Prognostic evaluation of transcatheter arterial embolization for unresectable renal cell carcinoma with distant metastasis


T. Onishi, Department of Urology, Aoto University Hospital, Jikei University School of Medicine, 41–2, 6-chome, Aoto, Katsushika-ku, Tokyo 125–8506, Japan.


Objective To evaluate the efficacy of transcatheter arterial embolization (TAE) in patients with unresectable renal cell carcinoma (RCC) and distant metastasis at the time of diagnosis.

Patients and methods The study included 54 patients with histologically defined RCC (by biopsy in 28 and autopsy in 26) who were unable to undergo nephrectomy mainly because their performance status was poor (score geqslant R: gt-or-equal, slanted 2). The patients were classified into two groups; 24 patients who underwent TAE with ethanol and 30 patients who did not. The two groups were compared for several clinical factors, mainly focusing on the prognosis.

Results There were no significant differences in the clinical factors, including performance status, tumour diameter, vascular invasion, lymph node involvement, adjuvant therapy, metastatic organs or the number of metastases between the groups. However, the proportion of patients with para-neoplastic signs in those undergoing TAE was greater than that in those not, and the difference was significant (chi squared 0.35, P < 0.05). The median survival of the two groups was 229 days (TAE) and 116 days (no TAE). The 1-, 2- and 3-year survival rates in the TAE group were 29%, 15% and 10%, respectively, and in those not undergoing TAE were 13%, 7% and 3%, respectively. Those undergoing TAE had a significantly better prognosis than those who did not (P = 0.019). The adverse effects in patients undergoing TAE with ethanol included fever, back pain on the affected side, nausea and vomiting, but all the patients recovered from these adverse effects.

Conclusion TAE with ethanol is a safe and effective treatment for patients with unresectable disseminated RCC and a poor performance status; TAE with ethanol not only induces ablation of the primary tumour, but also prolongs survival.


With the wide clinical application of abdominal CT and ultrasonography in recent years, the incidental detection of RCC in asymptomatic patients has increased [1]. This increase in incidental cases had led to an improvement in the prognosis of patients with RCC [1]. However, the proportion of patients with RCC and distant metastasis, i.e. advanced cases, has not decreased and management using non-surgical therapies, e.g. interleukin-2 and interferon, has generally produced disappointing results [2].

In the early 1980s, we introduced transcatheter arterial embolization (TAE) with absolute ethanol both for patients with localized RCC as a preoperative treatment, and for high-risk patients with disseminated RCC, to replace nephrectomy. There was controversy as to whether the efficacy of preoperative embolization for both localized and unresectable cases would result in improved survival [3,4].

From this clinical background, we retrospectively analysed the clinical efficacy of TAE with absolute ethanol for patients with disseminated RCC who did not undergo nephrectomy, mainly because their performance status was poor.

Patients and methods

Of patients diagnosed with RCC between January 1980 and December 1998, 227 had distant metastases at the time of diagnosis. From these patients, 54 (24%) were selected who did not undergo nephrectomy because they had a poor performance status (score geqslant R: gt-or-equal, slanted 2). Furthermore, over 60% of these patients had more than two unfavourable factors, i.e. anaemia (haemoglobin <100 g/L), a high ESR (> 30 mm/h), an increased α-2 globulin level, fever of unknown origin (over 37.5°C), weight loss and a positive reaction to C-reactive protein (CRP). All patients had histologically defined RCC (by renal biopsy in 28 and autopsy in 26).

The patients were classified into two groups; 24 who underwent TAE and 30 who did not (mainly because they or their families refused this treatment). The two groups were compared to assess the clinical effects of TAE, mainly focusing on survival. The clinical characteristics of the patients are shown in Table 1.

Table 1.  The clinical characteristics of patients with stage M1 RCC who did not undergo nephrectomy
CharacteristicTAENo TAE
  • *

    Patients with > 2 para-neoplastic signs were more frequent in the TAE than in the no TAE group; the difference was significant (P<0.05).

