Second Prize: Recurrence Rates After Percutaneous and Laparoscopic Renal Cryoablation of Small Renal Masses: Does the Approach Make a Difference?

Introduction

I n the United States, the population of persons ≥65 years and ≥80 years is projected to more than double from 35 and 9.3 million in 2000 to 71 and 19.5 million, respectively, by 2030. ¹ In 2007, renal-cell carcinoma (RCC) had a median age at diagnosis of 65 years and an incidence of 51,590 cases. ² At a minimum, the aging population and increasing use of radiographic imaging portends an ever-increasing incidence of RCC in the future. ³ Early detection has led to the downward stage migration and an increase in the proportion of small renal masses (SRM). Currently, 70% of RCC are clinical stage T_1a at presentation. ⁴ When and in whom observation or treatment of a SRM is warranted is the subject of intense debate. ⁵ Interventions in which more nephrons are lost for stage T_1a disease put the patient at risk for chronic kidney disease, which has been shown to increase the risk of hospitalization, cardiovascular events, and death. ⁶

Minimally invasive nephron-sparing surgery (MINSS) treatment options are increasingly being included in management decisions because of an increasingly morbid and aged population. ⁷ Percutaneous renal cryoablation (PRC) can be performed with local sedation, is less costly, and has shorter hospitalization and recovery times than transperitoneal laparoscopic renal cryoablation (TLRC). ^8,9 We compare these approaches, focusing on their intermediate-term oncologic outcomes and morbidity.

Patients and Methods

With Institutional Review Board approval, multicenter, retrospective data of 145 patients who were treated with either PRC (61) or TLRC (84) were compiled from September 1998 to May 2010 in five academic medical centers in the United States. Patients with fewer than 12 months of follow-up were excluded from analysis.

Patient selection for intervention was based on radiographic clinical stage T_1a (cT_1a) enhancing renal masses and/or Bosniak classification III and IV renal cysts. Most often, patients were offered PRC and/or TLRC rather than partial nephrectomy if they were assessed to be at higher risk for complications from excisional surgery. In this series, maximum renal mass diameter on preoperative imaging for non-cT_1a lesions ranged between >4.0 and ≤6.0 cm (cT_1b) in 10/145 (6.9%) patients. This subgroup of cT_1b patients were treated, because intervention was deemed necessary and more invasive treatments options were not possible because of patient comorbidities. Primarily, patients with anterior lesions were offered TLRC, while PRC was used for posterior lesions. Patients with severe coagulopathy were not offered cryoablation. Both PRC and TLRC were performed as described by Derweesh and associates ¹⁰ and Malcolm and colleagues. ¹¹ using argon-cooled and helium-warmed cryoprobes (Cryocare System,™ Endocare, Irvine, CA; SeedNet Gold Cryosurgery System,™ Galil Medical, Yokneam, Israel).

PRC was performed using CT-guided placement of the cryoprobes. The majority of cases were performed with conscious sedation and local anesthesia. General anesthesia was used when necessary. CT of the lesion was initially performed, followed by renal biopsy and placement of an appropriate number and type of cryoprobe to create an iceball that would extend 1.0 cm beyond the margin of the lesion. Cryoablation was most often performed with two 10-minute freeze cycles, to a temperature of −40°C at the margin, interspersed by an 8-minute thaw cycle. Treatment was considered adequate if intermittent noncontrast CT showed iceball extension 1.0 cm beyond the tumor margin during freezing and a probe temperature of −130°C was attained. A final post-thaw CT without ± with intravenous (IV) contrast was obtained, with or without the cryoprobes in place.

TLRC was performed after medial reflection of the colon with exposure of the renal hilum and renal tumor or capsule. Hilar clamping was not used. Most often, biopsy of the lesion was performed unless tumor seeding of a cystic lesion was a concern. Direct laparoscopic visualization was used in concert with laparoscopic ultrasonography to define the lesion, perform renal biopsy, and guide placement of tumor size and shape-appropriate number of cryoprobes. Probes were placed to ensure that an iceball extended 1.0 cm beyond the margin, allowing for a temperature of −40°C at the margin. Usually two 10-minute freeze cycles interspersed by an 8-minute thaw were administered with ultrasonographic monitoring of the iceball. If necessary, hemostatic agents were applied to the cryolesion afterward.

Follow-up included serial physical examination, serum biochemical indices, chest radiography and CT (without and with IV contrast) imaging every 3 months for the first year, every 6 months during year 2, and annually thereafter. Contrast enhancement and/or interval growth of the cryolesion was deemed a recurrence and treatment failure. Repeated biopsy and re-treatment were recommended if appropriate.

