PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The study was approved by the local Institutional Review Board. In a retrospective review, we identified patients with clinical stage T1a renal tumours treated with PN or RFA between July 1996 and January 2004, and who had a complete 2-year follow-up. Patients with bilateral synchronous or metachronous tumours, metastatic disease at presentation, or hereditary renal cancer syndromes were excluded. Of the 110 PN identified, 28 patients were diagnosed with multifocal disease and 45 were eliminated because they had tumours of >4 cm, <2 years of follow-up or a history of a previous renal mass.
The present patients had open PN (30), laparoscopic PN (seven), percutaneous RFA (26) or laparoscopic RFA (14), based on tumour size, location and the surgeons’ preferred approach. In this series open PN was used since 1996, laparoscopic PN since 2000, and our renal RFA programme began in 2001. All patients had preoperative contrast-enhanced (CE) abdomino-pelvic CT or MRI. Treatment was based on the presence of a solid enhancing renal mass suggestive of malignancy. The choice of surgical approach was individualized, based on discussions between the patient and surgeon about the advantages and disadvantages of PN vs RFA.
Survival was analysed using life-tables and the Kaplan–Meier method; two-sided tests of significance were used, and the log-rank statistic used to compare survival rates, with P < 0.05 taken to indicate significance.
We previously described our RFA technique for renal tumour ablation . The laparoscopic or CT-guided percutaneous approach was chosen, based on tumour location and proximity of adjacent organs. RF energy was applied using a model 1500 RF generator (RITA Medical Systems, Mountain View, CA, USA). A 14 G Starburst XL probe was used and tines were deployed to create an ablation zone diameter 0.5–1.0 cm beyond the CT-measured maximum tumour diameter. The generator modulated power up to 150 W to achieve an average temperature of 105 °C. Tumours of <2 cm were treated for two cycles of 5 min, those 2–3 cm for two cycles of 7 min and those >3 cm for two cycles of 8 min . Extra cycles were used at the surgeon’s discretion if ablation was considered to be incomplete on radiological inspection. Biopsies were obtained immediately before ablation in the percutaneous RFA group using a 20-G Tru-cut needle, or during surgery and after ablation in the laparoscopic RFA group, using a toothed biopsy forceps .
The follow-up for each RFA patient included a physical examination, chest radiography, liver function tests, and CE-CT or CE-MRI at 6 weeks, 3 months, 6 months, 1 year and every 6 months thereafter. Incomplete ablation was defined as any portion(s) of the treated lesion that enhanced on the 6-week CE-CT or CE-MRI. Recurrence was defined as growth of the tumour or any new enhancing portion(s) after initial imaging showed no lesion enhancement. Shrinkage of the ablated lesion was not a requirement for the success of ablation, as long as there was no growth and enhancement . Patients were deemed to have no evidence of disease (NED) if there were no clinical or radiological evidence of distant metastasis or local disease.
For the PN technique, the open PN used a flank approach; the renal vessels were generally dissected to allow tumour excision under cold ischaemia. The tumour was excised sharply, the pelvicalyceal openings and vessels were oversutured and the cut surface cauterized. Haemostatic pledgets, fibrin sealant and chromic suture were used for renorrhaphy.
For laparoscopic PN, using the transperitoneal approach, laparoscopic ultrasonography was used to delineate the depth and extent of the mass. The renal vessels were temporarily occluded with bulldog clamps or a Satinsky clamp for tumour excision under warm ischaemia. The cortical surface of the resection bed was cauterized with the argon-beam coagulator, and renorrhaphy completed using sutures, Floseal (Baxter Inc, Deerfield, IL, USA), Surgicel and fibrin glue. The follow-up included a physical examination, chest radiographs, liver function tests and annual CE-CT or CE-MRI.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The baseline characteristics of the 77 eligible patients are shown in Table 1; one patient in the RFA group had <2 years of follow up. This patient developed a PUJ obstruction and had nephrectomy 18 months after the original ablation of a Fuhrman grade 2, 2.3 cm clear cell RCC. Final pathology confirmed no evidence of residual cancer in that kidney and she remains with no evidence of disease recurrence. She was included in the present analysis because she represents a major complication from ablative surgery and is pathologically NED. The mean (range) follow-up for the RFA and PN groups was 30 (18–42) and 47 (24–93) months, respectively (P < 0.001); the mean tumour size was 2.41 and 2.43 cm (P = 0.45), respectively. The confirmed RCC rate was similar between the techniques (Table 1; 81% vs 89%; P = 0.47). The non-diagnostic rate was also similar in both groups, at three (8%) for the RFA and three (8%) for the PN group.
