm was significantly associated with improved primary and assisted efficacy rates (P = 0.024 and P = 0.015, respectively). The median OS was 34 months (range, 2–155 months), and the OS rates were 75.0%, 52.8%, and 28.8% at 1, 3, and 5 years, respectively. Adverse events occurred in 31 sessions (58.5%); all were grade 1 or 2, and no grade ≥ 3 adverse events or 30-day mortality were observed.

Conclusions

In this selected single-institution cohort of nonsurgical candidates with pulmonary metastases of soft tissue and bone sarcomas, percutaneous cryoablation was feasible and was associated with local tumor control and an acceptable safety profile.

Background

The prevalence of soft tissue and bone sarcomas is < 1% of all malignant tumors. The most common metastatic form is hematogenous spread to the lungs, which occurs in isolation in 20–40% of patients [1–3]. Pulmonary metastases represent the leading cause of mortality of soft tissue and bone sarcomas, and appropriate management of pulmonary metastases may contribute to prolonged survival in selected patients [4–6]. Chemotherapy is the standard treatment of choice for multiple metastases of soft tissue and bone sarcomas [7], but surgical resection of pulmonary metastases is considered feasible when complete resection of all metastatic lesions is achievable [1, 7–12]. However, surgery is contraindicated in some patients because of previous thoracotomy, respiratory dysfunction, comorbidities, age, or other reasons.

Stereotactic body radiotherapy (SBRT) [13–15] and thermal ablation treatments [16–23], including radiofrequency ablation (RFA) [16–21] and microwave ablation (MWA) [22, 23], are considered favorable for such patients because they are minimally invasive alternatives with acceptable local tumor control rates. mong the various thermal ablation modalities, computed tomography (CT)-guided percutaneous cryoablation has emerged as a minimally invasive treatment for patients with inoperable pulmonary tumors [24–26].

CT-guided percutaneous cryoablation has been studied as a minimally invasive treatment option for selected patients with primary or metastatic pulmonary tumors [27–37]. Although previous studies have evaluated percutaneous thermal ablation, including cryoablation, for pulmonary metastases of soft tissue and bone sarcomas [16–23, 35–37], disease-specific evidence regarding percutaneous cryoablation for pulmonary metastases of soft tissue and bone sarcomas remains limited. Because pulmonary metastases of soft tissue and bone sarcomas are rare, additional disease-specific data regarding outcomes after local treatment remain scarce yet clinically relevant. Moreover, image-guided thermal ablation, including cryoablation, continues to be used as a minimally invasive local treatment option for selected patients with primary or metastatic lung tumors, particularly those who are not suitable candidates for surgery. Therefore, this study aimed to evaluate clinical outcomes, including local tumor progression, overall survival (OS), and procedural safety, in patients undergoing percutaneous cryoablation for pulmonary metastases of soft tissue and bone sarcomas.

Methods

Ethics

This single-institution, retrospective cohort study was conducted in accordance with the Declaration of Helsinki (revised in 2013) and approved in 2021 by the Keio University School of Medicine Ethics Committee (approval ID: 20210003). The requirement for informed consent for participation in this retrospective study was waived.

Patients

This single-institution, retrospective cohort study was conducted at Keio University Hospital, Tokyo, Japan. We retrospectively reviewed 227 consecutive patients who underwent percutaneous cryoablation for malignant pulmonary tumors between July 2002 and December 2016. Among them, patients who underwent percutaneous cryoablation for pulmonary metastases originating from soft tissue and bone sarcomas were eligible for this study. Patients with malignant pulmonary tumors other than pulmonary metastases of soft tissue and bone sarcomas, including primary lung cancer and pulmonary metastases of other malignancies, were excluded. The final study cohort comprised 28 patients who underwent percutaneous cryoablation for 84 pulmonary metastases originating from soft tissue and bone sarcomas over 53 sessions. Figure 1 shows the flow of patient selection.

Fig. 1.

Flow diagram of patient selection

Use of the CRYOcare cryosurgical unit was suspended after December 2016 because domestic maintenance support became unavailable. No alternative cryoablation system was used for malignant thoracic tumors at our institution between December 2016 and April 2023. Cryoablation was restarted in April 2023 after introduction of the Visual-ICE system.

All patients underwent a comprehensive evaluation by interventional radiologists and thoracic surgeons before hospitalization, which included assessing inoperability and determining the technical feasibility of percutaneous cryoablation. The inclusion criteria for percutaneous cryoablation were as follows: (i) surgical ineligibility due to multiple prior surgeries, multiple tumors, patient refusal, poor pulmonary function, comorbidities, or advanced age; (ii) absence of uncontrolled extrathoracic metastases of soft tissue and bone sarcomas that would preclude local treatment of pulmonary metastases; (iii) performance status of 0 or 1 on the Eastern Cooperative Oncology Group (ECOG) scale; (iv) platelet count of > 50,000 per µL; (v) prothrombin time/international normalized ratio of < 1.5; (vi) estimated life expectancy longer than 12 months; and (vii) written informed clinical consent for the cryoablation procedure.