Number of patients2430
Mean (range) age (years)66.2 (48–83)65.2 (29–83)
Sex, n/n (M : F)22/2 (10 : 1)24/6 (4 : 2)
Side L/R14/1012/18
Mean (range) tumour size (cm)9.7 (4.6–17)8.3 (3.5–16)
ECOG score (% in category)
216 (67)20 (67)
38 (33)9 (30)
401 (3)
Para-neoplastic signs* (%)
positive19 (79)17 (39)
negative5 (21)13 (57)
T stage (%)
T13 (13)2 (7)
T29 (38)12 (40)
T3a9 (38)11 (37)
T3b2 (8)4 (13)
T3c1 (4)1 (3)
Nodal involvement (%)
Positive10 (42)10 (33)
Negative14 (58)20 (67)
Adjuvant treatments (%)
Chemotherapy3 (13)2 (7)
Interferon-α9 (38)13 (43)
Radiation5 (21)7 (23)
None7 (29)8 (27)
Metastatic sites (%)
Lung21 (88)21 (70)
Bone9 (38)9 (30)
Brain3 (13)5 (17)
Liver2 (8)6 (20)
Others2 (8)3 (10)
No. of metastatic lesions (%)
110 (42)13 (43)
210 (42)12 (40)
≥  34 (17)5 (17)

TAEs were carried out with 5–7 F Cobra- or Simmonds-type balloon catheters (Cook, Bloomington, IN, USA, and Clinical Supply, Gifu, Japan), i.e. by balloon occlusion arterial infusion, except in six patients in whom the same type of catheters, but not balloon-assisted, were used. Because absolute ethanol was infused stepwise and multiple renal artery embolization was necessary in eight patients, the mean (range) amount of ethanol used was 15 (4–20) mL. Furthermore, embolization with ethanol was used in patients who had a large ‘parasitic’ blood supply to the tumour. Six patients had a second TAE 2–4 weeks after the first to completely embolize the tumour-related neovascular vessels. Angiography before TAE showed parasitic feeding arteries from the lumbar artery and/or the inferior phrenic artery in 13 patients. These parasitic arteries were embolized with gelatine sponge. The main renal arteries were embolized with a Gianturco coil in four patients. To evaluate the tumour size and area of necrosis before and after TAE, serial changes in tumour lesions were compared on CT. Lymph node involvement was determined by CT in 28 patients and in the remaining 26 cases was diagnosed at autopsy.

The stage of the tumour was defined according to the TNM system proposed by the UICC [5]. The time of survival in both groups was calculated as starting from the day of diagnosis and the survival rate calculated using the Kaplan–Meier method, with the significance of differences assessed using the generalized Wilcoxon method and the Cox–Mantel method. When the difference in survival was significant by both methods, the difference was taken to be significant (with the P obtained using the generalized Wilcoxon method). The chi-squared test was used to evaluate numerical differences between the groups. The patients' performance status was evaluated as five grades (0–4), according to scheme of the Eastern Cooperative Oncology Group (ECOG; Table 2; [6]).

Table 2.  Performance status according to the ECOG criteria [6]
GradeEastern Co-operative Oncology Group criteria
0Fully active, able to carry on all activities without
1Restricted in physically strenuous activity but
ambulatory and able to carry out light or sedentary
work, e.g. light housework, office work
2Ambulatory and capable of all self-care but unable to
carry out any work activities. Active > 50% of waking
3Capable of only limited self-care, confined to bed or
chair >50% of waking hours
4Completely disabled. Can not carry on any selfcare.
Totally confined to bed or chair


There were no significant differences in the age distribution, sex, affected side, tumour size, proportion with a particular performance status and T stage between the groups (Table 1). Similarly, there were no differences in the prognostic factors, e.g. nodal involvement from the results of imaging studies and features at autopsy, the proportion receiving adjuvant therapy, metastatic sites and the number of metastases between the groups. The para-neoplastic signs (anaemia, increased ESR, α-2 globulin, fever of unknown origin, weight loss and positive reaction to CRP) were more frequent in the TAE group and the difference was significant (P < 0.05).

After TAE, the tumour was smaller in 12 patients (50%) with a mean (range) reduction of 21 (14–46)%. Comparing the tumour features before and after TAE, as seen on contrast-enhanced CT, all tumours showed an enlarged area of necrosis after TAE. There was no response (defined as > 50% of tumour reduction) by the metastatic lesions after TAE . Among patients who had tumour-related symptoms (e.g. gross haematuria in 14, colicky pain induced by obstruction of the ureter with coagula in five, and a palpable renal mass) the symptoms and signs resolved in 18 (75%) after TAE.

