- Email: [email protected]
Unclassified Renal Cell Carcinoma: Clinical Features and Prognostic Impact of a New Histological Subtype
0022-5347/02/1683-0950/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 168, 950 –955, September 2002 Printed in U.S.A.
DOI: 10.1097/01.ju.0000026869.95198.bb
UNCLASSIFIED RENAL CELL CARCINOMA: CLINICAL FEATURES AND PROGNOSTIC IMPACT OF A NEW HISTOLOGICAL SUBTYPE AMNON ZISMAN, DEBBY H. CHAO, ALLAN J. PANTUCK, HYUN J. KIM, JEFF A. WIEDER,* ROBERT A. FIGLIN, JONATHAN W. SAID AND ARIE S. BELLDEGRUN From the Divisions of Urologic Oncology, Hematology/Oncology and Anatomic Pathology, Departments of Biostatistics, Medicine and Urology, University of California-Los Angeles School of Medicine, Los Angeles, California
ABSTRACT
Purpose: We characterized the histopathological features and clinical behavior of unclassified renal cell carcinoma and compared the prognostic outcome in patients with unclassified and conventional (clear cell) renal cell carcinoma. Materials and Methods: A total of 31 patients with unclassified renal cell carcinoma are included in the kidney cancer database at our institution. Another 317 matched patients with clear cell carcinoma were used for comparing demographic, clinical, pathological and survival data. Results: The incidence of unclassified renal cell carcinoma was 2.9%. At initial diagnosis 29 patients (94%) with unclassified and 264 (83%) with clear cell renal cell carcinoma had metastatic disease (p ⫽ 0.143). Compared with the clear cell variety unclassified disease was associated with larger tumors (p ⫽ 0.005), increased risk of adrenal gland involvement (25% of cases, p ⫽ 0.0001), direct invasion to adjacent organs (42%, p ⫽ 0.00001), bone (52%, p ⫽ 0.022), regional (52%, p ⫽ 0.0042) and nonregional lymph node (41%, p ⫽ 0.03) metastases. Nephrectomy was less likely to be attempted or completed in unclassified renal cell carcinoma cases (61%, p ⫽ 0.00007). Unclassified histology was a significant indicator for poor prognosis on multivariate analysis (p ⬍0.0001). Median survival in patients with unclassified renal cell carcinoma was 4.3 months. Nephrectomy alone did not confer any survival advantage in these cases (p ⫽ 0.1086), while immunotherapy did (p ⫽ 0.008). The combination of nephrectomy and immunotherapy yielded improved survival over immunotherapy alone (p ⫽ 0.0356) but patients with unclassified renal cell carcinoma were significantly less likely than those with clear cell disease to be eligible for immunotherapy regimens (p ⫽ 0.05). Conclusions: Unclassified renal cell carcinoma is associated with distinct and highly aggressive biological behavior, and poor clinical outcome. Whenever feasible, immunotherapy with nephrectomy is warranted. KEY WORDS: kidney; carcinoma, renal cell; neoplasms by histologic type; immunotherapy; nephrectomy
In 1997 a multidisciplinary workshop on renal cell carcinoma held by the WHO resulted in an international agreement on the histological classification of renal cortical epithelial neoplasms.1 The 5 subtypes of renal cell carcinoma designated were conventional (clear cell), chromophil (papillary), chromophobe, collecting duct and unclassified. In some surgical series unclassified renal cell carcinoma comprised approximately 3% to 5% of cases.2 This classification includes morphologically distinct tumors that are generally high grade with unrecognizable cell types or pure sarcomatoid histology without typical renal cell epithelial components. To our knowledge there has been no study of the biology of unclassified renal cell carcinoma. Collecting duct carcinoma typically presents at advanced stages and is associated with an overall poor prognosis.3–5 Although sarcomatoid transformation is not considered a true subtype of renal cell carcinoma, it is also a strong indicator of poor outcome.6 –9 Prognostic comparisons of papillary and clear cell renal cell carcinoma have yielded discrepant results showing that papillary disease may be associated with a better10, 11 or worse12 prognosis, or have no prognostic differences7, 13 from clear cell renal cell carcinoma. There are also discrepant results in comparisons of chromophobe and
clear cell renal cell carcinoma.7, 11 In the literature on the biology of renal cell carcinoma scant attention has been directed toward unclassified epithelial tumors of the kidney. We defined the occurrence of unclassified renal cell carcinoma in the population of patients with kidney cancer treated at our institution, characterized the clinical behavior of unclassified renal cell carcinoma tumors and compared prognostic differences in patients with unclassified and clear cell renal cell carcinoma. PATIENTS AND METHODS
A total of 1,087 patients with renal cell carcinoma are included in the kidney cancer database at our institution, of whom the majority were treated in the context of the multidisciplinary kidney cancer program. Between July 1989 and June 2000, 31 patients (2.9%) were diagnosed with unclassified renal cell carcinoma and 814 had tumors with clear cell histology. Bilateral cases were excluded from study due to possible confounding factors and the special modeling required, which resulted in the exclusion of 51 of bilateral clear cell renal cell carcinoma. Clear cell renal cell carcinoma of mixed histological types was also excluded from study, including coexistent papillary disease in 67 patients, sarcomatoid disease in 37, chromophobe disease in 3 and undifferentiated features in 3, resulting in 653 with pure clear cell renal cell carcinoma.
Accepted for publication April 5, 2002. * Financial interest in or other relationship with TAP Pharmaceuticals and Vical. 950
951
UNCLASSIFIED RENAL CELL CARCINOMA
Demographic, clinical and pathological data were obtained from patient medical records as part of an institutional review board approved study (Protocol 99-233). Incidental diagnosis was defined as being completely asymptomatic at diagnosis or having symptoms attributable to nonrenal cell carcinoma related disease. Performance status using the criteria established by the Eastern Cooperative Oncology Group (ECOG) was assigned to each patient at initial diagnosis. Radiographic, operative and pathology reports were used to assess the extent of disease, including local involvement and distant metastasis. The histological classification and Fuhrman nuclear grade14 in all patients included in the kidney cancer database were confirmed by a small group of experienced pathologists at our institution. In all nonnephrectomy cases biopsy was done for definitive tissue diagnosis. Renal cell carcinoma was categorized as unclassified when it did not meet histological criteria for the recognized variations of established renal cell carcinoma entities. Many cases had features of undifferentiated carcinoma with pleomorphic or anaplastic nuclear morphology. In addition, malignant tumors with pure sarcomatoid or mixed sarcomatous and anaplastic features were also designated unclassified renal cell carcinoma, as proposed by the 1997 consensus statement of the Union Internationale Contre le Cancer and American Joint Commission on Cancer (fig. 1).1 When the diagnosis of renal cell carcinoma was in doubt, immunohistochemistry was performed to confirm renal epithelial origin and exclude true sarcoma of the kidney. Tumors that stained for vimentin and cytokeratins, and were negative for chromogranin, HMB45, desmin and muscle specific actins were considered to be of epithelial origin. Tumors that did not meet these criteria were excluded from study. Patients were staged according to the Union Internationale Contre le Cancer and American Joint Committee on Cancer 1997 TNM staging system,15 and University of California-Los Angeles Integrated Staging System (UISS).16 The latter classification is a clinically relevant staging system that was recently validated17 and predicts patient survival based on the integration of 1997 TNM stage, Fuhrman grade and ECOG performance status, and stratifies survival into 5 groups. According to this system projected 2 and 5-year survival is categorized as I—96% and 94%, II— 89% and 67%, III— 66% and 39%, IV— 42% and 23%, and V—9% and 0%, respectively. Incorporating UISS into our analysis made it possible to account for the complex interactions that occur among stage, grade and performance status, and allows the analysis of additional variables that may have significant prognostic impact on this patient population. All patients underwent radical or partial nephrectomy. Most of those with metastases, including all with unclassified renal cell carcinoma who received immunotherapy, were treated with recombinant interleukin (IL)-2 based immunotherapy regimens within the framework of 11 clinical trials. Due to the rarity of this tumor patients receiving immunotherapy were considered 1 group because as there would be an insufficient number to comment on individual immunotherapy regimens. Patients with measurable and evaluable disease underwent clinical measurements of indicator le-