Histological subtype was determined from the pathological diagnosis of the primary soft tissue or bone sarcomas, or from previously biopsied or resected metastatic lesions when available. Biopsy of every treated pulmonary lesion before cryoablation was not routinely performed. In patients with a known history of soft tissue or bone sarcomas and no history of another malignancy, pulmonary lesions were considered metastatic when the clinical course and imaging findings were consistent with sarcoma lung metastases. In selected cases, biopsy was performed before or during the cryoablation procedure to confirm the histology.

Cryoablation technique

A detailed description of the cryoablation procedure has been previously reported [28, 31, 33, 34]. All cryoablation procedures were performed using a multidetector CT scanner (Aquilion 64; Toshiba Med. Co. Ltd., Tokyo, Japan) under local anesthesia by interventional radiologists (S.N., M.I., and H.Y.). The chief interventional radiologist (S.N.) had more than 30 years of ablation experience and performed all 366 cryoablation sessions at our institution. A 21-gauge guiding needle was inserted into the targeted tumor under CT fluoroscopic guidance. Subsequently, a custom-made 8- or 11-gauge coaxial stainless-steel introducer system, consisting of an inner guiding sheath and an outer sheath (Silux, Kawaguchi, Japan), was used to establish the access tract. After removal of the 21-gauge guiding needle and inner guiding sheath, a 2- or 3-mm cryoprobe (CRYOcare Cryosurgical Unit, Endocare, Irvine, CA, USA) was inserted through the outer sheath. Cryoablation was performed using a protocol of three freeze–thaw cycles. Until July 2006, the freezing durations were 5, 5, and 10 min for the first, second, and third cycles, respectively. After July 2006, the second freeze cycle was extended, and the freezing durations were adjusted to 5, 10, and 10 min. The first freeze cycle was kept relatively short to facilitate tissue edema formation before subsequent freezing, whereas the later freeze cycles were prolonged to obtain a stable and sufficient ablation zone. Active thawing was performed using a high-pressure helium gas system after each freeze procedure until the cryoprobe temperature reached 20 °C. CT was performed after each cycle to visualize and evaluate the ablation zone. The ablation margin was assessed on sequential CT slices three-dimensionally to confirm tumor coverage as comprehensively as possible. The target ablation margin was at least 5 mm beyond the tumor border; however, for lesions adjacent to critical structures, such as large vessels, bronchi, or pleura, procedural safety was prioritized. After removal of the cryoprobe, fibrin glue (Bolheal; Chemo-Sero-Therapeutic Research Institute, Kumamoto, Japan) was injected into the outer sheath to prevent hemothorax and pneumothorax.

Follow-up and outcome measures

Follow-up CT scans were performed 1 day, 1 week, and 1 month after the procedure and subsequently at 3–4-month intervals. Follow-up CT images were retrospectively reviewed by three authors (Y.A., K.K., and K.A.), and final determinations of local tumor progression were made by consensus based on serial imaging findings and the predefined criteria. Local tumor progression was defined as an increase of more than 20% in the longest transaxial diameter of the treated lesion or ablation zone compared with the smallest post-ablation measurement, or new or enlarging nodular growth within or at the margin of the ablation zone [24, 25].

Clinical variables were collected from the medical charts by three authors (Y.A., K.K., and K.A.) and categorized into patient and tumor factors as follows: (1) patient factors: age, sex, Eastern Cooperative Oncology Group performance status, reasons for cryoablation, total number of pulmonary lesions, smoking history, past treatment history, and histological subtype; (2) tumor factors: histological classification according to the 2020 World Health Organization (WHO) classification of soft tissue and bone tumors, maximum tumor diameter, tumor distribution (central/non-central), and presence of extrathoracic metastases prior to cryoablation. Central location was defined as within the inner one-third of the lung field.

Complications occurring within 30 days after cryoablation were assessed on a per-session basis and graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. During routine clinical care, adverse events were prospectively recorded in the medical charts by senior thoracic surgeons and interventional radiologists. For the present retrospective analysis, these prospectively recorded adverse event data were retrospectively graded according to CTCAE version 5.0 and summarized by Y.A., a thoracic surgeon.