Table 1. The patients’ characteristics and tumour pathology
|Mean (sd) or (range):|
| tumour size, cm|| 2.41 (0.70)|| 2.43 (0.80)||0.45|
| age, years||60.5 (13.5)||56.4 (12.5)||0.63|
| follow-up, months||29.8 (13–42)||46.7 (24–93)||<0.001|
|Complications, n (minor/major)|| 2/3|| 2/1|| |
|Failures, n (%)|
| incomplete RFA|| 1 (2)||–|| |
| recurrence|| 2 (5)|| 2 (5)|| |
|Pathology*|| || ||0.47|
| clear cell||24||23|| |
| chromophobe|| 1|| 4|| |
| papillary|| 5|| 3|| |
| cystic renal cell|| 0|| 3|| |
| oncocytoma|| 3|| 1|| |
| angiomyolipoma|| 1|| 0|| |
|Not diagnostic|| 3|| 3|| |
|No biopsy|| 3|| 0|| |
There were three treatment failures in the RFA group, i.e. one incomplete ablation (3%) and two local recurrences (5%) with no distant metastases. Of these failures, two were endophytic tumours and one exophytic; each was posterior. The patient with incomplete ablation did not have a biopsy before ablation; she was re-ablated percutaneously and has been NED for 42 months. Of the recurrences, one was diagnosed with clear cell carcinoma, recurred 18 months after RFA, was re-ablated percutaneously, and is NED 1 year after re-ablation and 2.5 years after initial treatment. The second patient had percutaneous RFA of a 3.3-cm clear cell carcinoma. After a negative 6-week CE-CT she was lost to follow-up until the 24-month CE-CT showed local recurrence. This patient had a laparoscopic nephrectomy and final pathology showed a T1a clear cell RCC. She was disease-free at 1 year after nephrectomy. The RFA kidney preservation rate was 97.5%.
There were two recurrences (6%) in the PN group; one patient developed a 2-cm enhancing mass in the contralateral kidney 52 months after open PN, but has had no further intervention. The second patient developed a 1.2-cm enhancing mass, adjacent to the previous resection site 2 years after PN. Subsequent radical nephrectomy revealed a 3-cm pT3b grade 2 clear cell RCC.
The 3-year recurrence-free survival rate for all patients was 93.4% and 95.8% for the RFA and PN groups, respectively (P = 0.67; Fig. 1a). For patients with pathologically confirmed RCC at initial RFA or PN, the 3-year recurrence-free survival rate was similar (91.4% vs 95.2%, respectively, P = 0.58; Fig. 1b). There were no disease-specific deaths in either group, i.e. a 100% disease-specific survival.
Figure 1. Kaplan–Meier estimates of recurrence-free survival probabilities according to surgical approach for patients: (a) treated for clinical stage T1a renal masses; and (b) with diagnosed RCC.
Download figure to PowerPoint
There was one major and two minor complications in the PN group. One patient developed a flank-site hernia that was repaired with no complications. Two patients, one after LPN and one after a thoraco-abdominal approach, developed prolonged ileus and were both managed conservatively. There were three major and two minor complications in the RFA group. One patient developed a PUJ obstruction and subsequently had a nephrectomy for a non-functioning kidney. A second patient was incorrectly diagnosed with a recurrence at the margin of the ablated tumour and had a salvage nephrectomy, from which pathology revealed only a giant-cell reaction with no viable tumour. We included this case as a complication of failing to properly interpret this patient’s CT. The third patient, with multiple medical comorbidities and American Society of Anesthesia Class 3, developed pneumonia after surgery, but recovered. The minor complications consisted of one asymptomatic lower-pole hydrocalyx and one case of temporary probe-site numbness.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Historically, surgical oncology was guided by the Halsteadian principle of wide, en bloc tumour excision. Surgery has developed into a more organ-sparing approach, and newer technologies have added in situ needle ablation to the treatment of urological malignancies. To our knowledge this is the first attempt to compare intermediate (>2 years) outcomes of patients with cT1a renal masses treated with RFA vs standard PN.
PN is known to offer excellent long-term oncological cure; Fergany et al. reported their 10-year follow-up for PN, with a cancer-specific survival of 88.2% at 5 years and 73% at 10 years. Of their 107 patients, 68% were symptomatic at presentation, with a mean tumour size of 4.7 cm. A retrospective review from Memorial Sloan Kettering evaluated 252 patients who had either nephrectomy or PN for renal masses of <4 cm . The disease-specific survival for the PN group at 3 and 5 years was 96%. Herr  also reported the 10-year follow-up for a group of 70 patients who had PN for tumours predominantly of <4 cm and a normal contralateral kidney, with two cancer-specific deaths and a similar, 97% disease-free survival at 10 years.