sions, as indicated by standard practice, to establish criteria for the response to immunotherapy according to certain definitions. Complete response was defined as the complete disappearance of all measurable and evaluable disease with no new lesions. Partial response was defined as a 50% or greater decrease below baseline in the sum of the products of the perpendicular diameters of all measurable lesions with no new lesions. Progressive disease was defined as the lesser of a 25% increase or an increase of 10 cm.2 in the sum of the products of measurable lesions of greater than the least sum observed (greater than baseline if there was no decrease), clear worsening of any evaluable disease or any new lesion/ site. Patients with stable disease did not qualify for complete or partial response, or progressive disease. All patients with unclassified renal cell carcinoma had 1997 TNM stages 3 and 4 disease, and were grouped into UISS III and V. Therefore, the control group comprised 317 patients with clear cell renal cell carcinoma with 1997 TNM stages 3 and 4, and UISS III and V disease. Patients were grouped for analysis according to histological tumor type and whether nephrectomy was successfully completed. Comparisons of patient age, sex, family history of cancer, the percent of incidental and nonincidental diagnoses, performance status at presentation, disease stage and various tumor characteristics were made of those with clear cell and unclassified renal cell carcinoma in each group. In patients who underwent nephrectomy survival was calculated from the date of nephrectomy to the date of the last followup or death. In those who did not undergo nephrectomy survival was calculated from the first date of a contact specific for renal cell carcinoma at our institution to the date of the last followup or death. The 3 patients with unclassified renal cell carcinoma with missing grade (2) and ECOG performance status (1) data were presumptively assigned to UISS IV (2) and V (1) based on available clinical and pathological data. Of the 349 patients with 1997 TNM stages 3 and 4 clear cell renal cell carcinoma in the database 32 were excluded from study due to missing grade and ECOG performance status, while 20 with missing grade (19) or ECOG performance status (1) data were included in analysis. The latter group of 20 patients was presumptively assigned to UISS IV based on available clinical and pathological data. Survival data was missing for on 1 patient with clear cell disease. To determine variables significant for survival univariate analysis was performed in the combined population of patients with unclassified and clear cell renal cell carcinoma. The impact of patient age, sex, unclassified renal cell carcinoma histology, tumor size, nephrectomy, regional lymph node involvement, distant metastasis, UISS stage and immunotherapy were analyzed using the Cox proportional hazards regression model at a level of 0.05. To determine confounding factors variables significant on univariate analysis were subjected to backward stepwise multivariate analysis with entry and exclusion criteria of 0.1 and 0.15, respectively. Survival results were also assessed by the product-time method of Kaplan-Meier. Comparisons of groups were performed by the log rank method to assess the significance of the Kaplan-Meier curves. Results were based on the assumption check of the Cox regression model. Statistical analyses were performed using commercially available computer software. RESULTS
FIG. 1. Various histological appearances of unclassified renal cell carcinoma. a, pure anaplastic histology. b, mixed anaplastic and sarcomatoid histology. c, pure sarcomatoid histology. H&E, reduced from ⫻200.
The incidence of unclassified renal cell carcinoma in our patient population was 2.8%. Table 1 lists the characteristics of patients with unclassified and clear cell renal cell carcinoma. There was no history of von Hippel Lindau disease or other familial cancer syndromes. Patients with unclassified and clear cell disease did not differ in regard to a family history positive for cancer (14 or 45% versus 112 or 35%,
952
UNCLASSIFIED RENAL CELL CARCINOMA TABLE 1. Demographic, clinical and staging data on patients with unclassified and clear cell renal cell carcinoma Unclassified
No. pts. % Male Mean age ⫾ SD (range) No. family ca history (%) No. incidental diagnosis (%): Asymptomatic Symptomatic No. ECOG performance status (%): 0 1 2 3 X No. 1997 TNM stage (%): 3 4 No. UISS (%): 3 4 5 No. N0M0 (%) No. N0M⫹ (%) No. N⫹M0 (%) No. N⫹M⫹ (%)
Clear Cell
Nephrectomy Completed
Nephrectomy Not Attempted or Completed
Nephrectomy Completed
Nephrectomy Not Attempted or Completed
19 63 55.3 ⫾ 13.9 (25–81) 10 (53) 3 (16) 2 (11) 1 (5)
12 75 55.4 ⫾ 15.2 (27–81) 4 (33) 1 (8) 0 1 (8)
286 67 58.9 ⫾ 11.0 (25–89) 95 (33) 55 (19) 24 (8) 31 (11)
31 68 60.3 ⫾ 11.3 (30–80) 17 (55) 2 (6) 1 (3) 1 (3)
2 (11) 12 (63) 4 (21) 0 1 (5)
0 9 (75) 2 (17) 1 (8) 0
47 (16) 227 (79) 10 (3) 1 (0.3) 1 (0.3)
2 (6) 24 (77) 5 (16) 0 0
3 (16) 16 (84)
0 12 (100)
56 (20) 230 (80)
0 31 (100)
79 203 4 53 160 17 56
1 (3) 30 (97) 0 0 18 (58) 1 (3) 12 (39)
3 (16) 11 (58) 5 (26) 2 (11) 11 (58) 1 (5) 5 (26)
0 9 3 0 2 1 9
p ⫽ 0.28) or patient history of nonrenal malignancy (3 or 10% versus 23 or 7%, p ⫽ 0.0624). The rate of detecting an incidental renal mass also did not differ in the 2 groups (4 or 21% versus 57 or 18%, p ⫽ 0.64). No symptomatic patients presented with the classic triad. At diagnosis 29 patients (94%) with unclassified and 264 (83%) with clear cell renal cell carcinoma had regional lymph node or distant metastases (p ⫽ 0.143). All patients with unclassified and clear cell renal cell carcinoma who did not undergo attempted or completed nephrectomy presented with regional lymph node or distant metastases at initial diagnosis. As defined by 1997 TNM staging criteria, regional lymph nodes were involved in 16 unclassified (52%) and 86 clear cell (27%) cases (p ⫽ 0.0042). Of the 27 unclassified and 246 clear cell cases that presented with distant metastatic disease, the lung (24 or 89% versus 199 or 81%, p ⫽ 0.45), bone (14 or 52% versus 74 or 30%, p ⫽ 0.022) and nonregional lymph nodes (11 or 41% versus 54 or 22%, p ⫽ 0.03, respectively) were most commonly involved. Patients with unclassified disease were also 20 times more likely than those with
(75) (25) (17) (8) (75)
(28) (71) (1) (19) (56) (6) (20)
clear cell disease to be grouped as UISS V (8 or 26% versus 4 or 1.3%, p ⫽ 0.00001). Nephrectomy was completed in 19 (61%) and 286 (90%), attempted but unresectable in 3 (10%) and 5 (1.6%), and not attempted in 9 (29%) and 26 (8%) patients, with unclassified and clear cell disease, respectively (p ⫽ 0.00007). Table 2 lists grade and stage data on patients who did and did not successfully undergo nephrectomy. Average tumor size in patients who did not undergo nephrectomy was slightly greater than in those in whom it was completed in the 2 histological groups of unclassified and clear cell renal cell carcinoma (p ⫽ 0.757 and 0.812, respectively). Regardless of whether nephrectomy was completed average unclassified tumor size plus or minus standard deviation was greater than that of clear cell renal cell carcinoma tumors (10.8 ⫾ 3.9 cm., range 3.5 to 18.0 versus 8.8 ⫾ 3.6, range 1.8 to 22.0, p ⫽ 0.0054). Compared with clear cell cases unclassified cases were more likely to have tumor involvement of the adrenal glands (8 or 25% versus 13 or 4%, p ⫽ 0.0001) and local tumor extension to adjacent organs (13 or 42%
TABLE 2. Tumor characteristics in patients with unclassified and clear cell renal cell carcinoma Unclassified Nephrectomy Completed No. pts. No. rt. tumor (%) Mean cm. tumor size ⫾ SD (range) No. Fuhrman grade (%): 1 2 3 4 Missing No. adrenal involvement (%) No. renal vein extension (%) No. inferior vena cava extension (%) No. adjacent organ extension (%): Psoas muscle Liver Bowel Diaphragm Inferior vena cava Abdominal/chest wall