Statistical analysis

No formal a priori sample size calculation was performed because this was a retrospective cohort study involving a rare clinical condition: pulmonary metastases of soft tissue and bone sarcomas treated with percutaneous cryoablation under local anesthesia. All consecutive eligible patients treated during the study period were included. Accordingly, the analyses were exploratory and should be interpreted with caution. The sample size was determined by the number of eligible patients treated during the study period.

In accordance with the consensus recommendations for image-guided tumor ablation endpoints by Puijk et al. [38], primary efficacy was defined as the absence of radiographic local tumor progression after the initial cryoablation procedure. Tumors were considered to have lost primary efficacy at the time of first documented local tumor progression after the initial procedure. Some patients underwent repeat cryoablation for local progression after the initial cryoablation procedure. Assisted efficacy was defined as the absence of radiographic local tumor progression after the final cryoablation procedure, allowing for repeat cryoablation for tumors that developed local progression after the initial procedure. For tumors without documented local progression, follow-up was censored at the date of the last CT evaluation. Death without documented local tumor progression was not treated as a competing event in the present tumor-level analysis. Given the small cohort size and the limited number of local progression and competing events, Kaplan–Meier analysis was retained as a descriptive and exploratory approach to estimate primary and assisted efficacy.

OS was defined as the duration from the initial cryoablation procedure to the date of death or last follow-up and was estimated using the Kaplan–Meier method. Variables potentially predictive of local tumor progression were analyzed using the log-rank test. A tumor diameter cutoff of 2 cm was used for exploratory subgroup analysis based on a previous report [33]. P-values less than 0.05 were considered to be statistically significant. All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) software (version 29; IBM Corporation, Armonk, NY, USA).

Results

Patient and tumor characteristics

The patient and tumor characteristics are summarized in Table 1. Twenty-eight patients with 84 metastatic lung tumors originating from soft tissue and bone sarcomas underwent 53 cryoablation sessions. The median age of the patients was 41.5 years (range, 16–83 years), 18 (64.3%) were male, and 11 (39.3%) had a history of smoking. Previous treatments for pulmonary metastases before cryoablation included chemotherapy for 17 (60.7%), surgery for 15 (53.6%), and radiotherapy for 4 (14.3%) patients. All targeted pulmonary metastases met the inclusion criteria. Six patients (21.4%) had extrathoracic metastases involving the liver, brain, bone, kidney, pancreas, or non-regional lymph nodes, which had been treated or were being clinically managed with chemotherapy, surgery, or radiotherapy. A median of 2 tumors was treated with cryoablation per patient (range, 1–17) and a median of 1 cryoablation session was performed per patient (range, 1–9). The histological subtypes of the tumor were as follows: alveolar soft part sarcoma (26 tumors in 4 patients), undifferentiated pleomorphic sarcoma (16 tumors in 4 patients), synovial sarcoma (9 tumors in 3 patients), leiomyosarcoma (8 tumors in 4 patients), malignant peripheral nerve sheath tumor (8 tumors in 2 patients), osteosarcoma (7 tumors in 4 patients), spindle cell sarcoma (3 tumors in 2 patients), liposarcoma (2 tumors in 2 patients), chondrosarcoma (2 tumors in 1 patient), epithelioid sarcoma (2 tumors in 1 patient), and Ewing sarcoma (1 tumor in 1 patient). Among these subtypes, alveolar soft part sarcoma, synovial sarcoma, undifferentiated pleomorphic sarcoma, malignant peripheral nerve sheath tumor, and epithelioid sarcoma are categorized as tumors of uncertain differentiation (61 tumors in total) that lack an identifiable line of differentiation, according to the 2020 WHO classification of soft tissue and bone tumors. The median tumor diameter was 1.1 cm (range, 0.4–6.5 cm) for 84 tumors, and 17 tumors (20.2%) were centrally located.

Table 1.

Demographic and tumor characteristics of the study cohort (n = 28, tumors = 84)