Entered data were analyzed in subgroups based on treatment approach (PRC and TLRC). Analyzed data included basic demographic information (age, sex, body mass index [BMI, kg/m²], American Society of Anesthesiologists (ASA) score, comorbidities including diabetes mellitus, hypertension and smoking), disease characteristics (renal lesion/cyst diameter and location, precryosurgery biopsy pathology), outcomes (local recurrence, disease-free survival [DFS], overall survival [OS]), and complications as classified by Clavien and coworkers. ¹²

Results

Sixty-one (69.7% male; 30.3% female) patients underwent PRC, having a mean age of 68.6 ± 12.8 years (range 42.3–87.6 y) and a mean BMI of 29.9 ± 6.0 kg/m² (range 19.4–49.0 kg/m²). Eighty-four (58.3% male; 41.7% female) patients underwent TLRC and had a mean age of 65.7 ± 10.0 years (range 33.8–82.6 y) (P = 0.114) and a mean BMI of 29.8 ± 6.6 kg/m² (range 18.9–48.9 kg/m²) (P = 0.249). Comorbidities in the PRC and TLRC cohorts included a mean ASA score of 3.3 ± 0.7 (range 3–4) vs 3.5 ± 0.7 (range 2–4) (P = 0.227), diabetes mellitus in 23.0% vs 33.3% (P = 0.198), hypertension in 73.8% vs 78.6 (P = 0.554), and a smoking history in 44.3% vs 58.3% (P = 0.134), respectively (Table 1).

Mean tumor size in the PRC and TLRC groups was 2.7 ± 1.1 and 2.5 ± 0.8 cm (P = 0.090), respectively. Tumor size >4.0 cm but ≤6.0 cm was noted in 6/61 (9.1%) and 4/84 (4.8%) (P = 0.262). Renal masses, 76.4 vs 58.3%, (P = 0.035) were biopsy-confirmed RCC, and mean follow-up was 31.0 ± 15.9 vs 42.3 ± 21.2 months (P = 0.008), respectively. In the PRC and TLRC groups, location was anterior, central, or posterior in 24.6%, 19.7%, and 55.7% vs 11.9% (P = 0.046), 65.5% (P < 0.001) and 22.6% (P < 0.001), respectively. Collecting system involvement was noted in 23.0% vs 20.2% (P = 0.688), respectively (Table 1).

Local recurrence in the PRC and TLRC groups was noted in 10/61 (16.4%) and 5/84 (5.9%) kidneys (P = 0.042), respectively. Of those patients with recurrence, 8/10 (80.0%) vs 4/5 (80.0%) (P = 1.000) were biopsy-proven RCC before the initial cryosurgery treatment and 6/10 (60.0%) vs 5/5 (100%) were upper pole tumors (P = 0.231). In patients with treated tumors >4.0 and ≤6.0 cm, 2/10 (20.0%) vs 0/5 (0.0%) occurred (P = 0.524). In the PRC cohort, recurrences were managed with nephron-sparing surgery (NSS) in 7/10 (70.0%) and radical surgery in 3/10 (30.0%) patients. In comparison with PRC recurrences, NSS was used for recurrences in 2/5 (40.0%) (P = 0.329) and radical surgery was used in 3/5 (40.0%) patients (P = 0.329) after TLRC. Characteristics of recurrences and the modality of NSS or radical surgery used for them are further described in Table 2.

In the PRC cohort, DFS and OS were 93.7% and 88.9%, with four patients having evidence of disease at last follow-up. DFS and OS were 91.7% and 89.3% in the TLRC group, with seven patients having evidence of disease at last follow-up. DFS (P = 0.654) and OS (P = 0.939) were similar (Table 1).

The number of perioperative complications (9 vs 13, P = 1.000) and patients experiencing complications (8/61 [13.1%] vs 10/84 [11.9%], P = 1.000) were similar in both PRC and TLRC groups. Complications in the PRC and TLRC groups included bronchospasm in 1.6 % vs 1.1% (P = 0.736), cellulitis in 0.0% vs 1.1% (P = 0.387), perinephric hematoma in 6.6% vs 3.5% (P = 0.234), hypotension in 1.6% vs 0.0% (P = 0.251), ileus in 0.0% vs 2.4% (P = 0.219), pleural effusion in 0.0% vs 1.1% (P = 0.387), pneumothorax in 1.6% vs 0.0% (P = 0.251), transfusion in 3.2% vs 3.6% (P = 0.775), urinary retention in 0.0% vs 1.1% (P = 0.387), and urinary tract infection in 0.0% vs 1.1% (P = 0.387), respectively. Complications as classified by Clavien and colleagues ¹² were as follows: Grade I (4 vs 7), Grade II (4 vs 6) and Grade IIIa (1 vs 0) (Table 1).