19
Clear Cell
Nephrectomy Not Attempted or Completed 9/3
10 (53) 10.6 ⫾ 3.4 (4.5–17.0)
6 (50) 11.0 ⫾ 4.8 (3.5–18.0)
0 0 9 (47) 10 (53) 0 4 (21) 3 (16) 2 (11)
0 0 8 (67) 2 (17) 2 (17) 4 (33) 2 (17) 3 (25)
1 (5) 1 (5) 1 (5) 2 (11) 3 (16) 0
1 (8) 0 3 (25) 0 0 1 (8)
Nephrectomy Completed 286 163 (57) 8.8 ⫾ 3.6 (3.0–22.0) 6 125 142 6 7 7 55 54
Nephrectomy Not Attempted or Completed 26 Not attempted, 5 not completed 15 (50) 9.0 ⫾ 3.5 (4.0–19.0)
(2) (44) (50) (2) (2) (2) (19) (19)
2 (6) 7 (23) 10 (32) 0 12 (39) 6 (19) 2 (6) 5 (16)
2 (0.7) 1 (0.3) 1 (0.3) 0 0 0
3 (10) 2 (6) 1 (3) 1 (3) 1 (3) 1 (3)
953
UNCLASSIFIED RENAL CELL CARCINOMA
versus 25 or 8%, p ⫽ 0.00001). However, the incidence of tumor extension into the renal vein or inferior vena cava was similar. Of the patients who underwent attempted or completed nephrectomy 2 of the 22 with unclassified (9%) and 4 of the 291 with clear cell (1.4%) disease died of perioperative and postoperative complications (Yates correction p ⫽ 0.08). The patients with unclassified disease died of sepsis and multiorgan failure on postoperative day 7 and of cardiac arrest perioperatively, respectively. The 4 patients with clear cell disease died of sepsis and multi-organ failure on postoperative day 13, coagulopathy on postoperative day 1, inferior vena caval tumor emboli to the atrium and pulmonary artery intraoperatively, and pulmonary embolism on postoperative day 12, respectively. Major complications developed postoperatively in 5 cases of unclassified (23%) and 52 of clear cell (17%) disease in which surgery was done for kidney cancer (p ⫽ 0.59). Univariate analysis of the combined unclassified and clear cell renal cell carcinoma groups indicated that unclassified histology (p ⬍0.0001), UISS (p ⬍0.0001), completed nephrectomy (p ⬍0.0001), regional lymph node disease (p ⬍0.0001), tumor size (p ⫽ 0.0148) and distant metastasis (p ⫽ 0.0209) were significantly associated with patient survival, while immunotherapy approached significance (p ⫽ 0.0912). Patient age (p ⫽ 0.3604) and sex (p ⫽ 0.4429) were not significant factors. On multivariate analysis unclassified histology (p ⬍0.0001) and UISS (p ⫽ 0.0001) remained highly significant factors, whereas completed nephrectomy (p ⫽ 0.0004) and immunotherapy (p ⫽ 0.0015) were protective (table 3). Because unclassified renal cell carcinoma and UISS were the most significant prognostic factors, survival in patients with unclassified and clear cell disease within a UISS group was compared. We compared those with UISS IV disease because most patients with the 2 types of renal cell carcinoma were categorized in this group. Figure 2 shows that when matched for UISS IV, patients with unclassified disease had significantly lower survival compared with those with clear cell disease (median 3.9 versus 17.2 months, p ⬍0.0001). This finding indicates that unclassified histology had a detrimental impact beyond the UISS categories, which groups cases of similar stage, tumor grade and performance status. A comparison of patients with distant metastasis also indicated significantly lower survival for unclassified versus clear cell disease (median 3.8 versus 19.9 months, p ⬍0.0001). Overall median survival in patients with unclassified renal cell carcinoma was 4.3 months. There was no significance difference in survival in those who did and did not undergo completed nephrectomy (median 4.7 and 3.9 months, respectively, p ⫽ 0.1086). This result suggests that the protective effect of nephrectomy on multivariate analysis was only attributable to its effect on those with clear cell renal cell carcinoma. Patients with unclassified disease who received immunotherapy had significantly improved survival compared with those who did not (median 11.1 and 2.9 months, respectively, p ⫽ 0.008). Of the patients with metastatic disease at initial diagnosis, 14 (52%) and 173 (70%) with unclassified and clear cell disease received immunotherapy (p ⫽ 0.05), including 6 and 17 in whom the primary tumor was not removed and 8 and 156 who underwent completed nephrectomy, respectively. All 14 patients with unclassified
FIG. 2. Kaplan-Meier comparison of survival in patients with UISS IV unclassified and clear cell renal cell carcinoma (RCC) shows significant survival advantage for clear cell variety.
renal cell carcinoma treated with immunotherapy received IL-2 based immunotherapy, including 6 who received high dose IL-2 in a nonintensive care setting. There were no complete responses, 1 patient achieved a partial response, 3 had stable disease and 10 had progressive disease after treatment. Patients with clear cell renal cell carcinoma who received immunotherapy had improved survival compared with patients with unclassified disease whether nephrectomy was (p ⫽ 0.0042) or was not (p ⫽ 0.0479) performed. Treating clear cell renal cell carcinoma with combined nephrectomy and immunotherapy conferred a survival advantage over immunotherapy with the primary tumor still in place (median 25.1 versus 10.9 months, p ⫽ 0.0007). The advantage for unclassified renal cell carcinoma was less dramatic (12.9 versus 7.7 months, p ⫽ 0.0356). At diagnosis there were 2 cases (7%) of stage N0M0 unclassified renal cell carcinoma. One patient who was symptomatic at initial diagnosis underwent nephrectomy for a grade 4 stage pT3b (UISS III) tumor. Local recurrence developed 3 months after nephrectomy. After disease recurrence he was treated with a course of high dose IL-2. Disease progressed despite immunotherapy and the patient died 7 months postoperatively. The other patient was asymptomatic at diagnosis and underwent nephrectomy for a grade 4 stage pT3a (UISS III) pure sarcomatoid tumor. He had no evidence of disease 18.5 months postoperatively. All other patients with unclassified renal cell carcinoma died of cancer. DISCUSSION
This study indicates that unclassified renal cell carcinoma is an uncommon variant of renal cell carcinoma (2.8% of cases) that typically presents with significantly larger, more aggressive tumors and is associated with poor clinical outcomes. Only 10% of these patients are asymptomatic at initial diagnosis, while greater than 90% present with regional lymph node or distant metastases. Compared with clear cell renal cell carcinoma there is significantly increased propensity for unclassified disease to involve local tumor invasion of the adrenal gland and adjacent organs as well as metastatic
TABLE 3. Variables with a significant impact on the survival of patients with unclassified renal cell carcinoma on multivariate analysis
Unclassified renal cell Ca UISS Tumor size Nephrectomy completed Immunotherapy
Standard Estimate ⫾ Error
Chi-Square
p Value
Hazards Ratio (95% CI)
1.003 ⫾ 0.230 0.601 ⫾ 0.158 0.033 ⫾ 0.019 ⫺0.753 ⫾ 0.214 ⫺0.438 ⫾ 0.151
18.959 14.468 2.814 12.355 8.386
⬍0.0001 0.0001 0.0934 0.0004 0.0038
2.726 (1.736–4.281) 1.824 (1.338–2.487) 1.034 (0.994–1.074) 0.471 (0.309–0.717) 0.645 (0.480–0.868)
954
UNCLASSIFIED RENAL CELL CARCINOMA
involvement of the bone and regional/nonregional lymph nodes. Metastatic unclassified renal cell carcinoma confers a significant survival disadvantage compared with metastatic clear cell renal cell carcinoma, as also illustrated by the survival difference in the more homogenous UISS IV population (p ⬍0.0001). On multivariate analysis unclassified histology was a significant indicator of poor prognosis. Only 1 patient with unclassified disease (pure sarcomatoid tumor) achieved prolonged disease-free survival. Long-term disease-free survival for sarcomatoid renal cell carcinoma is uncommon but it has previously been reported.18, 19 The lack of clinical symptoms in this patient is indicative of a relatively early diagnosis and may explain his improved survival. Overall such a presentation is rare among patients with unclassified renal cell carcinoma. The dismal median survival of 4.3 months in these cases was not significantly impacted by surgical extirpation. While nephrectomy was completed in 90% of stage 3 and 4 clear cell renal cell carcinoma cases, it was completed in only 60% of the unclassified population. This finding reflects the unique biological characteristics of the tumor and the advanced state of disease at presentation. Although unclassified renal cell carcinoma comprised just 2.8% of the kidney cancer database population, the 14% rate of unresectability exceeded it 5-fold. In