Fax: (011)33 1 01 58 41 27 55
Chromophobe renal cell carcinoma
Analysis of 61 cases
Article first published online: 13 FEB 2004
Copyright © 2004 American Cancer Society
Volume 100, Issue 7, pages 1406–1410, 1 April 2004
How to Cite
Peyromaure, M., Misrai, V., Thiounn, N., Vieillefond, A., Zerbib, M., Flam, T. A. and Debré, B. (2004), Chromophobe renal cell carcinoma. Cancer, 100: 1406–1410. doi: 10.1002/cncr.20128
- Issue published online: 18 MAR 2004
- Article first published online: 13 FEB 2004
- Manuscript Accepted: 12 JAN 2004
- Manuscript Received: 15 OCT 2003
- renal carcinoma;
Chromophobe renal cell carcinoma (CRCC) is often associated with a favorable prognosis. However, to the authors' knowledge, only few clinical data are available regarding this variant of tumor. In the current study, the authors report their experience with CRCC over the last 14 years.
Since 1989, 61 patients have been treated at the study institution for CRCC. Tumor characteristics and patient outcome were analyzed retrospectively. Data were obtained from the patients' medical records.
The mean age of the patients was 58 years. Of the 61 tumors, 68.8% were discovered incidentally. The mean tumor size was 6.9 cm. Fifty-seven patients (93.4%) were treated with radical nephrectomy and 4 patients (6.6%) underwent partial nephrectomy. According to the 1997 TNM classification, the pathologic tumor stage was T1 in 65.6% of cases, T2 in 31.1% of cases, and T3a in 3.3% of cases. All tumors were staged as N0M0. Nuclear grade was low (1 or 2) in 88.5% of cases. In no case of CRCC was a sarcomatoid component observed. At a mean follow-up of 49.5 months (range, 5–135 months), no patient had experienced tumor recurrence or disease progression, and none had died of renal carcinoma.
In the authors' experience, CRCC carries an excellent prognosis, possibly due to the high rate of low-stage and low-grade tumors. Cancer 2004;100:1406–10. © 2004 American Cancer Society.
Chromophobe renal cell carcinoma (CRCC) is a rare variant of renal carcinoma, with distinct histochemical, ultrastructural, and genetic characteristics. CRCC accounts for approximately 4% of all kidney neoplasms.1, 2 Although the pathologic features and molecular genetic characteristics of this tumor have been described extensively, to our knowledge there is little information available regarding the clinical outcome of patients with CRCC. To analyze the clinical behavior of CRCC further, we reviewed the data from 61 consecutive patients treated in the study institution for this type of tumor.
MATERIALS AND METHODS
Between January 1989 and January 2003, 61 patients (34 men and 27 women) were treated for CRCC at Cochin Hospital in Paris. The mean patient age was 58 years (range, 23–82 years). Data were obtained from the patients' medical records. Presenting symptoms, pathologic features, treatment, and patient outcome were analyzed retrospectively.
Preoperative investigations included clinical examination, complete blood count, renal function test, abdominal computed tomography (CT) scan, and chest radiography. Patients with suspected metastatic disease underwent pulmonary CT scan and bone scintigraphy. Renal magnetic resonance imaging (MRI) was performed in those patients with suspected renal angiomyolipoma.
In all cases, CRCC was diagnosed on a nephrectomy specimen. Pathologic studies included light microscopy and immunohistochemistry. Each tumor was reviewed by the same pathologist and staged according to the 1997 TNM classification.
Patient outcome was assessed by clinical examination and abdominal and pulmonary CT scans every 3 months for the first year, then every 6 months for the next 2 years, and annually thereafter.
The 61 cases of CRCC represented 4.1% of the 1488 renal tumors resected in the study department during this period.
The majority (68.9%) of the cases were discovered incidentally. The most common presenting symptoms were flank discomfort (18%) and macroscopic hematuria (13.1%). No patient presented with weight loss, poor general condition, or fever. At the time of presentation, no patient was found to have paraneoplastic anemia or polycythemia; only 2 patients (3.3%) had a palpable lumbar mass.
