Percutaneous CT- and MRI-guided Cryoablation of cT1 Renal Cell Carcinoma: Intermediate- to Long-term Outcomes in 307 Patients

Introduction

Widespread use of cross-sectional imaging has led to increased detection of incidental renal masses, some diagnosed as cT1 renal cell carcinoma (RCC) (1–3). Partial nephrectomy is the standard of care for solitary cT1 tumors (<7 cm and confined to the kidney) by several consensus guidelines, including those from the American Urological Association and the European Association of Urology (4,5), but management strategies are evolving. Imaging-guided percutaneous ablation is an alternative nephron-sparing approach and a primary treatment option for select cT1 tumors in recent National Comprehensive Cancer Network guidelines (6). Recent studies show favorable periprocedural outcomes and complication rates after ablation compared with those of partial nephrectomy (7–9).

However, ablation (percutaneous or surgical) is used in only 6%–12% of cT1 tumors (7,10,11) and is often reserved for suboptimal surgical candidates (4). Consensus guidelines warn of higher treatment failure rates after ablation (4,6), and large meta-analyses suggest reduced overall survival compared with partial nephrectomy (7,9,12,13). These conclusions rely on ablation outcome studies of heterogeneous patient populations, typically with substantial baseline comorbidities, as well as variable treatment methods that include laparoscopic and percutaneous approaches and radiofrequency, microwave, or cryoablation methods or a combination thereof (7,9,13). Recent studies focusing on percutaneous cryoablation as a primary treatment option for solitary cT1 RCC have yielded outcomes competitive with those of partial nephrectomy (8,14–16). However, more data with inclusion of long-term (>5–10 years) follow-up would be helpful to evaluate cryoablation as a durable first-line treatment for solitary cT1 tumors and inform future consensus guidelines.

Recent concerns regarding radiation exposure may also limit the use of CT-guided percutaneous ablation (17). CT-guided renal cryoablation requires CT to guide the placement of several cryoprobes and to monitor freeze-thaw cycles, which typically take longer than heat-based methods (18,19). MRI guidance remains an attractive alternative to CT because it does not use ionizing radiation and it provides superior depiction of the ice ball and renal tumor in multiple planes. However, MRI guidance is not widely adopted, perhaps because of a lack of familiarity, perceived workflow inefficiencies, or limited availability of MRI-compatible equipment. A few small early studies demonstrated the feasibility of MRI (20,21). However, none had large numbers of patients, long-term follow-up, or a direct comparison with CT to assess MRI as a viable first-line image guidance modality for renal cryoablation.

The purpose of the current study was to assess intermediate- to long-term (3-, 5-, 10-, and 15-year) outcomes of imaging-guided percutaneous cryoablation of solitary cT1 RCC and to compare CT- and MRI-guidance outcomes.

Materials and Methods

Patients

The institutional review board approved this retrospective study, which complied with the Health Insurance Portability and Accountability Act. Written informed consent was waived. A search of our interventional radiology database demonstrated consecutive patients who underwent renal ablation procedures between August 2000 and July 2017. Patients suspected of having cT1 RCC on the basis of imaging appearance were evaluated and referred by urologists or oncologists and were subsequently seen in the interventional radiology ablation clinic. Exclusion criteria were an alternative ablation modality (n = 19), lack of pathology confirmation (n = 74), and multiple or prior RCC (n = 104), resulting in patients with solitary pathology-proven cT1 RCC treated with imaging-guided percutaneous cryoablation (Fig 1).

Figure 1:

Download as PowerPointOpen in Image Viewer

Procedure and Periprocedural Care

All procedures from November 2, 2006, to September 2, 2010, were performed with CT guidance because the interventional MRI suite was unavailable; otherwise, MRI or CT guidance was determined by operator preference.

Procedures were typically performed with the patient under conscious sedation administered by an anesthesiologist (monitored anesthesia care). General anesthesia was occasionally used for patients with high risk or anxious patients. Image guidance by CT (n = 152) (Somatom Plus Sensation Open; Siemens Medical Solutions, Malvern, Pa), 3-T MRI (n = 95) (Verio; Siemens Medical, Erlangen, Germany), or 0.5-T MRI (n = 60) (Signa SP; GE Healthcare, Milwaukee, Wis) was performed by one of five staff interventional radiologists with a minimum of 7 years of experience in interventional radiology (K.T., P.B.S., S.G.S.). Tumors were ablated by using one to seven cryoprobes with one of two U.S. Food and Drug Administration–approved cryoablation delivery systems (BTG/Galil Medical, St Paul, Minn), one of which was MRI compatible.