Characteristics	Value
Age (median, range)	41.5, 16–83
Sex
Male  Female	18 (64.3%) 10 (35.7%)
ECOG performance status
0  1	27 (96.4%) 1 (3.6%)
Reasons for cryoablation (not mutually exclusive)
Refusal to undergo surgery  Multiple previous thoracotomies  Multifocal disease  Severe comorbidity  Technically unresectable  Advanced age	15 (53.6%) 9 (32.1%) 6 (21.4%) 3 (10.7%) 2 (7.1%) 1 (3.6%)
Number of tumors targeted per patient
Median, range  1  2  3  ≥4	2, 1–17 10 (35.7%) 10 (35.7%) 2 (7.1%) 6 (21.4%)
Total number of cryoablation sessions per patient
Median, range  1  2  3  5  9	1, 1–9 16 (57.1%) 7 (25.0%) 3 (10.7%) 1 (3.6%) 1 (3.6%)
Smoking status
Nonsmoker  Current or former smoker	17 (60.7%) 11 (39.3%)
Treatment history for pulmonary metastases before cryoablation
Chemotherapy  Surgery  Radiotherapy	17 (60.7%) 15 (53.6%) 4 (14.3%)
Sites of extrathoracic metastases before cryoablation
Liver  Brain  Bone  Kidney  Pancreas  Inguinal lymph node	1 (3.6%) 1 (3.6%) 1 (3.6%) 1 (3.6%) 1 (3.6%) 1 (3.6%)
Histological type of sarcoma (number of tumors / patients)
Alveolar soft part sarcoma  Undifferentiated pleomorphic sarcoma  Synovial sarcoma  Leiomyosarcoma  Malignant peripheral nerve sheath tumor  Osteosarcoma  Spindle cell sarcoma  Liposarcoma  Chondrosarcoma  Epithelioid sarcoma  Ewing sarcoma	26 (31.0%) / 4 (14.3%) 16 (19.0%) / 4 (14.3%) 9 (10.7%) / 3 (10.7%) 8 (9.5%) / 4 (14.3%) 8 (9.5%) / 2 (7.1%) 7 (8.3%) / 4 (14.3%) 3 (3.6%) / 2 (7.1%) 2 (2.4%) / 2 (7.1%) 2 (2.4%) / 1 (3.6%) 2 (2.4%) / 1 (3.6%) 1 (1.2%) / 1 (3.6%)
Primary site
Extremities  Abdomen  Thorax	21 (75.0%) 5 (17.9%) 2 (7.1%)
Tumor diameter, cm
Median, range  ≤1  1<, ≤2  2<, ≤3  3<	1.1, 0.4–6.5 39 (46.4%) 35 (41.7%) 4 (4.8%) 6 (7.1%)
Tumor distribution
Central  Non-central	17 (20.2%) 67 (79.8%)

Values are shown as n (%) unless otherwise indicated. Patient-level variables are presented per patient, whereas tumor diameter and tumor distribution are presented per tumor. Histological type is shown as the number of tumors / number of patients. Reasons for cryoablation and treatment history before cryoablation were not mutually exclusive. Severe comorbidity was defined as a comorbid medical condition that made surgical resection unsuitable or high-risk after multidisciplinary evaluation

ECOG Eastern Cooperative Oncology Group

Procedural safety

The post-treatment outcomes are summarized in Table 2. The median post-cryoablation hospitalization duration was 2 days (range, 2–8 days), and discharge occurred on day 2 in 41 sessions (77.4%). Of the 53 sessions, adverse events were recorded in 31 sessions (58.5%), whereas 22 sessions (41.5%) had no adverse events. A total of 36 adverse events were observed, including 26 grade 1 events and 10 grade 2 events. No adverse events of grade ≥ 3 were observed. Because some sessions involved more than one adverse event, the numbers of grade 1 and grade 2 events were not mutually exclusive at the session level. Table 3 shows the adverse events associated with cryoablation. Pneumothorax was the most common complication, occurring in 21 sessions (39.6%). Among these, 11 events were grade 1 and 10 events were grade 2. Manual aspiration and chest tube drainage were required in 5 sessions (9.4%) each. Hemoptysis was observed in 14 sessions (26.4%), and hemothorax was observed in 1 session (1.9%). No patient required additional treatment for hemoptysis or hemothorax. No 30-day mortality was recorded.

Table 2.

Post-treatment outcomes following cryoablation (n = 53)

Characteristics	Value
Duration of post-cryoablation hospitalization, days (median, range)	2, 2–8
Sessions with adverse events
Present  Absent	31 (58.5%) 22 (41.5%)
Individual adverse events by CTCAE grade
Grade 1  Grade 2  Grade 3 or higher	26 (49.1%) 10 (18.9%) 0 (0.0%)
30-day mortality
Present  Absent	0 (0.0%) 53 (100.0%)

Values are shown as n (%) unless otherwise indicated. Percentages were calculated using the total number of cryoablation sessions as the denominator (n = 53). Adverse event “Present” and “Absent” are shown on a per-session basis. CTCAE grades are shown as the number of individual adverse events and are not mutually exclusive because some sessions involved more than one adverse event

CTCAE Common Terminology Criteria for Adverse Events

Table 3.