Discussion

In this multicenter, retrospective study, DFS and OS were similar for patients who were treated with either PRC or TLRC (Table 1). In a meta-analysis that compared surgical cryoablation (TLRC and open RC) vs PRC of renal masses, Hui and associates ¹³ demonstrated that PRC was no less effective in terms of DFS after repeated treatment was taken into account. Series with relatively short follow-up intervals that compared PRC and TLRC have reported low and similar local recurrence rates. ^13,14 Our data demonstrate a significantly increased risk for local recurrence with the percutaneous approach (10/61 [16.4%] vs 5/84 [5.9%], P = 0.042) at an intermediate follow-up (mean 37.6 mos). At this point, it is the comparative study with the longest follow-up. Some recent studies in the literature are summarized in Table 3. ^{8,14 –20}

It has been suggested that PRC may have a higher recurrence rate for a variety of reasons, including the fact that it is a relatively newer modality in the early stage of the learning curve and complicated upper pole treatments in close proximity to surrounding structures make treatment difficult and complex. ¹¹ Compared with our PRC cohort, the follow-up was significantly longer for our TLRC patients: 42.3 ± 21.2 vs 31.0 ± 15.9 months (P = 0.008), respectively. Also, the fact that PRC is more often performed without general anesthesia may lead to a higher risk of “off-target” treatment secondary to a lack of paralysis and/or control of respiration. Furthermore, unlike PRC, TLRC offers direct visualization of the tumor laparoscopically, in addition to sonographic monitoring, adding another dimension in tracking iceball coverage.

Of the local recurrences, upper pole tumors that were treated with cryoablation by either PRC or TLRC (11/15 recurrences in the upper pole, 6 PRC and 5 TLRC) comprised a significantly greater proportion than either midpolar or lower pole tumors (P < 0.001). When comparing upper pole recurrences between the PRC (6/10) and TLRC (5/5) groups, however, there were no significant differences (P = 0.231). For the other four PRC local recurrences, all occurred in those patients with a midpole tumor adjacent to the collecting system, consistent with the study by Wright and coworkers ¹⁵ that reported a higher failure rate for TLRC of endophytic lesions. Unlike radiofrequency ablation of renal masses, surveillance of cryosurgery lesions is relatively straightforward where recurrence is signaled by incremental size increase and/or contrast enhancement of the cryolesion. ²¹

While a higher proportion of patients with PRC failures underwent nephron-sparing salvage treatment when compared with patients in the TLRC group (7/10 [70.0%] vs 2/5 [40.0%]), this did not reach statistical significance (P = 0.329). Definite conclusions about which initial treatment approach (PRC vs TLRC) may be more effectively salvaged by a further nephron-sparing procedure cannot be drawn from this analysis. Further study with longer-term follow-up and greater numbers may answer this question.

As with previous studies, ours had a relatively low rate of complications. ⁸ Malcolm and associates ¹¹ reported that with experience, cryoablation is increasingly being offered to younger and healthier patients with SRMs at their center. While the results are promising, Nguyen and colleagues ²² suggest that offering cryosurgery to a wider and younger range of patients should be done with prudence, because “extensive perinephric scarring” is often encountered if re-treatment is necessary, rendering extirpative surgery, especially laparoscopic and nephron-sparing modalities, more difficult and morbid.

Strengths of the study include the number of patients and the duration of follow-up in each cohort. A limitation of the follow-up duration in this study, however long in comparison with other studies at this time, is that follow-up should be up to or beyond 60 months to come to more concrete oncologic conclusions. Study weaknesses include its retrospective nature and inconsistencies inherent in multicenter studies affecting reports and protocols for imaging, surveillance, and recurrence management. Also, the schedule for postcryotherapy imaging is extensive in this cohort. Not only is this costly, it may be putting patients at increased risk for radiation-induced malignancies. Postcryotherapy imaging protocols need to be optimized to provide the highest diagnostic yield while minimizing unnecessary radiation exposure. Future study is needed to determine the optimum radiologic follow-up interval.

Conclusions

PRC and TLRC both continue to provide oncologic control of SRMs. In this multicenter study of well-matched patient cohorts with intermediate follow-up, significantly higher local recurrence rates were observed in the PRC group. While no differences were noted between DFS and OS, the analysis is limited by its intermediate follow-up. Caution should continue to be exercised when offering both modalities to young and healthy patients, given the risk of re-treatment. Further study is necessary to discern the reasons for the higher recurrence rates in PRC and its associated long-term consequences.