All patients who underwent TAE had a fever (> 38°C) which continued for 3–16 days (median 7), 14 (58%) had back pain on the embolized side, which continued for 3–9 days (median 5) and seven (29%) had nausea and/or vomiting, which continued for 2–6 days (median 3). Hypertension occurred in four patients (17%) and paralytic ileus in two (8%). Most of these TAE-related symptoms disappeared after giving appropriate medication, e.g. antipyretics, analgesics or a hypotensive agent, and the remaining patients recovered spontaneously.

All patients in the TAE group died from RCC except one who remains alive with cancer; the median (range) survival in this group was 229 (61–1283) days. The 1-, 2- and 3-year survival rates were 29%, 15% and 10%, respectively. Similarly, all patients who did not undergo TAE also died from RCC, except for one alive with cancer, and their median (range) survival was 116 (55–1867) days. In this group, the 1-, 2- and 3-year survival rates were 13%, 7% and 3%, respectively. Thus patients who underwent TAE had a better prognosis than those who did not; the difference in survival rates between the groups was significant (P = 0.016; Fig. 1).

Figure 1.

Comparison of the survival rates between patients who underwent TAE (red) and those who did not (green). Those undergoing TAE had a better prognosis than those who did not; the difference in survival rates between the groups was significant (P = 0.016).


In general, the accepted role of TAE in treating locally advanced RCC is to reduce the patient's problems, including gross haematuria, and other tumour-related signs and symptoms [7]. As described recently, embolization is used as a palliative treatment for spinal cord compression by RCC, and this conventional radiological treatment is now providing opportunities for treating RCC [8]. The clinical efficacy of TAE contributes to an improved prognosis in patients with organ-confined RCC, but whether it does so in patients with disseminated disease (metastatic disease) is unclear.

The management of patients with advanced RCC has been influenced by the lack of any effective method of treatment other than surgical extirpation; nephrectomy remains the only effective treatment for the disease. However, there have been reports indicating that, among patients with an acceptable performance status, those who underwent nephrectomy survived significantly longer than those who did not [9]. In predicting the outcome in patients with advanced RCC, favourable factors include a good performance status, single-organ metastasis, lower tumour grade, lower T stage and two or fewer of the six clinical measurements (anaemia, high ESR, α-2 globulin increase, fever of unknown origin, weight loss and a positive reaction to CRP) [10,11]. There are many patients who are unable to undergo nephrectomy because of adverse factors and the prognosis in these patients is unfavourable even if treated with interferon or interleukin-2.

From previous reports, the median survival in patients with metastasized RCC who underwent nephrectomy was > 1 year [12] or 17.8 months [10]. Overall, the 1-year survival rate was 28% and the median survival 7 months in patients with metastasized RCC treated by TAE and followed by delayed nephrectomy [13]. Considering these reports [10,12,13] and the present results, nephrectomized patients with metastasized RCC had a better survival rate and median survival than those who did not undergo nephrectomy. However, according to Park et al.[14], the overall median survival in patients with unresectable disseminated RCC treated with TAE (using a mixture of ethanol and iodized oil) was 7 months; this median survival was similar to that in the present study (229 days).

Considering the clinical background of the present patients, especially that over half had more than two organs with metastases, and all had a poor ECOG status (geqslant R: gt-or-equal, slanted 2; an important prognostic factor [15]), TAE seems to be an adequate treatment for such patients with unresectable disseminated RCC, as no other effective means of therapy is available for these patients.

There were no significant background differences between the present groups except for the proportion of patients who showed para-neoplastic signs, where significantly more patients in the TAE group had such signs. As these para-neoplastic signs are accepted as important prognostic factors [9,11] we suggest that the therapeutic effect of TAE is not only on the primary lesion but also has general effects in such high-risk patients, e.g. immunological reactions against the tumour might be induced in the host [16]. Previously, the temporary remission of suspected pulmonary metastases was reported in two patients with advanced disease after treatment with TAE [17]. Therefore, further studies are needed to determine how TAE helps to improve the condition of patients with disseminated RCC, e.g. through the general induction of immunological reactions against the tumour.

In conclusion, although our experience is limited, renal TAE for patients with unresectable disseminated RCC is a convenient and tolerable treatment option. From the present results TAE with ethanol is effective in devascularizing unresectable RCC and consequently leads to improved survival.


T. Onishi, MD, Associate Professor.

Y. Oishi, MD, Chairman and Professor.

Y. Suzuki, MD, Lecturer.

K. Asano, MD, Lecturer.