Surgical excision of isolated renal-bed recurrence after radical nephrectomy for renal cell carcinoma

Authors


Sarbjinder S. Sandhu, The Royal Marsden Hospital, Fulham Road, London, SW3 6JJ, UK.
e-mail: sarb@sandhu-one.freeserve.co.uk

Abstract

OBJECTIVE

To present our results on managing loco-regional recurrence of renal cell carcinoma (RCC) with surgical excision, as local recurrence at the site of a previous nephrectomy is resistant to both systemic therapy and radiotherapy.

PATIENTS AND METHODS

In all, 16 patients were operated on between 1994 and 2003 for local recurrence of RCC. The median (mean, range) age at the time of local recurrence was 57.9 (57.4, 28.9–71.7) years, and the median interval from primary surgery 2.22 (3.88, 0.27–14.46) years. Before surgery eight patients had been given systemic immunotherapy, with no response of their local recurrence.

RESULTS

Two patients were deemed inoperable because of direct invasion of the great vessels and the liver by tumour. The remaining 14 patients had recurrence in residual adrenal tissue (two), para-aortic nodes (three), para-caval nodes (two), retrocaval nodes (one), renal bed (six), liver, spleen and stomach (one each), and diaphragm (two). Although complete macroscopic en-bloc clearance was achieved in these patients, only eight had tumour-free margins on histological examination. The histology was consistent with RCC recurrence in all cases. All of the patients were followed with computed tomography at regular intervals. At a median follow-up of 1.0 (1.65, 0.25–6.5) years, five patients remain disease-free, four have local and distant relapse, and five developed distant metastasis only. The presence of tumour at the resection margin was a significant factor in predicting local and distant disease-free survival (P < 0.05).

CONCLUSIONS

En bloc excision of isolated locally recurrent RCC is possible, and complete surgical extirpation can lead to prolonged disease-free survival.

INTRODUCTION

Radical nephrectomy for localized RCC is carried out with curative intent, but the advent of stringent follow-up protocols and modern imaging techniques showed that isolated loco-regional recurrence will occur in ≈ 2% of patients [1,2]. Recurrent disease may be a result of metastatic disease within the ipsilateral adrenal gland which was left in situ at the time of the primary surgery, inadequate excision of regional lymph nodes, or recurrent or residual disease in Gerota's fat, the renal bed or within the psoas muscle. Also, inadvertent perioperative implantation of tumour may lead to local relapse.

Systemic treatments have had some benefit in managing patients with distant metastasis but they appear to be of limited benefit in local recurrence [1]. The relative radio-resistance of RCC has limited the role of radiotherapy in managing this disease to the palliation of symptoms. Surgical excision therefore remains the only option for cure in patients with locally recurrent renal tumour. We present our results on managing loco-regional recurrence of RCC with surgical excision.

PATIENTS AND METHODS

This study included 16 patients (12 men and four women) who were referred to our unit between 1994 and 2004 for operative intervention (Table 1). The patients’ details were retrieved from a prospective database. All patients had their primary surgery elsewhere and were referred for the management of renal bed recurrence. Their original histology was reviewed at our institution (Table 1). The patients were staged with cross-sectional imaging of the renal bed and lungs, and a bone scan. All patients had blood tests for a full blood count and biochemistry. The median (mean, range) age at the time of local recurrence was 57.9 (57.4, 28.9–71.7) years, the median interval between the nephrectomy and recurrence 2.22 (3.88, 0.27–14.46) years and the median interval between surgery for recurrence and primary radical nephrectomy 3.63 (5.06, 0.27–14.57) years. One patient had a previous attempt at removal of an ipsilateral adrenal metastasis; the remaining patients had only had a primary radical nephrectomy. All patients were given full bowel preparation and an intravenous infusion of saline for 12 h before surgery. The patients had cross-matched blood available.

Table 1.  The management of 16 patients with local recurrence
Age (years) atnephrectomyT stageFuhrmangradeLymphnodes*Systemic therapy before surgeryInterval, years surgery/recurTumour atexcision marginFollow-up, monthsRelapse
  • *

    0, no positive nodes; NPA, no pathology available; NN, no nodes in specimen;

  • †died from disease;

  • ‡resection not possible.