CT-guided cryoneedle placement used CT fluoroscopy and axial helical CT, with coronal and sagittal reformats, as needed, to confirm cryoneedle positions and proximity to adjacent structures. For MRI-guided procedures, localizer images were typically performed for planning and initial cryoneedle placement, with subsequent imaging in oblique planes parallel and perpendicular to the cryoneedle axis to facilitate precise needle positioning and ice ball monitoring. Techniques of CT- and MRI-guided procedures have been described previously (20,22).

Two 15-minute freeze cycles separated by a 10-minute thaw were typically applied. The technical objective was visualization of a complete ice ball covering the tumor with at least a 5-mm margin on all sides. In selected procedures, hydrodissection with saline (n = 44), manual displacement (n = 18), or a ureteric stent (n = 3) were used to protect adjacent structures. Twenty-one patients underwent biopsy as part of the procedure.

Patients were typically observed for 6–24 hours after the procedure. Blood tests (serum hematocrit, platelets, creatinine, and myoglobin) were performed immediately after the procedure and the next morning for patients who required overnight admission. Adverse events were graded by using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0.

Follow-up

Electronic medical records and radiology reports were reviewed as of August 1, 2019. Clinical follow-up with electronic medical records and death certificate records were used to determine whether a patient was alive as of August 1, 2019. If the patient was deceased, cause of death was recorded. Imaging follow-up was assessed between the first ablation procedure and August 1, 2019. Imaging follow-up with contrast material–enhanced abdominal renal mass protocol CT (including unenhanced, nephrographic, and excretory phases), MRI (including unenhanced and nephrographic phases with subtraction images), or both was typically performed at 3, 6, and 12 months for the 1st year and then yearly thereafter. New nodular or thick crescentic tumoral enhancement was used as evidence of local progression.

Study Definitions

Primary end points were overall survival (OS), disease-free survival (DFS), local progression–free survival (LPFS), and disease-specific survival (DSS). OS was defined as the period from first ablation procedure to date of death from any cause, with censoring of patients still alive as of August 1, 2019. DSS was defined as the period between first ablation and date of death from RCC, with censoring of patients alive as of August 1, 2019, or at the date of death unrelated to RCC.

DFS was defined from the first ablation to first local progression (at the ablation site), metachronous (elsewhere in the ipsilateral or contralateral kidney), or metastatic progression (distant nodal or organ site), as detected with imaging, with censoring of patients without recurrence at the last CT or MRI date.

Secondary end points were technique efficacy, adverse event rate, and procedure duration. Primary technique efficacy was defined as complete tumor ablation without local progression at any follow-up imaging examination through August 1, 2019 (23). Secondary technique efficacy was defined as complete tumor ablation after reablation of local progression.

Procedure times were calculated as the duration from the first scout or localizer image to the final end-of-procedure image. Radiation dose-length products were recorded for 149 of 155 CT-guided procedures (dose-length products for procedures before 2005 were not available).

Statistical Analysis

The 3-, 5-, 10-, and 15-year OS and DSS estimates with 95% confidence intervals (CIs) of the overall (n = 307), CT guidance (n = 155), and MRI guidance (n = 152) cohorts were calculated by using the Kaplan-Meier method. The 3-, 5-, 10-, and 15-year DFS and LPFS estimates with 95% CIs of all patients with imaging follow-up (n = 291), CT guidance (n = 149), and MRI guidance (n = 142) were also calculated by using the Kaplan-Meier method.

Univariable Cox regression was performed to assess effects of age, tumor size (T1a vs T1b), sex, and imaging modality (CT vs MRI) on OS, DSS, DFS, and LPFS, with intention to analyze variables with a P value less than .05 in a multivariable regression model. Univariable logistic regression was used to compare primary technique efficacy between CT and MRI guidance. Statistical significance was defined as P < .05.