Adverse events associated with cryoablation procedures (n = 53)

Characteristics	Value
Pneumothorax  Chest tube insertion  Manual aspiration	21 (39.6%) 5 (9.4%) 5 (9.4%)
Hemoptysis	14 (26.4%)
Hemothorax	1 (1.9%)

Values are shown as n (%). Percentages were calculated per cryoablation session (n = 53). Chest tube insertion and manual aspiration are shown as interventions for pneumothorax and are included within the total number of pneumothorax events. Some sessions included more than one adverse event

Local tumor control and overall survival

Of the 84 tumors, local tumor progression after initial cryoablation was observed in 12 tumors (14.3%) in 11 patients, with 1 patient experiencing local tumor progression of 2 distinct lesions. The primary efficacy rates were 88.8% at 1 year, 79.1% at 3 years, and 79.1% at 5 years, with a median follow-up duration of 16.3 months (range, 0–216.4 months) (Fig. 2A). Of the 12 tumors with local progression, management included additional cryoablation for 4 tumors (3 patients), surgery for 3 tumors, and surgery following radiation therapy for 1 tumor, while 4 tumors were managed with observation without any additional treatment. Among the 4 tumors that underwent additional cryoablation, 2 underwent 2 cryoablation sessions, including the initial procedure, 1 underwent 3 sessions, and 1 underwent 6 sessions. Of the 2 tumors treated with 2 cryoablation sessions, 1 ultimately showed no further local progression (CT follow-up, 107.5 months), while the other exhibited local progression after the second procedure (CT follow-up, 36.6 months). The clinical course of the former lesion is shown in Fig. 3. The tumor treated with 3 cryoablation sessions eventually showed no local progression after the final procedure (CT follow-up, 36.6 months), whereas the tumor treated with 6 sessions exhibited local progression after the final procedure (CT follow-up, 33.4 months). The 2 lesions that ultimately demonstrated local progression after multiple cryoablation procedures were managed with observation without any additional treatment. The assisted efficacy rates were 95.6% at 1 year, 82.8% at 3 years, and 77.9% at 5 years, with a median CT follow-up duration of 22.8 months (range, 0–216.4 months) (Fig. 2B). The 10-year estimate was 70.8%, but the number of tumors at risk was limited at this late time point. Two patients (7.1%) with 3 tumors survived for more than ten years without any local progression. Tumor diameters ≤ 2 cm were associated with significantly better primary and assisted efficacy rates (P = 0.024 and P = 0.015, respectively) (Figs. 4A and B). However, no significant differences were observed in either the primary or assisted efficacy rates based on tumor distribution (Figs. 5A and B) or the presence of extrathoracic metastases before cryoablation (Figs. 6A and B). Additionally, there were no significant differences in primary or assisted efficacy rates between the tumors of uncertain differentiation and all other tumors (Figs. 7A and B).

Fig. 2.

Efficacy rates for all tumors. A Primary efficacy rates. B Assisted efficacy rates

Fig. 3.

Serial computed tomography images of repeat percutaneous cryoablation for a pulmonary metastasis in the right middle lobe. A Lung-window computed tomography image obtained before the initial cryoablation showing a pulmonary metastasis in segment 4 of the right middle lobe. B, C Lung- and mediastinal-window images obtained immediately after the initial cryoablation showing that the ablation zone encompassed the target tumor with an adequate margin. D, E Lung-window images obtained 1 and 6 months after the initial cryoablation showing gradual regression of the ablation zone. F Lung-window image obtained 1 year after the initial cryoablation showing new nodular growth at the margin of the ablation zone, consistent with local tumor progression. (G) Lung-window image obtained 2 years after the initial cryoablation showing further enlargement of the locally progressed lesion, for which a second cryoablation procedure was subsequently performed. H, I Lung- and mediastinal-window images obtained immediately after the second cryoablation showing adequate coverage of the lesion by the ablation zone. J, K Lung-window images obtained 1 and 3 years after the second cryoablation showing progressive regression of the ablation zone, leaving only a residual scar-like opacity. L Lung-window image obtained 7 years after the second cryoablation showing maintained local tumor control at the treated site, although multiple pulmonary metastases had developed

Fig. 4.

Efficacy rates for all tumors according to tumor diameter. A Primary efficacy rates. B Assisted efficacy rates

Fig. 5.

Efficacy rates for all tumors according to tumor distribution. A Primary efficacy rates. B Assisted efficacy rates

Fig. 6.

Efficacy rates for all tumors based on extrathoracic metastases before cryoablation. A Primary efficacy rates. B Assisted efficacy rates

Fig. 7.

Efficacy rates for all tumors according to histological differentiation. A Primary efficacy rates. B Assisted efficacy rates

The median OS of the 28 patients was 34 months (mean, 48 months; range, 2–155); the OS rates at 1, 3, and 5 years were 75.0%, 52.8%, and 28.8%, respectively (Fig. 8). The 10-year OS estimate was 19.7%, but the number of patients at risk was limited at this late time point. Three patients (10.7%) survived for more than ten years after the initial cryoablation.