56.4NPANPANPAYes7.43No3.1No
71.5231Yes0.27No21.7No
50.8231No7.31No19.4No
61.53b2NNNo1.43No9.0Distant
49.13a30Yes1.98No29.2Distant
53.6220No9.18No22.2Distant
61.03a2NNYes8.76No52.2Local + distant
47.1220No3.14No77.9No
52.422NNYes11.31Yes9.6Local + distant
58.1121No2.17Yes7.4Local + distant
49.83a3NNYes6.37Yes14.2Distant
64.33b31Yes1.62Yes4.1Distant
50.4230Yes3.17Yes3.3Local + distant
28.8220No0.60Yes3.0No
48.920No14.572.6
49.39NPANPANPAYes4.0834.4

The most appropriate surgical incision for the procedure was used; 14 patients had a thoraco-abdominal incision with removal of the eighth rib, one had a midline incision and one had the loin incision re-explored. The thoraco-abdominal extra peritoneal approach allows direct access to the renal hilum with minimal manipulation of bowel. An aggressive attempt was made at surgical removal of all recurrent disease. Two patients were found to be inoperable; the first had disease invading the liver, and the second had disease invading both of the great vessels. In the remaining 14 patients, an en-bloc macroscopic excision of the recurrence was achieved. The sites of recurrence were; residual adrenal tissue (two), para-aortic nodes (three), para-caval nodes (two), retrocaval nodes (one), renal bed (six), liver, spleen and stomach (one each), and diaphragm (two).

The median operative duration was 67 (75, 60–135) min, the median transfusion rate 2.5 (3.88, 0–12) units, and the median inpatient stay 10.5 (10.0, 5–16) days. One patient developed a wound infection, one had a postoperative chest infection and two had incisional hernias. There were no deaths during or immediately after surgery.

Eight patients had had systemic therapy before surgery, which consisted of interferon alone in five, interferon combined with interleukin (low-dose) and 5-fluorouracil in two, and vinflunine alone in one. In the event of progression after surgery, patients either received radiotherapy (three), or systemic treatment with provera (two), interferon (three) or sorafanib (three).

RESULTS

All patients were followed with cross-sectional imaging at 3-monthly intervals for the first 2 years, 6-monthly for the next 3 years, and annually thereafter, or more frequently if clinically warranted. Only 14 patients were amenable to surgical excision, and were followed for a median of 1.0 (1.65, 0.25–6.5) years (Fig. 1A). Five patients remain free of disease at 0.25+, 0.26+, 1.62+, 1.81+ and 6.51+ years. Five further patients remain free of local disease but have evidence of metastasis, and four have developed both local and distant disease. The median time to local relapse was 2.4 (13.4, 2.1–48.8) months, and the median time to developing distant metastasis was 4.33 (11.22, 1.9–48.8) months after surgery for removal of recurrent local disease. Of the nine patients with distant metastasis, three died from their disease during the follow-up (Table 1).

Figure 1.

A, Overall survival of patients after surgical of recurrence; and B, disease-free survival of patients with tumour at margins of specimen and tumour-free margins (both Kaplan-Meier). Tumour at margins of specimen, six (green); no tumour at specimen margins, eight (red). Hazard ratio 0.21, 95% CI 0.04–1.04; chi square 6.72, d.f. 1, P= 0.0095.

Therefore, at a median follow-up of 1.0 (1.65, 0.25–6.5) years, 10 patients have no evidence of local disease and two of these are long-term disease-free survivors. The interval to local recurrence did not affect the outcome of surgical intervention for renal-bed recurrence.

On pathological examination the histology was consistent with recurrence or residual RCC, and when compared with the original histology there was no evidence of a change in Fuhrman grade. Six of the specimens had positive margins, with the remaining eight having tumour-free margins (Fig. 1B). The presence of tumour at the resection margin influenced the local and distant disease-free survival (P < 0.05).

DISCUSSION

The incidence of renal-bed recurrence in patients with metastatic RCC has been reported at 66%[3]. In retrospective series the incidence of isolated renal-bed recurrence has been documented as low as 0.77–1.8%[1,2]. It was suggested that the risk of recurrence appears to be related to the stage of the disease at primary nephrectomy, with most recurrences being in patients whose disease extended beyond the renal capsule [4]. However, in the present series, patients with stage pT1 and pT2 disease had evidence of local recurrence, as confirmed by others [2,5,6].