Matlab software (version R2014b; Mathworks, Natick, Mass) was used to calculate standard summary statistics of medians, means, ranges, percentages, Kaplan-Meier survival, and logistic regression.

Results

Patient Characteristics

Of 504 consecutive patients who underwent renal ablation procedures, 197 were excluded because they underwent treatment with an alternate ablation method (n = 19), lacked pathology confirmation (n = 74), or had multiple or previously treated RCC at the time of ablation (n = 104) (Fig 1), resulting in 307 patients (mean age ± standard deviation, 68 years ± 11; 192 men) included in our study. Baseline patient characteristics are summarized in Table 1.

Follow-up

Clinical survival follow-up was available for all 307 patients (median follow-up, 95 months; range, 8–219 months). The only patients included in the study who did not have clinical follow-up of at least 2 years duration were those who died within 2 years of the initial ablation. All other patients were followed for at least 2 years. Clinical follow-up duration was more than 3 years in 263 patients, more than 5 years in 214 patients, more than 10 years in 101 patients, and more than 15 years in 11 patients. There was follow-up imaging for 291 of 307 (95%) patients (median, 41 months; range, 0–189 months); 290 of 307 (94%) had imaging follow-up of at least 3 months duration; one patient had only a 24-hour imaging follow-up examination; and 16 patients had no imaging follow-up. We did not include the 16 patients without imaging follow-up in assessment of DFS, LPFS, or technique efficacy.

Primary End Points

Univariable regression analysis of patient age, sex, tumor size (T1a vs T1b), and imaging modality (CT vs MRI) did not demonstrate a significant effect on OS, DFS, LPFS, or DSS (Table 2). Therefore, multivariable regression analysis was not performed.

Overall survival.—During the clinical follow-up period, 55 of 307 (18%) patients died, with a median survival time after ablation of 37 months (range, 8–159 months). The Kaplan-Meier OS curve is shown in Figure 2, and estimated 3-, 5-, 10-, and 15-year OS for the total cohort, CT- and MRI-guided ablations, and T1a and T1b tumors are listed in Table 3.

Figure 2:

Download as PowerPointOpen in Image Viewer

Disease-specific survival.—Two of 307 (0.7%) patients died of RCC progression or metastatic disease within the follow-up period. One of these patients underwent ablation of a T1b (4.9-cm) clear-cell RCC and died 49 months after the initial ablation with metastatic pulmonary and osseous RCC. The other patient underwent ablation of a T1a (3.2-cm) clear-cell RCC and died 31 months later. The Kaplan-Meier DSS curve is shown in Figure 2 (Fig E1 [online] shows T1a and T1b survival curves), and estimated 3-, 5-, 10-, and 15-year overall survival rates for the total cohort, and T1a and T1b tumors are listed in Table 3.

Imaging documented DFS and LPFS.—Twenty-three of 291 (7.9%) patients with imaging follow-up developed evidence of RCC after the initial ablation. Local progression occurred in 13 of 291 (4.5%) patients; 11 underwent repeat cryoablation, and two underwent surgery. One patient developed two recurrences at the initial ablation site. Both were treated with repeat cryoablation. Local progression occurred at a median follow-up imaging time of 21 months (range, 0–42 months). Metachronous RCC (in the ipsilateral or contralateral kidney) developed in 10 of 291 (3.4%) patients; all underwent repeat percutaneous cryoablation. Metachronous tumors were seen at median follow-up of 24 months (range, 3–73 months). Three of 10 were detected at more than 70 months (72, 73, and 73 months, respectively).

Metastatic disease developed in three of 291 patients (1.0%). One patient had an initial case of local progression treated with reablation before subsequently developing lung metastases requiring wedge resection and systemic treatment. The second patient developed lung and bone metastases and was treated systemically. The third patient had both local progression and metastatic ipsilateral retroperitoneal lymph nodes; both sites were treated with repeat cryoablation.

The Kaplan-Meier DFS curve is shown in Figure 3 (Fig E2 [online] shows T1a and T1b survival curves), and estimated 3-, 5-, 10-, and 15-year DFS and LPFS for total cohort, T1a, and T1b tumors are listed in Table 3.