the stages 3 and 4 clear cell cohort the 29% prevalence in the database paralleled the 23% rate of unresectability. Patients with unclassified renal cell carcinoma who undergo surgery tend to have higher morbidity as well as higher mortality. The poor prognosis associated with unclassified renal cell carcinoma can be improved by administering immunotherapy. However, patients with unclassified disease were significantly more likely than those with clear cell disease to have contraindications against immunotherapy, which reflects the poor clinical status of the former as well as rapidly progressive disease. When immunotherapy was administered in these cases, it was beneficial when nephrectomy was performed and when the primary tumor remained in place. Nephrectomy alone did not improve survival in the unclassified group, whereas combined nephrectomy and immunotherapy conferred a survival advantage over immunotherapy alone. The previous study of Litwin et al shows that this prolonged survival can be accompanied by significant quality of life in patients with metastasis who undergo nephrectomy and IL-2 based immunotherapy according to cross-sectional assessment with health related quality of life instruments.20 Compared with reference populations with chronic medical conditions or nonrenal malignancies patients with metastatic renal cell carcinoma have transient impairment in quality of life during immunotherapy. However, after immunotherapy those with renal cell carcinoma have better quality of life estimates than those with other malignancies as well as better physical function than those with congestive heart failure, suggesting that patients with metastatic renal cell carcinoma may have well preserved quality of life when nephrectomy is followed by immunotherapy. Despite this promising data the benefits of combined nephrectomy and immunotherapy for unclassified renal cell carcinoma is modest compared with clear cell renal cell carcinoma. A limitation of this study is the possible bias resulting from patients with the primary tumor in place, in whom the classification of unclassified renal cell carcinoma was based only on the biopsy specimen. Insufficient sampling or the poor quality of tumor samples may have impacted the histological evaluation. Therefore, it is impossible to be certain that some renal cell carcinoma cases with the primary tumor in place and classifiable histology were not inadvertently incorporated into the unclassified population. Other limitations of this study are its retrospective nature, the fact that treatments were not randomly assigned and the fact that there was no untreated control arm for comparison. However, the
contribution of and improvement in survival related to immunotherapy21 and the benefit of nephrectomy in the context of multimodal treatment using cytoreductive surgery and immunotherapy22 have been supported by other randomized controlled studies of metastatic renal cell carcinoma. Renal tumors of unclassified histology show highly aggressive biological behavior, have a tendency toward metastatic involvement to the bones and lymph nodes, and are associated with a poor clinical outcome in the majority of affected patients. Nephrectomy combined with immunotherapy confers some survival advantage. We advocate that when a histological diagnosis of unclassified renal cell carcinoma is made preoperatively, patients should be advised to undergo nephrectomy only when they have no contraindications for immunotherapy. This recommendation is due to the lack of survival advantage attained by nephrectomy alone, and the high associated surgical morbidity and mortality in this patient population. Because of the aggressive nature of these tumors, immunotherapy should also be considered preoperatively, especially when surgery must be delayed. A question that remains to be answered is whether adjuvant immunotherapy should be routinely administered in patients with stage N0M0 unclassified renal cell carcinoma. Due to the small study population and the aggressive nature of this tumor such an issue is unlikely to be addressed adequately by the experience at a single institution. In conclusion, a well coordinated nationwide effort similar to Wilms tumor studies is warranted to evaluate prospectively the merits of adjuvant immunotherapy in patients with stage N0M0 renal cell carcinoma who are at high risk for disease recurrence and progression, such as those with unclassified, collecting duct and sarcomatoid renal cell carcinoma. REFERENCES
1. Storkel, S., Eble, J. N., Adlakha, K., Amin, M., Blute, M. L., Bostwick, D. G. et al: Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 987, 1997 2. Kovacs, G., Akhtar, M., Beckwith, B. J., Bugert, P., Cooper, C. S., Delahunt, B. et al: The Heidelberg classification of renal cell tumours. J Pathol, 183: 131, 1997 3. Chao, D., Zisman, A., Pantuck, A. J., Gitlitz, B. J., Freedland, S. J., Said, J. W. et al: Collecting duct renal cell carcinoma: clinical study of a rare tumor. J Urol, 167: 71, 2002 4. Dimopoulos, M. A., Logothetis, C. J., Markowitz, A., Sella, A., Amato, R. and Ro, J.: Collecting duct carcinoma of the kidney. Br J Urol, 71: 388, 1993 5. Schoenberg, M., Cairns, P., Brooks, J. D., Marshall, F. F., Epstein, J. I., Isaacs, W. B. et al: Frequent loss of chromosome arms 8p and 13q in collecting duct carcinoma (CDC) of the kidney. Genes Chromosomes Cancer, 12: 76, 1995 6. Mani, S., Todd, M. B., Katz, K. and Poo, W. J.: Prognostic factors for survival in patients with metastatic renal cancer treated with biological response modifiers. J Urol, 154: 35, 1995 7. Moch, H., Gasser, T., Amin, M. B., Torhorst, J., Sauter, G. and Mihatsch, M. J.: Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer, 89: 604, 2000 8. Delahunt, B. and Nacey, J. N.: Renal cell carcinoma. II. Histological indicators of prognosis. Pathology, 19: 258, 1987 9. Cangiano, T., Liao, J., Naitoh, J., Dorey, F., Figlin, R. and Belldegrun, A.: Sarcomatoid renal cell carcinoma: biologic behavior, prognosis, and response to combined surgical resection and immunotherapy. J Clin Oncol, 17: 523, 1999 10. Amin, M. B., Corless, C. L., Renshaw, A. A., Tickoo, S. K., Kubus, J. and Schultz, D. S.: Papillary (chromophil) renal cell carcinoma: histomorphologic characteristics and evaluation of conventional pathologic prognostic parameters in 62 cases. Am J Surg Pathol, 21: 621, 1997 11. Fromowitz, F. B. and Bard, R. H.: Clinical implications of pathologic subtypes in renal cell carcinoma. Semin Urol, 8: 31, 1990 12. Medeiros, L. J., Gelb, A. B. and Weiss, L. M.: Renal cell carci-
UNCLASSIFIED RENAL CELL CARCINOMA
13.
14. 15.
16.
17.
noma. Prognostic significance of morphologic parameters in 121 cases. Cancer, 61: 1639, 1988 Giberti, C., Oneto, F., Martorana, G., Rovida, S. and Carmignani, G.: Radical nephrectomy for renal cell carcinoma: long-term results and prognostic factors on a series of 328 cases. Eur Urol, 31: 40, 1997 Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 Guinan, P., Sobin, L. H., Algaba, F., Badellino, F., Kameyama, S., MacLennan, G. et al: TNM staging of renal cell carcinoma: Workgroup No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 992, 1997 Zisman, A., Pantuck, A. J., Dorey, F., Said, J. W., Schvarts, O., Quintana, D. et al: Improved prognostication of renal cell carcinoma using an integrated staging system. J Clin Oncol, 19: 1649, 2001 Zisman, A., Pantuck, A. J., Figlin, R. A. and Belldegrun, A. S.:
18. 19. 20.
21.
22.
955
Validation of the UCLA integrated staging system for patients with renal cell carcinoma. J Clin Oncol, 19: 3792, 2001 Ro, J. Y., Ayala, A. G., Sella, A., Samuels, M. L. and Swanson, D. A.: Sarcomatoid renal cell carcinoma: clinicopathologic. A study of 42 cases. Cancer, 59: 516, 1987 Bertoni, F., Ferri, C., Benati, A., Bacchini, P. and Corrado, F.: Sarcomatoid carcinoma of the kidney. J Urol, 137: 25, 1987 Litwin, M. S., Fine, J. T., Dorey, F., Figlin, R. A. and Belldegrun, A. S.: Health related quality of life outcomes in patients treated for metastatic kidney cancer: a pilot study. J Urol, 157: 1608, 1997 Interferon-alpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Medical Research Council Renal Cancer Collaborators. Lancet, 353: 14, 1999 Flanigan, R. C., Salmon, S. E., Blumenstein, B. A., Bearman, S. I., Roy, V., McGrath, P. C. et al: Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med, 345: 1655, 2001
Vol. 168, 950 –955, September 2002 Printed in U.S.A.