Fig. 8.

Overall survival rates for all patients

Discussion

Percutaneous cryoablation was performed for pulmonary metastases of soft tissue and bone sarcomas in 28 patients with 84 tumors during 53 cryoablation sessions over 15 consecutive years. Adverse events occurred in 31 sessions (58.5%), with no grade ≥ 3 events. Pneumothorax was the most common complication (21 sessions, 39.6%), and 5 sessions (9.4%) required chest tube drainage. No 30-day mortality was recorded. Although the overall pneumothorax rate in the present cohort was higher than that reported in some contemporary studies, this rate included all radiographically detected pneumothorax, including grade 1 events. Of the 21 pneumothorax events, 11 were grade 1 and 10 were grade 2, with no grade ≥ 3 events. The grade 2 pneumothorax rate was 18.9%, which was comparable to that in our previous institutional study of lung cryoablation, in which CTCAE grade ≥ 2 pneumothorax was reported in 18.0% of sessions [34]. These data may indicate that percutaneous cryoablation was feasible and was associated with an acceptable safety profile in this selected cohort.

In the present study, the primary efficacy rates were 88.8%, 79.1%, and 79.1%; the assisted efficacy rates were 95.6%, 82.8%, and 77.9%; and the OS rates were 75.0%, 52.8%, and 28.8% at 1, 3, and 5 years after cryoablation, respectively. Three tumors (3.6%) in 2 patients showed no local progression for more than ten years after the initial cryoablation. Although previous studies have evaluated percutaneous ablation for sarcoma lung metastases in cohorts with comparable numbers of treated lesions [36, 37], data specifically addressing cryoablation for pulmonary metastases of soft tissue and bone sarcomas remain limited. Therefore, the present study should be interpreted as additional disease-specific evidence from a selected cohort of nonsurgical candidates, rather than as a definitive assessment of the efficacy of cryoablation.

Although the patients in this cohort were treated between 2002 and 2016, the present findings may still be clinically relevant. Image-guided thermal ablation, including percutaneous cryoablation, remains a minimally invasive local treatment option for selected patients with primary or metastatic lung tumors, particularly those who are not suitable candidates for surgery. In addition, the underlying principle of cryoablation—local tumor destruction by repeated freeze–thaw cycles under CT guidance—has remained unchanged. In the present cohort, the basic procedural protocol for percutaneous lung cryoablation has also remained largely consistent, although ablation systems, imaging technology, and peri-procedural management have evolved over time. Because pulmonary metastases of soft tissue and bone sarcomas are rare, disease-specific data remain valuable, particularly when interpreting outcomes in highly selected nonsurgical candidates.

Previous studies provide important context for interpreting the present findings. Several prospective and retrospective studies of percutaneous cryoablation for lung tumors included small numbers of patients with metastases of soft tissue and bone sarcomas. The multicenter Study of Metastatic Lung Tumors Targeted by Interventional Cryoablation Evaluation (SOLSTICE) [24] and Evaluating Cryoablation of Metastatic Lung Tumors in Patients-Safety and Efficacy (ECLIPSE) study [25] reported that 4.7% (6/128) and 10% (4/40) of cases, respectively, involved pulmonary metastases of soft tissue and bone sarcomas. Both studies were multicenter, single-arm, and prospective. A single-institution retrospective study by McDevitt et al. reported 47 lung tumors that were treated with percutaneous cryoablation, including three pulmonary metastases of sarcomas [26]. Direct comparison of OS may not be appropriate due to the heterogeneity in primary tumor types across the studies, but the efficacy rates per tumor appear broadly comparable. The SOLSTICE study reported primary efficacy rates of 85.1% and 77.2% at 1 and 2 years, respectively, and assisted efficacy rates of 91.1% and 84.4% at 1 and 2 years, respectively [24]. On the other hand, the ECLIPSE study reported primary efficacy rates of 87.9% and 79.2% at 3 and 5 years, respectively [25]. McDevitt et al. reported primary efficacy rates of 41% and 23% for metastatic lung tumors at 1 year and 3 years, respectively [26]. Both primary and assisted efficacy rates in the present study were broadly comparable to those in the SOLSTICE and ECLIPSE studies. The lower efficacy rates in the study by McDevitt may be attributed to diameters of > 2 cm for 56% of the tumors relative to 6% in the SOLSTICE study and 20% in the ECLIPSE study. The current study included 12% of such tumors, a proportion similar to that of the SOLSTICE and ECLIPSE studies. In support of this finding, a tumor diameter of ≤ 2 cm was associated with significantly better primary and assisted efficacy rates in the present study, which was consistent with our previous reports [33].