Figure 3:

Download as PowerPointOpen in Image Viewer

Secondary End Points

Technique efficacy.—Local progression occurred in 13 of 291 (4.5%) tumors after the initial ablation, including 11 of 13 (85%) T1a and two of 13 (15%) T1b tumors. Primary technique efficacy based on follow-up imaging until August 1, 2019, was 96% (278 of 291). Ten patients underwent successful reablation once, one patient underwent reablation twice, and two patients underwent surgical resection. Secondary technique efficacy (successful reablation after initial local progression) was 99% (288 of 291). One patient underwent reablation for primary recurrence and did not develop further local progression but later developed metastatic disease.

Adverse event rate.—Forty-three of 307 (14%) patients developed adverse events that met Common Terminology Criteria for Adverse Events criteria: 11 with grade 1 events (pain, perinephric hematoma, skin induration, mild paresthesia, mild headache); 18 with grade 2 events (myoglobinemia, urinary retention); 11 with grade 3 events (urinary tract infection, anemia, pneumonia, non–life-threatening pulmonary embolus); and three with grade 4 events (cerebrovascular accident, aspiration pneumonia, hypertensive emergency). No grade 5 adverse events occurred.

CT versus MRI Guidance

Ice ball was well seen with CT and MRI guidance (Fig 4). Survival did not significantly differ between CT and MRI guidance, with univariable Cox regression analysis hazard ratio of 0.97 (95% CI: 0.57, 1.67; P = .92) for OS, 0.63 (95% CI: 0.26, 1.52; P = .30) for DFS, and 0.83 (95% CI: 0.26, 2.74; P = .77) for LPFS (Table 2). Both disease-specific deaths followed CT-guided ablation. Kaplan-Meier survival curves for CT and MRI guidance are presented in Figures 5 and 6, and Kaplan-Meier–estimated survival with 95% CIs is detailed in Table 3.

Figure 4:

Download as PowerPointOpen in Image Viewer

Figure 5:

Download as PowerPointOpen in Image Viewer

Figure 6:

Download as PowerPointOpen in Image Viewer

Mean MRI procedure time (139.9 minutes ± 31.2) was 13 minutes longer than mean CT procedure time (126.8 minutes ± 31.1) (P < .001). Mean CT radiation dose-length product was 4427 mGy · cm ± 2527.

Discussion

Percutaneous imaging-guided cryoablation enables visualization of the ice ball at CT or MRI (14,24,25) and may be more effective than radiofrequency or other heat-based ablation methods for renal cell carcinoma (RCC) (24). However, there are limited data on the evaluation of intermediate- to long-term outcomes of percutaneous cryoablation alone or comparison of outcomes after CT and MRI guidance. We report intermediate- to long-term outcomes of percutaneous cryoablation in 307 patients with solitary pathology-proven cT1 RCC, with clinical follow-up of up to 219 months (median, 95 months) and imaging follow-up of up to 189 months (median, 42 months). We report Kaplan-Meier–estimated overall survival (OS) of 76%, disease-specific survival (DSS) of 99%, disease-free survival (DFS) of 88%, and local progression–free survival (LPFS) of 95% at 15 years. Survival did not differ between CT and MRI guidance, with hazard ratio of 0.97 (P = .92) for OS, 0.63 (P = .30) for DFS, and 0.83 (P = .77) for LPFS. Only two of 307 (0.7%) patients in our study died of RCC, both after CT-guided cryoablation. Median detection time was 8 months (range, 0–42 months) for local progression and 36 months (range, 3–73 months) for metachronous disease. Primary and secondary technique efficacy rates were 96% (278 of 291) and 99% (288 of 291), respectively. Our overall complication rate was 14% (43 of 307), with 4.6% (14 of 307) classified as grade 3 or higher.

Our OS is within the range of 5-year (9,26,27) and 10-year (28) rates reported for partial nephrectomy. Our OS is higher than prior 3- and 5-year (26,29) and 10-year (30) rates reported for radiofrequency ablation. Breen et al (14) recently reported 3- and 5-year OS with use of cryoablation, which is similar to our findings. Although our T1b sample size was small, our 3-year OS for this subset is higher than that found in prior partial nephrectomy and cryoablation series at 3 years (29).