DOI: 10.1097/01.ju.0000026869.95198.bb
UNCLASSIFIED RENAL CELL CARCINOMA: CLINICAL FEATURES AND PROGNOSTIC IMPACT OF A NEW HISTOLOGICAL SUBTYPE AMNON ZISMAN, DEBBY H. CHAO, ALLAN J. PANTUCK, HYUN J. KIM, JEFF A. WIEDER,* ROBERT A. FIGLIN, JONATHAN W. SAID AND ARIE S. BELLDEGRUN From the Divisions of Urologic Oncology, Hematology/Oncology and Anatomic Pathology, Departments of Biostatistics, Medicine and Urology, University of California-Los Angeles School of Medicine, Los Angeles, California
ABSTRACT
Purpose: We characterized the histopathological features and clinical behavior of unclassified renal cell carcinoma and compared the prognostic outcome in patients with unclassified and conventional (clear cell) renal cell carcinoma. Materials and Methods: A total of 31 patients with unclassified renal cell carcinoma are included in the kidney cancer database at our institution. Another 317 matched patients with clear cell carcinoma were used for comparing demographic, clinical, pathological and survival data. Results: The incidence of unclassified renal cell carcinoma was 2.9%. At initial diagnosis 29 patients (94%) with unclassified and 264 (83%) with clear cell renal cell carcinoma had metastatic disease (p ⫽ 0.143). Compared with the clear cell variety unclassified disease was associated with larger tumors (p ⫽ 0.005), increased risk of adrenal gland involvement (25% of cases, p ⫽ 0.0001), direct invasion to adjacent organs (42%, p ⫽ 0.00001), bone (52%, p ⫽ 0.022), regional (52%, p ⫽ 0.0042) and nonregional lymph node (41%, p ⫽ 0.03) metastases. Nephrectomy was less likely to be attempted or completed in unclassified renal cell carcinoma cases (61%, p ⫽ 0.00007). Unclassified histology was a significant indicator for poor prognosis on multivariate analysis (p ⬍0.0001). Median survival in patients with unclassified renal cell carcinoma was 4.3 months. Nephrectomy alone did not confer any survival advantage in these cases (p ⫽ 0.1086), while immunotherapy did (p ⫽ 0.008). The combination of nephrectomy and immunotherapy yielded improved survival over immunotherapy alone (p ⫽ 0.0356) but patients with unclassified renal cell carcinoma were significantly less likely than those with clear cell disease to be eligible for immunotherapy regimens (p ⫽ 0.05). Conclusions: Unclassified renal cell carcinoma is associated with distinct and highly aggressive biological behavior, and poor clinical outcome. Whenever feasible, immunotherapy with nephrectomy is warranted. KEY WORDS: kidney; carcinoma, renal cell; neoplasms by histologic type; immunotherapy; nephrectomy
In 1997 a multidisciplinary workshop on renal cell carcinoma held by the WHO resulted in an international agreement on the histological classification of renal cortical epithelial neoplasms.1 The 5 subtypes of renal cell carcinoma designated were conventional (clear cell), chromophil (papillary), chromophobe, collecting duct and unclassified. In some surgical series unclassified renal cell carcinoma comprised approximately 3% to 5% of cases.2 This classification includes morphologically distinct tumors that are generally high grade with unrecognizable cell types or pure sarcomatoid histology without typical renal cell epithelial components. To our knowledge there has been no study of the biology of unclassified renal cell carcinoma. Collecting duct carcinoma typically presents at advanced stages and is associated with an overall poor prognosis.3–5 Although sarcomatoid transformation is not considered a true subtype of renal cell carcinoma, it is also a strong indicator of poor outcome.6 –9 Prognostic comparisons of papillary and clear cell renal cell carcinoma have yielded discrepant results showing that papillary disease may be associated with a better10, 11 or worse12 prognosis, or have no prognostic differences7, 13 from clear cell renal cell carcinoma. There are also discrepant results in comparisons of chromophobe and
clear cell renal cell carcinoma.7, 11 In the literature on the biology of renal cell carcinoma scant attention has been directed toward unclassified epithelial tumors of the kidney. We defined the occurrence of unclassified renal cell carcinoma in the population of patients with kidney cancer treated at our institution, characterized the clinical behavior of unclassified renal cell carcinoma tumors and compared prognostic differences in patients with unclassified and clear cell renal cell carcinoma. PATIENTS AND METHODS
A total of 1,087 patients with renal cell carcinoma are included in the kidney cancer database at our institution, of whom the majority were treated in the context of the multidisciplinary kidney cancer program. Between July 1989 and June 2000, 31 patients (2.9%) were diagnosed with unclassified renal cell carcinoma and 814 had tumors with clear cell histology. Bilateral cases were excluded from study due to possible confounding factors and the special modeling required, which resulted in the exclusion of 51 of bilateral clear cell renal cell carcinoma. Clear cell renal cell carcinoma of mixed histological types was also excluded from study, including coexistent papillary disease in 67 patients, sarcomatoid disease in 37, chromophobe disease in 3 and undifferentiated features in 3, resulting in 653 with pure clear cell renal cell carcinoma.
Accepted for publication April 5, 2002. * Financial interest in or other relationship with TAP Pharmaceuticals and Vical. 950
951
UNCLASSIFIED RENAL CELL CARCINOMA
Demographic, clinical and pathological data were obtained from patient medical records as part of an institutional review board approved study (Protocol 99-233). Incidental diagnosis was defined as being completely asymptomatic at diagnosis or having symptoms attributable to nonrenal cell carcinoma related disease. Performance status using the criteria established by the Eastern Cooperative Oncology Group (ECOG) was assigned to each patient at initial diagnosis. Radiographic, operative and pathology reports were used to assess the extent of disease, including local involvement and distant metastasis. The histological classification and Fuhrman nuclear grade14 in all patients included in the kidney cancer database were confirmed by a small group of experienced pathologists at our institution. In all nonnephrectomy cases biopsy was done for definitive tissue diagnosis. Renal cell carcinoma was categorized as unclassified when it did not meet histological criteria for the recognized variations of established renal cell carcinoma entities. Many cases had features of undifferentiated carcinoma with pleomorphic or anaplastic nuclear morphology. In addition, malignant tumors with pure sarcomatoid or mixed sarcomatous and anaplastic features were also designated unclassified renal cell carcinoma, as proposed by the 1997 consensus statement of the Union Internationale Contre le Cancer and American Joint Commission on Cancer (fig. 1).1 When the diagnosis of renal cell carcinoma was in doubt, immunohistochemistry was performed to confirm renal epithelial origin and exclude true sarcoma of the kidney. Tumors that stained for vimentin and cytokeratins, and were negative for chromogranin, HMB45, desmin and muscle specific actins were considered to be of epithelial origin. Tumors that did not meet these criteria were excluded from study. Patients were staged according to the Union Internationale Contre le Cancer and American Joint Committee on Cancer 1997 TNM staging system,15 and University of California-Los Angeles Integrated Staging System (UISS).16 The latter classification is a clinically relevant staging system that was recently validated17 and predicts patient survival based on the integration of 1997 TNM stage, Fuhrman grade and ECOG performance status, and stratifies survival into 5 groups. According to this system projected 2 and 5-year survival is categorized as I—96% and 94%, II— 89% and 67%, III— 66% and 39%, IV— 42% and 23%, and V—9% and 0%, respectively. Incorporating UISS into our analysis made it possible to account for the complex interactions that occur among stage, grade and performance status, and allows the analysis of additional variables that may have significant prognostic impact on this patient population. All patients underwent radical or partial nephrectomy. Most of those with metastases, including all with unclassified renal cell carcinoma who received immunotherapy, were treated with recombinant interleukin (IL)-2 based immunotherapy regimens within the framework of 11 clinical trials. Due to the rarity of this tumor patients receiving immunotherapy were considered 1 group because as there would be an insufficient number to comment on individual immunotherapy regimens. Patients with measurable and evaluable disease underwent clinical measurements of indicator le-