SBRT is used to treat inoperable lung metastases of soft tissue and bone sarcomas. OS in the present study tended to be lower than that reported in previous studies using SBRT, whereas local tumor control was comparatively favorable. Navarria et al. reported 51 lesions in 28 patients with lung metastases of soft tissue sarcomas, demonstrating 5-year local tumor control rates of 96% and OS rates of 96.2% and 60.5% at 2 and 5 years, respectively [13]. Soyfer et al. reported the outcomes of 53 pulmonary lesions in 22 patients with metastatic sarcoma, with an OS rate of 62% at 5 years and no local tumor progression in any SBRT-treated lesion during a median follow-up of 95 months [14]. Moreover, Frakulli et al. analyzed SBRT for lung metastases from sarcoma in 24 patients with 68 lesions, showing a local tumor control rate of 85.9% and an OS rate of 66.4% at 2 years [15]. Although SBRT has shown favorable local tumor control for pulmonary metastases of soft tissue and bone sarcomas, it is associated with a risk of pulmonary toxicity [13, 15]. Consequently, SBRT has a limited ability to treat multiple lesions with overlapping radiation fields and manage local tumor progression following previous SBRT treatment. Compared to SBRT, cryoablation may have advantages in treating neighboring lesions and in allowing repeat treatment for local tumor progression. Furthermore, SBRT is deemed unsuitable for targeting lesions adjacent to hilar structures because of the risk of significant complications [39, 40]. Therefore, tumors with local progression after initial SBRT, as well as those adjacent to the hilar structures, may be considered potential targets for cryoablation.

Other local image-guided thermal ablation modalities, such as RFA and MWA, have been utilized for pulmonary metastases and have been reported to provide local tumor control and favorable OS in patients with oligometastatic soft tissue and bone sarcomas [16, 41]. Sato et al. reported the outcomes of RFA for 144 pulmonary metastases in 46 patients with sarcoma, demonstrating primary efficacy rates of 83.5% at 1 year and 76.3% at 2 years; assisted efficacy rates of 90.0% at 1 year and 81.4% at 2 years; and OS rates of 80.6%, 70.1%, and 47.1% at 1, 2, and 3 years, respectively [21]. According to other reports of RFA for pulmonary metastases of soft tissue and bone sarcomas, the OS rates were 88.9–100% at 1 year and 59–85% at 3 years [18–20]. As stated in two reports focusing on various pulmonary tumors, not limited solely to soft tissue and bone sarcomas, MWA yielded local tumor progression rates of 2.3–7.7% [22, 23].

Several studies focusing on pulmonary metastases of soft tissue and bone sarcomas have included percutaneous cryoablation. Yevich et al. evaluated 26 pulmonary metastases in 11 children with osteosarcoma, with cryoablation performed on 2 lesions and RFA on 24 lesions [35]. One patient died of rapidly progressive lumbar metastasis before the first follow-up (0.7 months). However, the remaining 10 patients showed no local tumor progression, with a median local tumor control duration of 16.7 months (range, 4.1–41.8 months). Bourgouin et al. compared the efficacies of cryoablation and MWA for the treatment of 65 sarcoma pulmonary metastases in 27 patients [36]. Cryoablation was performed for 35 tumors while MWA was performed for 30 tumors. The estimated local tumor control rates for tumors measuring ≤ 1 cm in diameter were 97% and 95% for MWA and 99% and 98% for cryoablation at 1 and 2 years, respectively. For tumors measuring > 1 cm in diameter, the estimated local tumor control rates were 74% and 62% for MWA and 86% and 79% for cryoablation at 1 and 2 years, respectively. More recently, Pal et al. reported outcomes of percutaneous cryoablation for 250 recurrent or metastatic soft tissue sarcoma tumors in adults, including 85 lung metastases [37]. In that study, the 1- and 2-year local tumor control rates after cryoablation for lung metastases were 96% and 94%, respectively, which were higher than those observed for metastases at other sites. Although these studies differed in patient population, tumor site, ablation modality, and follow-up duration, they provide relevant context for interpreting the present disease-specific cohort of pulmonary metastases of soft tissue and bone sarcomas.

Compared with other thermal ablation modalities, cryoablation may have potential advantages for treating hilar lesions. In contrast, RFA and MWA may be less suitable for lesions adjacent to large vessels or bronchi because of the potential risk of damage to these structures [42]. Cryoablation spares the collagenous architecture and preserves the tracheobronchial tree [43]. Additionally, the circulating blood in large vessels generates heat during cryoablation, contributing to the preservation of the vessel structure [33]. Furthermore, RFA and MWA are associated with worse intraprocedural pain than cryoablation because the latter has an analgesic cooling effect on the nerves [44]. In the present cohort, tumor distribution was not significantly associated with local tumor control. This result, however, should be interpreted with caution. Only 17 tumors were centrally located, and the resulting imbalance between central and non-central lesions limited the ability to detect a difference between the groups. Thus, the absence of statistical significance does not exclude a clinically relevant effect of tumor location.