Renal ablative surgery is a newer technology, but abstracted intermediate follow-up data appear to mirror the outcomes seen with PN in contemporary series [4,9–11]. McDougal et al. recently reported their experience in 16 patients, from their larger cohort, who were followed for ≥ 4 years; only one in this smaller group was found to have recurrence by CE-CT. Varkarakis et al. also reported on their experience with RFA for renal masses of ≤ 4 cm; in their study of 49 patients, 43 were successfully ablated on the first attempt. Six patients required re-ablation for residual tumour enhancement at the 6-month follow-up CE-CT. Overall, only three of 49 patients (5%) recurred, at 24, 25 and 31 months.
In the present series of PN there were two late recurrences in 37 patients (5%), and one contralateral asynchronous recurrence. However, we understand that such a recurrence almost certainly does not represent metastatic disease, but include it for completeness. The recurrence-free probability for the RFA and the PN groups, respectively, was 93.4% and 95.8% (P = 0.674); Fig. 1). For the patients with pathologically confirmed RCC the results were also comparable. Importantly, while one of the RFA failures was re-ablated and remains disease-free, that patient is still designated as a failure in the actuarial analyses. Based on this direct comparison and the experience of others, we think that RFA has, and will continue to have, technical and oncological success when used for cT1a renal tumours.
The present study also underscores several points about renal RFA. First, clinical results after RFA are consistent among experienced centres. Second, treatment with RFA requires long-term follow-up imaging. Third, RFA need not be offered to patients as a ‘one time only’ treatment. Unlike PN, renal tumour ablation, uniquely, allows for re-treatment with no increase in morbidity or surgical difficulty. This is highlighted in our and other series [2,7,12]. In the present series, we reported two recurrences (5%) in 40 patients, and one incomplete ablation (3%). One of the recurrences was safely re-ablated and the patient is cancer-free 30 months after initial ablation. The one incomplete ablation was re-ablated and the patient is tumour-free at 42 months from initial ablation. Thus, two-thirds of the patients who needed additional treatment were re-treated percutaneously, on an outpatient basis, and remain tumour-free at >2 years.
RFA can offer certain advantages over PN as a nephron-sparing treatment. Endophytic lesions that would be difficult to locate during open PN and technically challenging for laparoscopic PN can generally be targeted and ablated with CT-guided RFA. Renal ischaemia is not necessary for RFA, and RFA can be delivered either percutaneously or laparoscopically, dramatically decreasing the convalescence compared with open PN. Furthermore, RFA is technically easier than laparoscopic PN, which is challenging and associated with potential complications. Ramani et al. recently reported their expert experience with laparoscopic PN; overall, 33% of patients had one or more complication. Intraoperative haemorrhage occurred in seven (3.5%), delayed haemorrhage in eight (4%) and urine leakage in nine (4.5%). In the present series of predominantly open PN, we report one (3%) major and two (5%) minor complications.
However, ablative surgery requires training and experience, and has associated complications. A recent multi-institutional review of 133 cases of RFA outlined the 30 complications that these groups encountered. Probe-site pain and or parasthesia were the most common complications . Weizer et al. reported their experience with 24 patients who had percutaneous RFA. Of their five patients who had complications, three were considered major, including one ureteric stricture and two colonic injuries. At our institution, to avoid bowel injury, we recommend laparoscopic RFA for anterior tumours. Another drawback to all of the renal ablative techniques is the required protocol of aggressive follow-up imaging. At present, the radiographic follow-up for RFA is extensive and financially cumbersome. As a longer follow-up for RFA accrues, it is not unreasonable to expect these imaging intervals to widen.
The current analysis has several limitations. First, this is a retrospective review with relatively few patients, but we are comparing contemporary cohorts matched in tumour and patient characteristics who presented for management at a single tertiary referral centre. Second, the success of RFA was assessed by radiographic findings, as opposed to pathological margin-free status. As such, despite the intermediate follow-up reported, it might take much longer to detect RFA failures radiographically. Similarly, T1a renal masses have very low metastatic potential, and hence it might take a longer follow-up to unmask a metastatic foci from an incomplete ablation or local recurrence after RFA. Nevertheless, in this preliminary study directly comparing RFA and PN, the outcomes were favourable.
In conclusion, the intermediate oncological outcome for patients with cT1a renal cortical tumours treated with RFA compare well with those from PN. However, a longer follow-up and prospective randomized studies are required to further define the role of RFA in the management of small renal tumours.