sions, as indicated by standard practice, to establish criteria for the response to immunotherapy according to certain definitions. Complete response was defined as the complete disappearance of all measurable and evaluable disease with no new lesions. Partial response was defined as a 50% or greater decrease below baseline in the sum of the products of the perpendicular diameters of all measurable lesions with no new lesions. Progressive disease was defined as the lesser of a 25% increase or an increase of 10 cm.2 in the sum of the products of measurable lesions of greater than the least sum observed (greater than baseline if there was no decrease), clear worsening of any evaluable disease or any new lesion/ site. Patients with stable disease did not qualify for complete or partial response, or progressive disease. All patients with unclassified renal cell carcinoma had 1997 TNM stages 3 and 4 disease, and were grouped into UISS III and V. Therefore, the control group comprised 317 patients with clear cell renal cell carcinoma with 1997 TNM stages 3 and 4, and UISS III and V disease. Patients were grouped for analysis according to histological tumor type and whether nephrectomy was successfully completed. Comparisons of patient age, sex, family history of cancer, the percent of incidental and nonincidental diagnoses, performance status at presentation, disease stage and various tumor characteristics were made of those with clear cell and unclassified renal cell carcinoma in each group. In patients who underwent nephrectomy survival was calculated from the date of nephrectomy to the date of the last followup or death. In those who did not undergo nephrectomy survival was calculated from the first date of a contact specific for renal cell carcinoma at our institution to the date of the last followup or death. The 3 patients with unclassified renal cell carcinoma with missing grade (2) and ECOG performance status (1) data were presumptively assigned to UISS IV (2) and V (1) based on available clinical and pathological data. Of the 349 patients with 1997 TNM stages 3 and 4 clear cell renal cell carcinoma in the database 32 were excluded from study due to missing grade and ECOG performance status, while 20 with missing grade (19) or ECOG performance status (1) data were included in analysis. The latter group of 20 patients was presumptively assigned to UISS IV based on available clinical and pathological data. Survival data was missing for on 1 patient with clear cell disease. To determine variables significant for survival univariate analysis was performed in the combined population of patients with unclassified and clear cell renal cell carcinoma. The impact of patient age, sex, unclassified renal cell carcinoma histology, tumor size, nephrectomy, regional lymph node involvement, distant metastasis, UISS stage and immunotherapy were analyzed using the Cox proportional hazards regression model at a level of 0.05. To determine confounding factors variables significant on univariate analysis were subjected to backward stepwise multivariate analysis with entry and exclusion criteria of 0.1 and 0.15, respectively. Survival results were also assessed by the product-time method of Kaplan-Meier. Comparisons of groups were performed by the log rank method to assess the significance of the Kaplan-Meier curves. Results were based on the assumption check of the Cox regression model. Statistical analyses were performed using commercially available computer software. RESULTS
FIG. 1. Various histological appearances of unclassified renal cell carcinoma. a, pure anaplastic histology. b, mixed anaplastic and sarcomatoid histology. c, pure sarcomatoid histology. H&E, reduced from ⫻200.
The incidence of unclassified renal cell carcinoma in our patient population was 2.8%. Table 1 lists the characteristics of patients with unclassified and clear cell renal cell carcinoma. There was no history of von Hippel Lindau disease or other familial cancer syndromes. Patients with unclassified and clear cell disease did not differ in regard to a family history positive for cancer (14 or 45% versus 112 or 35%,
952
UNCLASSIFIED RENAL CELL CARCINOMA TABLE 1. Demographic, clinical and staging data on patients with unclassified and clear cell renal cell carcinoma Unclassified
No. pts. % Male Mean age ⫾ SD (range) No. family ca history (%) No. incidental diagnosis (%): Asymptomatic Symptomatic No. ECOG performance status (%): 0 1 2 3 X No. 1997 TNM stage (%): 3 4 No. UISS (%): 3 4 5 No. N0M0 (%) No. N0M⫹ (%) No. N⫹M0 (%) No. N⫹M⫹ (%)
Clear Cell
Nephrectomy Completed
Nephrectomy Not Attempted or Completed
Nephrectomy Completed
Nephrectomy Not Attempted or Completed
19 63 55.3 ⫾ 13.9 (25–81) 10 (53) 3 (16) 2 (11) 1 (5)
12 75 55.4 ⫾ 15.2 (27–81) 4 (33) 1 (8) 0 1 (8)
286 67 58.9 ⫾ 11.0 (25–89) 95 (33) 55 (19) 24 (8) 31 (11)
31 68 60.3 ⫾ 11.3 (30–80) 17 (55) 2 (6) 1 (3) 1 (3)
2 (11) 12 (63) 4 (21) 0 1 (5)
0 9 (75) 2 (17) 1 (8) 0
47 (16) 227 (79) 10 (3) 1 (0.3) 1 (0.3)
2 (6) 24 (77) 5 (16) 0 0
3 (16) 16 (84)
0 12 (100)
56 (20) 230 (80)
0 31 (100)
79 203 4 53 160 17 56
1 (3) 30 (97) 0 0 18 (58) 1 (3) 12 (39)
3 (16) 11 (58) 5 (26) 2 (11) 11 (58) 1 (5) 5 (26)
0 9 3 0 2 1 9
p ⫽ 0.28) or patient history of nonrenal malignancy (3 or 10% versus 23 or 7%, p ⫽ 0.0624). The rate of detecting an incidental renal mass also did not differ in the 2 groups (4 or 21% versus 57 or 18%, p ⫽ 0.64). No symptomatic patients presented with the classic triad. At diagnosis 29 patients (94%) with unclassified and 264 (83%) with clear cell renal cell carcinoma had regional lymph node or distant metastases (p ⫽ 0.143). All patients with unclassified and clear cell renal cell carcinoma who did not undergo attempted or completed nephrectomy presented with regional lymph node or distant metastases at initial diagnosis. As defined by 1997 TNM staging criteria, regional lymph nodes were involved in 16 unclassified (52%) and 86 clear cell (27%) cases (p ⫽ 0.0042). Of the 27 unclassified and 246 clear cell cases that presented with distant metastatic disease, the lung (24 or 89% versus 199 or 81%, p ⫽ 0.45), bone (14 or 52% versus 74 or 30%, p ⫽ 0.022) and nonregional lymph nodes (11 or 41% versus 54 or 22%, p ⫽ 0.03, respectively) were most commonly involved. Patients with unclassified disease were also 20 times more likely than those with
(75) (25) (17) (8) (75)
(28) (71) (1) (19) (56) (6) (20)
clear cell disease to be grouped as UISS V (8 or 26% versus 4 or 1.3%, p ⫽ 0.00001). Nephrectomy was completed in 19 (61%) and 286 (90%), attempted but unresectable in 3 (10%) and 5 (1.6%), and not attempted in 9 (29%) and 26 (8%) patients, with unclassified and clear cell disease, respectively (p ⫽ 0.00007). Table 2 lists grade and stage data on patients who did and did not successfully undergo nephrectomy. Average tumor size in patients who did not undergo nephrectomy was slightly greater than in those in whom it was completed in the 2 histological groups of unclassified and clear cell renal cell carcinoma (p ⫽ 0.757 and 0.812, respectively). Regardless of whether nephrectomy was completed average unclassified tumor size plus or minus standard deviation was greater than that of clear cell renal cell carcinoma tumors (10.8 ⫾ 3.9 cm., range 3.5 to 18.0 versus 8.8 ⫾ 3.6, range 1.8 to 22.0, p ⫽ 0.0054). Compared with clear cell cases unclassified cases were more likely to have tumor involvement of the adrenal glands (8 or 25% versus 13 or 4%, p ⫽ 0.0001) and local tumor extension to adjacent organs (13 or 42%
TABLE 2. Tumor characteristics in patients with unclassified and clear cell renal cell carcinoma Unclassified Nephrectomy Completed No. pts. No. rt. tumor (%) Mean cm. tumor size ⫾ SD (range) No. Fuhrman grade (%): 1 2 3 4 Missing No. adrenal involvement (%) No. renal vein extension (%) No. inferior vena cava extension (%) No. adjacent organ extension (%): Psoas muscle Liver Bowel Diaphragm Inferior vena cava Abdominal/chest wall