Our DSS data are similar to those reported for partial nephrectomy (26,27,31), prior ablation studies that included radiofrequency ablation and laparoscopic approaches (15,26,27,30,32), and percutaneous cryoablation alone (14,15,33). One study found higher 5-year DSS with partial nephrectomy than with cryoablation in T1b tumors (34). Five-year DSS for cryoablation was higher in our study and more like that seen with partial nephrectomy.

LPFS in our study was similar to that reported in prior 3–5-year partial nephrectomy series (26,29) and higher than that found in prior combined ablation studies that included radiofrequency ablation and laparoscopic approaches (26,29,30,32). Our results are similar to prior 3–5-year LPFS rates for percutaneous cryoablation alone (14,15,33,35).

Several guidelines warn of a high rate of ablation treatment failure compared with partial nephrectomy (4,5). However, they are largely based on heterogeneous data combining laparoscopic and percutaneous approaches and both radiofrequency and cryoablation methods (9,36). Our technique efficacy rates were similar to previously reported 5-year efficacy rates for partial nephrectomy (26,37) and percutaneous cryoablation (14). Our complication rates were low and similar to those reported in prior ablation series (14,35,38).

These data add to the growing body of evidence that percutaneous cryoablation results in survival outcomes similar to those of partial nephrectomy in patients with cT1 tumors (13–15). Although RCC incidence has increased, mortality has remained stable despite treatment. This may indicate overdiagnosis and treatment of indolent tumors (39,40). As a result, active surveillance has gained favor in elderly patients with low-risk disease (41). Percutaneous ablation offers a potentially attractive treatment option with lower morbidity than surgery (7) and may alleviate anxiety associated with active surveillance. However, although a survival benefit of ablation has been suggested (42), the role of percutaneous ablation relative to active surveillance and surgery has remained somewhat vague, in part because of heterogeneous and limited long-term outcome data. Recent studies, including the current one, may be helpful in the development of future treatment guidelines.

Mean CT radiation dose per procedure (4427 mGy · cm) was similar to that in a previously reported series (18) and higher than previously reported doses for other common CT-guided procedures, including biopsy and catheter drainage (19). MRI-guided procedures were only slightly longer (approximately 13 minutes longer) than CT.

Our study had several limitations. It was a retrospective single-institution study. Although outcomes were competitive with previously reported partial nephrectomy studies, we did not compare patients in our study with other patients at our institution who underwent partial nephrectomy or active surveillance. We reported data for T1b tumors with favorable outcomes, but the sample size was small and the lack of survival differences between the T1a and T1b groups may be due to underpowering. More patients will be needed to evaluate cryoablation as an alternative to surgery in this subset. It may be unexpected that age did not significantly affect survival. However, on the basis of the 95% CIs reported in Table 2, this does not appear to be due to underpowering. Furthermore, most of our population was aged at least 60 years, and younger patients may have had more comorbidities than older patients, although this is speculative. Although many patients were referred for ablation because of surgical comorbidities, we did not evaluate patient performance status in our analysis. We did not assess tumor location (eg, using nephrometry scores) when comparing CT with MRI guidance, but it is unlikely tumor location made a difference because of the overall low local progression rate. Although the CT and MRI groups were similar, a more useful comparison would include a prospectively randomized design.

In conclusion, percutaneous CT- and MRI-guided cryoablation of cT1 renal cell carcinoma had similar excellent intermediate- and long-term outcomes, similar to previously reported outcomes after partial nephrectomy. These data can be used to support percutaneous cryoablation as a first-line modality in treatment candidates. MRI is a viable first-line image guidance modality that eliminates procedural radiation and may be beneficial, particularly in young patients.

Disclosures of Conflicts of Interest: S.K.B. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: holds a provisional patent (U.S. patent 20,140,187,999 A1). Other relationships: disclosed no relevant relationships. K.T. disclosed no relevant relationships. P.B.S. disclosed no relevant relationships. V.M.L. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is a consultant for Galil Medical. Other relationships: disclosed no relevant relationships. S.L.C. disclosed no relevant relationships. S.G.S. disclosed no relevant relationships.

Author Contributions

Author contributions: Guarantors of integrity of entire study, S.K.B., K.T., V.M.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, S.K.B., K.T., V.M.L., S.G.S.; clinical studies, S.K.B., K.T., P.B.S., V.M.L., S.G.S.; statistical analysis, S.K.B., K.T., V.M.L.; and manuscript editing, all authors