Tumors were located in the right kidney in 54% of patients and in the left kidney in 46% of patients; there were no bilateral tumors detected. The ultrasonographic and tomodensitometric appearance of CRCC was similar to that of other renal solid tumors; enhancement was observed in 93.4% of patients, necrosis was present in 31% of patients, calcification was present in 21.3% of patients, and cystic foci were present in 6.5% of patients. Two tumors (3.3%) were suspected to be oncocytomas because of the presence of a central scar. The renal vein was involved in only one case and the caval vein was normal in all cases. MRI was performed in five patients with suspected renal angiomyolipoma; the tumor was considered to be an usual solid carcinoma in four patients and an oncocytoma in one patient.
Of the 61 patients, 57 (93.4%) underwent radical nephrectomy and 4 (6.6%) underwent partial nephrectomy for small (< 3 cm) and peripheral tumors. No patient received adjuvant treatment. Although frozen sections were subjected to pathologic examination during surgery for 13 patients, a diagnosis of CRCC was established in only 2 patients.
The mean tumor dimension was 6.9 cm (range, 1.5–25 cm). Sixty tumors were solitary. One patient had 2 CRCC foci measuring 5 cm and 2.5 cm, respectively. Macroscopically, all tumors had a homogeneous, light brown (54.1%) or white (45.9%) surface (Fig. 1). Light microscopy showed that the tumors were comprised of sheets or trabeculae with variable proportions of clear cells and eosinophilic cells; 21 (34.4%) were classified as predominantly eosinophilic, 9 (14.8%) were classified as predominantly clear cell, and 31 (50.8%) tumors were found to be comprised of equal numbers of clear and eosinophilic cells (Fig. 2). In no case did the tumor have a sarcomatoid component. Some areas of necrosis were present in 36% of tumors. Hale colloidal iron stain was found to be positive in all cases (Fig. 3). At the start of the current series, ultrastructural investigation of seven tumors showed the presence of intracytoplasmic vesicles in all cases, thereby confirming the diagnosis of CRCC. Of the 61 patients, 3 had undergone renal biopsy before undergoing nephrectomy; however, in no patient had this led to a suspected diagnosis of CRCC.
The TNM stage and nuclear grade of all the tumors are summarized in Table 1. Tumors were confined to the kidney in 96.7% of patients and were of low grade (Führman grade of 1 or 2) in 88.5% of patients.
|Pathologic stagea||No. (%)|
|Nuclear Führman grade|
The mean follow-up after surgery was 49.5 months (range, 5–135 months). Nine patients were lost to follow-up, with a mean interval after nephrectomy of 13 months. During the follow-up period, no patient experienced tumor recurrence or metastases, and none died of renal carcinoma; 2 patients died of unrelated causes at 54 months and 82 months, respectively.
The first cases of CRCC in humans were described by Thoenes et al.3 in 1985. This variant of renal tumor is shown to be comprised of variable proportions of clear (“chromophobe”) cells and eosinophilic cells when stained with hematoxylin and eosin. The chromophobe cells contain opaque or finely reticular cytoplasm, and demonstrate deep staining with Hale colloidal iron. Electronmicroscopically, the cytoplasm contains microvesicular structures (150–300 nanometers), possibly derived from the endoplasmic reticulum or mitochondria.
The diffuse positivity for Hale colloidal iron stain in relation to the presence of cytoplasmic vesicles is one characteristic and distinguishing feature of CRCC; however, the differential diagnosis with renal oncocytoma can be challenging because this variant of tumor closely resembles CRCC. Moreover, renal oncocytoma occasionally is positive for Hale colloidal iron stain and in some cases contains cytoplasmic microvesicles.4, 5 Some authors therefore have investigated other methods with which to distinguish both tumors. Leroy et al.4 reported that immunohistochemical staining for cytokeratin 7 may be useful; in their study, all CRCC specimens showed strong cytoplasmic staining for cytokeratin 7, whereas the majority of oncocytomas were negative.
It is now acknowledged that CRCC and renal oncocytoma arise from the intercalated cells of the collecting duct system. Some cases of coexistent CRCC and oncocytoma (so-called renal oncocytosis) have been reported.5 These findings give strong support to a common origin. Conversely, chromosomal analyses of both tumors show different features. A low chromosome number (due to the loss of chromosomes 1, 2, 6, 10, 13, 17, and 21) is characteristic of CRCC.6 Recently, an oncogene (KIT) has been shown to be involved specifically in the development of CRCC.7 Renal oncocytoma shows different genetic characteristics, including combined loss of chromosomes Y and 1, rearrangements affecting band 11q12–13, involvement of 12q12–13, and loss of 14q.8 To our knowledge, the relation between both tumors has yet to be clarified, because some cases of hybrid tumors demonstrating CRCC features with components of oncocytoma have been reported.8
To our knowledge, only a few authors to date have studied outcome in patients with CRCC. In 1995, Crotty et al.1 reported a series of 50 patients treated for CRCC. With a mean follow-up of 6 years, the rates of specific mortality and recurrence were found to be only 2% and 6%, respectively. In this series, 2 patients developed early recurrence at 6 months and 12 months, respectively; another patient developed metastases 7 years after undergoing nephrectomy and died rapidly.