19
Clear Cell
Nephrectomy Not Attempted or Completed 9/3
10 (53) 10.6 ⫾ 3.4 (4.5–17.0)
6 (50) 11.0 ⫾ 4.8 (3.5–18.0)
0 0 9 (47) 10 (53) 0 4 (21) 3 (16) 2 (11)
0 0 8 (67) 2 (17) 2 (17) 4 (33) 2 (17) 3 (25)
1 (5) 1 (5) 1 (5) 2 (11) 3 (16) 0
1 (8) 0 3 (25) 0 0 1 (8)
Nephrectomy Completed 286 163 (57) 8.8 ⫾ 3.6 (3.0–22.0) 6 125 142 6 7 7 55 54
Nephrectomy Not Attempted or Completed 26 Not attempted, 5 not completed 15 (50) 9.0 ⫾ 3.5 (4.0–19.0)
(2) (44) (50) (2) (2) (2) (19) (19)
2 (6) 7 (23) 10 (32) 0 12 (39) 6 (19) 2 (6) 5 (16)
2 (0.7) 1 (0.3) 1 (0.3) 0 0 0
3 (10) 2 (6) 1 (3) 1 (3) 1 (3) 1 (3)
953
UNCLASSIFIED RENAL CELL CARCINOMA
versus 25 or 8%, p ⫽ 0.00001). However, the incidence of tumor extension into the renal vein or inferior vena cava was similar. Of the patients who underwent attempted or completed nephrectomy 2 of the 22 with unclassified (9%) and 4 of the 291 with clear cell (1.4%) disease died of perioperative and postoperative complications (Yates correction p ⫽ 0.08). The patients with unclassified disease died of sepsis and multiorgan failure on postoperative day 7 and of cardiac arrest perioperatively, respectively. The 4 patients with clear cell disease died of sepsis and multi-organ failure on postoperative day 13, coagulopathy on postoperative day 1, inferior vena caval tumor emboli to the atrium and pulmonary artery intraoperatively, and pulmonary embolism on postoperative day 12, respectively. Major complications developed postoperatively in 5 cases of unclassified (23%) and 52 of clear cell (17%) disease in which surgery was done for kidney cancer (p ⫽ 0.59). Univariate analysis of the combined unclassified and clear cell renal cell carcinoma groups indicated that unclassified histology (p ⬍0.0001), UISS (p ⬍0.0001), completed nephrectomy (p ⬍0.0001), regional lymph node disease (p ⬍0.0001), tumor size (p ⫽ 0.0148) and distant metastasis (p ⫽ 0.0209) were significantly associated with patient survival, while immunotherapy approached significance (p ⫽ 0.0912). Patient age (p ⫽ 0.3604) and sex (p ⫽ 0.4429) were not significant factors. On multivariate analysis unclassified histology (p ⬍0.0001) and UISS (p ⫽ 0.0001) remained highly significant factors, whereas completed nephrectomy (p ⫽ 0.0004) and immunotherapy (p ⫽ 0.0015) were protective (table 3). Because unclassified renal cell carcinoma and UISS were the most significant prognostic factors, survival in patients with unclassified and clear cell disease within a UISS group was compared. We compared those with UISS IV disease because most patients with the 2 types of renal cell carcinoma were categorized in this group. Figure 2 shows that when matched for UISS IV, patients with unclassified disease had significantly lower survival compared with those with clear cell disease (median 3.9 versus 17.2 months, p ⬍0.0001). This finding indicates that unclassified histology had a detrimental impact beyond the UISS categories, which groups cases of similar stage, tumor grade and performance status. A comparison of patients with distant metastasis also indicated significantly lower survival for unclassified versus clear cell disease (median 3.8 versus 19.9 months, p ⬍0.0001). Overall median survival in patients with unclassified renal cell carcinoma was 4.3 months. There was no significance difference in survival in those who did and did not undergo completed nephrectomy (median 4.7 and 3.9 months, respectively, p ⫽ 0.1086). This result suggests that the protective effect of nephrectomy on multivariate analysis was only attributable to its effect on those with clear cell renal cell carcinoma. Patients with unclassified disease who received immunotherapy had significantly improved survival compared with those who did not (median 11.1 and 2.9 months, respectively, p ⫽ 0.008). Of the patients with metastatic disease at initial diagnosis, 14 (52%) and 173 (70%) with unclassified and clear cell disease received immunotherapy (p ⫽ 0.05), including 6 and 17 in whom the primary tumor was not removed and 8 and 156 who underwent completed nephrectomy, respectively. All 14 patients with unclassified
FIG. 2. Kaplan-Meier comparison of survival in patients with UISS IV unclassified and clear cell renal cell carcinoma (RCC) shows significant survival advantage for clear cell variety.
renal cell carcinoma treated with immunotherapy received IL-2 based immunotherapy, including 6 who received high dose IL-2 in a nonintensive care setting. There were no complete responses, 1 patient achieved a partial response, 3 had stable disease and 10 had progressive disease after treatment. Patients with clear cell renal cell carcinoma who received immunotherapy had improved survival compared with patients with unclassified disease whether nephrectomy was (p ⫽ 0.0042) or was not (p ⫽ 0.0479) performed. Treating clear cell renal cell carcinoma with combined nephrectomy and immunotherapy conferred a survival advantage over immunotherapy with the primary tumor still in place (median 25.1 versus 10.9 months, p ⫽ 0.0007). The advantage for unclassified renal cell carcinoma was less dramatic (12.9 versus 7.7 months, p ⫽ 0.0356). At diagnosis there were 2 cases (7%) of stage N0M0 unclassified renal cell carcinoma. One patient who was symptomatic at initial diagnosis underwent nephrectomy for a grade 4 stage pT3b (UISS III) tumor. Local recurrence developed 3 months after nephrectomy. After disease recurrence he was treated with a course of high dose IL-2. Disease progressed despite immunotherapy and the patient died 7 months postoperatively. The other patient was asymptomatic at diagnosis and underwent nephrectomy for a grade 4 stage pT3a (UISS III) pure sarcomatoid tumor. He had no evidence of disease 18.5 months postoperatively. All other patients with unclassified renal cell carcinoma died of cancer. DISCUSSION
This study indicates that unclassified renal cell carcinoma is an uncommon variant of renal cell carcinoma (2.8% of cases) that typically presents with significantly larger, more aggressive tumors and is associated with poor clinical outcomes. Only 10% of these patients are asymptomatic at initial diagnosis, while greater than 90% present with regional lymph node or distant metastases. Compared with clear cell renal cell carcinoma there is significantly increased propensity for unclassified disease to involve local tumor invasion of the adrenal gland and adjacent organs as well as metastatic
TABLE 3. Variables with a significant impact on the survival of patients with unclassified renal cell carcinoma on multivariate analysis
Unclassified renal cell Ca UISS Tumor size Nephrectomy completed Immunotherapy
Standard Estimate ⫾ Error
Chi-Square
p Value
Hazards Ratio (95% CI)
1.003 ⫾ 0.230 0.601 ⫾ 0.158 0.033 ⫾ 0.019 ⫺0.753 ⫾ 0.214 ⫺0.438 ⫾ 0.151
18.959 14.468 2.814 12.355 8.386
⬍0.0001 0.0001 0.0934 0.0004 0.0038
2.726 (1.736–4.281) 1.824 (1.338–2.487) 1.034 (0.994–1.074) 0.471 (0.309–0.717) 0.645 (0.480–0.868)
954
UNCLASSIFIED RENAL CELL CARCINOMA
involvement of the bone and regional/nonregional lymph nodes. Metastatic unclassified renal cell carcinoma confers a significant survival disadvantage compared with metastatic clear cell renal cell carcinoma, as also illustrated by the survival difference in the more homogenous UISS IV population (p ⬍0.0001). On multivariate analysis unclassified histology was a significant indicator of poor prognosis. Only 1 patient with unclassified disease (pure sarcomatoid tumor) achieved prolonged disease-free survival. Long-term disease-free survival for sarcomatoid renal cell carcinoma is uncommon but it has previously been reported.18, 19 The lack of clinical symptoms in this patient is indicative of a relatively early diagnosis and may explain his improved survival. Overall such a presentation is rare among patients with unclassified renal cell carcinoma. The dismal median survival of 4.3 months in these cases was not significantly impacted by surgical extirpation. While nephrectomy was completed in 90% of stage 3 and 4 clear cell renal cell carcinoma cases, it was completed in only 60% of the unclassified population. This finding reflects the unique biological characteristics of the tumor and the advanced state of disease at presentation. Although unclassified renal cell carcinoma comprised just 