This study had several limitations. First, this was a retrospective, single-institution study without randomization, and the sample size was small. The inclusion of only 28 patients over a prolonged study period reflects the rarity of pulmonary metastases of soft tissue and bone sarcomas treated with percutaneous cryoablation under local anesthesia and the highly selected nature of this treatment. The small sample size limited the statistical power, precluded robust subgroup and multivariable analyses, and may reduce the robustness and generalizability of the findings. Several patients had multiple treated lesions, which may have introduced intra-patient correlation in the lesion-level analysis. Because the cohort included only 28 patients and 12 local progression events, generalized estimating equations or mixed-effects models were not applied owing to concerns regarding model instability and unreliable estimates. Accordingly, the lesion-level findings should be interpreted as exploratory.

Second, this was a historical cohort treated between 2002 and 2016, although the basic procedural protocol and treatment concept of cryoablation remained consistent throughout the study period at our institution. Advances in ablation devices, peri-procedural care, and systemic therapies after the study period may limit the direct applicability of these findings to contemporary practice. Systemic therapies administered after cryoablation were also heterogeneous in regimen, timing, duration, and indication. Because the number of local tumor progression events was limited, robust adjustment for post-ablation systemic therapy was not feasible. The local tumor progression analysis should therefore be interpreted cautiously, considering the potential influence of systemic therapy administered after cryoablation.

Third, the median CT follow-up duration for each tumor (22.8 months) was shorter than that for each patient (34 months), which may have led to an overestimation of efficacy rates after cryoablation. Late time-point estimates, particularly the 10-year estimates, should also be interpreted cautiously because the number of tumors and patients at risk was limited at later time points. This may have reduced the reliability of the Kaplan–Meier estimates beyond 5 years. Another limitation of the tumor-level time-to-event analysis was that death without documented local tumor progression was not treated as a competing event. Although such deaths may preclude subsequent detection of local tumor progression, cumulative incidence estimates were not used because the number of local progression and competing events was limited and later time-point estimates were considered statistically unstable. This analytic approach may also have overestimated the primary and assisted efficacy estimates. Local tumor progression was determined by consensus review of follow-up CT images rather than by independent blinded readings. Therefore, formal inter-reader agreement statistics were not available, and potential variability in imaging assessment may not have been fully captured.

Fourth, the histological subtypes of soft tissue and bone sarcomas and treatments performed before cryoablation varied widely. Therefore, the results cannot be directly generalized to all patients with pulmonary metastases of soft tissue and bone sarcomas. Larger, contemporary, multicenter prospective studies are warranted to validate these findings and to better define the role of percutaneous cryoablation in the multidisciplinary management of pulmonary metastases of soft tissue and bone sarcomas.

Conclusion

In this selected single-institution cohort, percutaneous cryoablation under local anesthesia was feasible and was associated with local tumor control and an acceptable safety profile in nonsurgical candidates with pulmonary metastases of soft tissue and bone sarcomas. These exploratory findings should be interpreted cautiously and require validation in larger contemporary studies.

Abbreviations

CT

Computed tomography

CTCAE

Common Terminology Criteria for Adverse Events

ECOG

Eastern Cooperative Oncology Group

MWA

Microwave ablation

OS

Overall survival

RFA

Radiofrequency ablation

SBRT

Stereotactic body radiation therapy

WHO

World Health Organization

Authors’ contributions

(I) Conception and design: YA, KK, and KA; (II) Administrative support: KK, TH, and KA; (III) Provision of study materials or patients: KK; (IV) Collection and assembly of data: YA, KK, YO, and KM; (V) Data analysis and interpretation: YA, KK, and KA; (VI) Manuscript writing: all authors; and (VII) Final approval of the manuscript: all authors.

Funding

This study was supported by unrestricted institutional funding. The funder played no role in the study design, data collection, data analysis, data interpretation, or manuscript writing.

Data availability

All data and material relevant to the study are available from the corresponding authors upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki (revised in 2013). The Medical Research Ethics Committee of Keio University School of Medicine approved this study (approval no.: 20210003), and the requirement for informed consent was waived due to the retrospective design. The authors are accountable for all aspects of the work and ensured that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Competing interests

The authors declare no competing interests.

Consent for publication

Not applicable.