Recently, all patients who had undergone radical nephrectomy between 1970 and 2000 at the Mayo Clinic were reviewed retrospectively.2 In this series of 2385 patients, 102 (4.3%) originally were diagnosed with CRCC. The 5-year disease-specific survival rate in this group was 86.7%. Patients with clear cell renal carcinoma were found to have a poorer prognosis, with a 5-year disease-specific survival rate of 68.9% (P < 0.001). The authors found no difference with regard to specific survival between CRCC and papillary renal carcinoma. In this study, the TNM stage, tumor size, nuclear grade, presence of a sarcomatoid component, and tumor necrosis were reported to be associated significantly with a poor prognosis. The disease-specific survival rates with and without necrosis were 60.7% and 94%, respectively. In the current series, necrosis was found to be present in 36% of cases. However, this feature did not appear to have any prognostic impact.
In the opinion of most authors, CRCC has a favorable prognosis. In the current series, no patient experienced tumor recurrence or disease progression, and none died of renal carcinoma after a mean follow-up of 49.5 months. It is interesting to note that the tumors for the most part were staged as T1/T2 (96.7% of cases), and were of low grade (88.5% of cases). No patient was found to have lymph node involvement or metastases at the time of presentation; moreover, in no case did the CRCC contain a sarcomatoid component. The patient outcome in the current study most likely can be explained in terms of these favorable features; indeed, it has been widely accepted that TNM stage and nuclear grade are the most important prognostic factors in patients with renal carcinoma.9, 10
However, in some series, aggressive variants of CRCC have been reported. Grabowski et al.11 evaluated the survival of 42 patients with rare subtypes of renal carcinoma; of their 9 patients with CRCC, 2 (22.2%) died during follow-up. Renshaw et al.12 reported a series of 25 cases of CRCC, comprised of 23 cases of solitary tumors and 2 cases of coexistent papillary carcinoma. In that report, 7 patients (28%) developed metastases. Of these patients, 5 had a solitary CRCC measuring > 8 cm in dimension and developed liver metastases; the other 2 patients (those with coexistent papillary carcinoma) developed metastases in the lung but it was not known which tumor (CRCC or papillary carcinoma) metastasized.
Onishi et al.13 have suggested that the eosinophilic subtype of CRCC has a better prognosis than the typical (chromophobe) subtype. In a long-term series of 35 patients, comprised of 26 with the typical subtype and 9 patients with the eosinophilic subtype, these authors noted a different outcome for each variant; there were no disease-related deaths reported in patients with eosinophilic CRCC whereas, of those patients with the typical subtype, 5 (19.2%) died within 2 years of surgery. Of these 5 patients, 3 with high-stage disease were found to have varying degrees of sarcomatoid change. Conversely, the other 2 patients developed disease recurrence > 10 years after surgery (late recurrence). No sarcomatoid changes were noted in either of these patients, nor in any of the patients who later died from carcinoma. To our knowledge, the hypothesis that the eosinophilic subtype of CRCC has a better prognosis than the typical subtype has not been confirmed to date. In the current series, 34.4% of the tumors were classified as predominantly eosinophilic. These tumors were not found to be associated with a better prognosis.
Finally, a radical nephrectomy was performed in > 93% of cases. However, a partial nephrectomy would have been more appropriate in a large proportion of patients because most of them had small, low-grade tumors.
The results of the current study provide clear evidence that CRCC carries a favorable prognosis. However, other authors have reported that this variant of tumor may sometimes have an aggressive course. Larger studies are needed to elucidate the clinical behavior of CRCC further.
In our experience, CRCC carries an excellent prognosis, which may be explained in terms of the high rate of low-stage and low-grade tumors. However, larger series with a longer follow-up period are required to improve our understanding of this type of tumor.