In the present series local recurrence occurred at a median of 2.2 and maximum of ≈ 15 years after the primary surgery, therefore a prolonged follow-up of patients after radical nephrectomy may be required. The interval between the primary nephrectomy, detection and surgery for the recurrence did not affect the outcome after excising the recurrence. This is in agreement with the findings of others [5,6] but Schrodter et al.[2], in a series of 13 patients with renal-bed recurrence, detected a better outcome in patients with a prolonged interval between nephrectomy and recurrence. The wide availability of CT and MRI has allowed renal fossa recurrences to be detected at an early stage. Early detection allows these lesions to be amenable to attempted surgical excision.

There is increasing evidence that excising local renal-bed disease can improve survival. In one series of 14 patients with residual disease after nephrectomy treated conservatively, all patients died within a year [7]. In another series of seven patients treated with renal-bed excision, four survived at 5 years [8]. In further small series, seven of 10 patients survived disease-free at a mean follow-up of 16.6 months [9], and three of 11 patients in another series did so at 3 years [6]. Itano et al.[1] followed 30 patients with local disease recurrence; 10 had surgical resection of the recurrence, with half surviving at 5 years, compared to 18% in those treated with medical therapy, and 13% in whom no treatment was given.

In the present series, at a median follow-up of 11.9 months, five patients remained completely free of disease, and the absence of positive surgical margins was an important prognostic indicator. Positive margins were predictive of local and distant relapse (P < 0.05). In all cases an attempt was made to totally excise the recurrent disease, and thus the presence of positive margins may indicate the aggressive nature of the disease, rather than inadequate surgical technique. Frozen-section sampling of the margins of the specimen may aid in determining the extent of excision. Notably, all the patients who had a relapsed locally developed distant metastasis, either before or simultaneously with the development of local relapse. As developing distant metastasis precludes further curative local surgical intervention, we consider that only one attempt should be made at surgical extirpation of the local recurrence.

Although systemic therapy does not appear to have a primary role in managing renal fossa recurrence, Tanguay et al.[5] suggested that it may have a role in conjunction with surgical extirpation, and they further suggest this may delay progression and recurrence. In their series of 16 patients, eight received adjuvant systemic therapy; half relapsed at a follow-up of 3–49 months in those receiving systemic therapy, compared to two of those treated with surgery alone, at a follow-up of 62 and 136 months.

Interleukin-2 is one of the best-characterized systemic treatments available for managing metastatic RCC. In a study of 255 patients treated with high-dose interleukin-2 [10], there was a 14% response (5% complete and 9% partial). No patient with a renal-bed recurrence had a complete response, with only three such patients achieving a partial response. The total number of patients with renal-bed recurrence recruited was not reported in that study. However, high-dose interleukin has well documented toxicity and several studies comparing the high and low dose showed no differences in efficacy, but a significant reduction in morbidity [11,12]. We therefore prefer a low-dose interleukin regimen.

In the present study, patients with tumour at the resection margins after surgical extirpation were likely to relapse locally and develop distant metastasis. Therefore, there may be a role for adjuvant systemic therapy in this selected group. Frydenberg et al.[13] suggested using radiotherapy before and after surgery; they reported eight patients treated with a combination of radiotherapy and surgery, in whom four were disease-free at a mean interval of 34 (15–50) months. Newer methods, e.g. cryoablation and radiofrequency ablation, may have a role in managing these patients, but this remains to be established [14,15].

Of concern, in four of the present patients lymph nodes were not removed at the time of the primary operation. This at least implies incomplete initial surgery, and may have prevented patients with potentially curable disease receiving definitive therapy [3,7].

In conclusion, cross-sectional imaging can accurately detect renal-fossa recurrence after primary radical nephrectomy for RCC. Imaging has been recommended in patients with T3/T4 disease after radical nephrectomy, but it may also have a role in those with T1/T2 disease. The long interval between developing recurrent disease and initial surgery implies that patients who have had a radical nephrectomy for RCC require a prolonged follow-up. The management of isolated renal fossa recurrence should be surgical. This is possible in most cases (88%), and if accompanied by tumour clearance can lead to a prolonged period of both distant and local disease-free survival.

CONFLICT OF INTEREST

None declared.

Ancillary