2.8% of the kidney cancer database population, the 14% rate of unresectability exceeded it 5-fold. In the stages 3 and 4 clear cell cohort the 29% prevalence in the database paralleled the 23% rate of unresectability. Patients with unclassified renal cell carcinoma who undergo surgery tend to have higher morbidity as well as higher mortality. The poor prognosis associated with unclassified renal cell carcinoma can be improved by administering immunotherapy. However, patients with unclassified disease were significantly more likely than those with clear cell disease to have contraindications against immunotherapy, which reflects the poor clinical status of the former as well as rapidly progressive disease. When immunotherapy was administered in these cases, it was beneficial when nephrectomy was performed and when the primary tumor remained in place. Nephrectomy alone did not improve survival in the unclassified group, whereas combined nephrectomy and immunotherapy conferred a survival advantage over immunotherapy alone. The previous study of Litwin et al shows that this prolonged survival can be accompanied by significant quality of life in patients with metastasis who undergo nephrectomy and IL-2 based immunotherapy according to cross-sectional assessment with health related quality of life instruments.20 Compared with reference populations with chronic medical conditions or nonrenal malignancies patients with metastatic renal cell carcinoma have transient impairment in quality of life during immunotherapy. However, after immunotherapy those with renal cell carcinoma have better quality of life estimates than those with other malignancies as well as better physical function than those with congestive heart failure, suggesting that patients with metastatic renal cell carcinoma may have well preserved quality of life when nephrectomy is followed by immunotherapy. Despite this promising data the benefits of combined nephrectomy and immunotherapy for unclassified renal cell carcinoma is modest compared with clear cell renal cell carcinoma. A limitation of this study is the possible bias resulting from patients with the primary tumor in place, in whom the classification of unclassified renal cell carcinoma was based only on the biopsy specimen. Insufficient sampling or the poor quality of tumor samples may have impacted the histological evaluation. Therefore, it is impossible to be certain that some renal cell carcinoma cases with the primary tumor in place and classifiable histology were not inadvertently incorporated into the unclassified population. Other limitations of this study are its retrospective nature, the fact that treatments were not randomly assigned and the fact that there was no untreated control arm for comparison. However, the
contribution of and improvement in survival related to immunotherapy21 and the benefit of nephrectomy in the context of multimodal treatment using cytoreductive surgery and immunotherapy22 have been supported by other randomized controlled studies of metastatic renal cell carcinoma. Renal tumors of unclassified histology show highly aggressive biological behavior, have a tendency toward metastatic involvement to the bones and lymph nodes, and are associated with a poor clinical outcome in the majority of affected patients. Nephrectomy combined with immunotherapy confers some survival advantage. We advocate that when a histological diagnosis of unclassified renal cell carcinoma is made preoperatively, patients should be advised to undergo nephrectomy only when they have no contraindications for immunotherapy. This recommendation is due to the lack of survival advantage attained by nephrectomy alone, and the high associated surgical morbidity and mortality in this patient population. Because of the aggressive nature of these tumors, immunotherapy should also be considered preoperatively, especially when surgery must be delayed. A question that remains to be answered is whether adjuvant immunotherapy should be routinely administered in patients with stage N0M0 unclassified renal cell carcinoma. Due to the small study population and the aggressive nature of this tumor such an issue is unlikely to be addressed adequately by the experience at a single institution. In conclusion, a well coordinated nationwide effort similar to Wilms tumor studies is warranted to evaluate prospectively the merits of adjuvant immunotherapy in patients with stage N0M0 renal cell carcinoma who are at high risk for disease recurrence and progression, such as those with unclassified, collecting duct and sarcomatoid renal cell carcinoma. REFERENCES
1. Storkel, S., Eble, J. N., Adlakha, K., Amin, M., Blute, M. L., Bostwick, D. G. et al: Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 987, 1997 2. Kovacs, G., Akhtar, M., Beckwith, B. J., Bugert, P., Cooper, C. S., Delahunt, B. et al: The Heidelberg classification of renal cell tumours. J Pathol, 183: 131, 1997 3. Chao, D., Zisman, A., Pantuck, A. J., Gitlitz, B. J., Freedland, S. J., Said, J. W. et al: Collecting duct renal cell carcinoma: clinical study of a rare tumor. J Urol, 167: 71, 2002 4. Dimopoulos, M. A., Logothetis, C. J., Markowitz, A., Sella, A., Amato, R. and Ro, J.: Collecting duct carcinoma of the kidney. Br J Urol, 71: 388, 1993 5. Schoenberg, M., Cairns, P., Brooks, J. D., Marshall, F. F., Epstein, J. I., Isaacs, W. B. et al: Frequent loss of chromosome arms 8p and 13q in collecting duct carcinoma (CDC) of the kidney. Genes Chromosomes Cancer, 12: 76, 1995 6. Mani, S., Todd, M. B., Katz, K. and Poo, W. J.: Prognostic factors for survival in patients with metastatic renal cancer treated with biological response modifiers. J Urol, 154: 35, 1995 7. Moch, H., Gasser, T., Amin, M. B., Torhorst, J., Sauter, G. and Mihatsch, M. J.: Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer, 89: 604, 2000 8. Delahunt, B. and Nacey, J. N.: Renal cell carcinoma. II. Histological indicators of prognosis. Pathology, 19: 258, 1987 9. Cangiano, T., Liao, J., Naitoh, J., Dorey, F., Figlin, R. and Belldegrun, A.: Sarcomatoid renal cell carcinoma: biologic behavior, prognosis, and response to combined surgical resection and immunotherapy. J Clin Oncol, 17: 523, 1999 10. Amin, M. B., Corless, C. L., Renshaw, A. A., Tickoo, S. K., Kubus, J. and Schultz, D. S.: Papillary (chromophil) renal cell carcinoma: histomorphologic characteristics and evaluation of conventional pathologic prognostic parameters in 62 cases. Am J Surg Pathol, 21: 621, 1997 11. Fromowitz, F. B. and Bard, R. H.: Clinical implications of pathologic subtypes in renal cell carcinoma. Semin Urol, 8: 31, 1990 12. Medeiros, L. J., Gelb, A. B. and Weiss, L. M.: Renal cell carci-
UNCLASSIFIED RENAL CELL CARCINOMA
13.
14. 15.
16.
17.
noma. Prognostic significance of morphologic parameters in 121 cases. Cancer, 61: 1639, 1988 Giberti, C., Oneto, F., Martorana, G., Rovida, S. and Carmignani, G.: Radical nephrectomy for renal cell carcinoma: long-term results and prognostic factors on a series of 328 cases. Eur Urol, 31: 40, 1997 Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 Guinan, P., Sobin, L. H., Algaba, F., Badellino, F., Kameyama, S., MacLennan, G. et al: TNM staging of renal cell carcinoma: Workgroup No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 992, 1997 Zisman, A., Pantuck, A. J., Dorey, F., Said, J. W., Schvarts, O., Quintana, D. et al: Improved prognostication of renal cell carcinoma using an integrated staging system. J Clin Oncol, 19: 1649, 2001 Zisman, A., Pantuck, A. J., Figlin, R. A. and Belldegrun, A. S.:
18. 19. 20.
21.
22.
955
Validation of the UCLA integrated staging system for patients with renal cell carcinoma. J Clin Oncol, 19: 3792, 2001 Ro, J. Y., Ayala, A. G., Sella, A., Samuels, M. L. and Swanson, D. A.: Sarcomatoid renal cell carcinoma: clinicopathologic. A study of 42 cases. Cancer, 59: 516, 1987 Bertoni, F., Ferri, C., Benati, A., Bacchini, P. and Corrado, F.: Sarcomatoid carcinoma of the kidney. J Urol, 137: 25, 1987 Litwin, M. S., Fine, J. T., Dorey, F., Figlin, R. A. and Belldegrun, A. S.: Health related quality of life outcomes in patients treated for metastatic kidney cancer: a pilot study. J Urol, 157: 1608, 1997 Interferon-alpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Medical Research Council Renal Cancer Collaborators. Lancet, 353: 14, 1999 Flanigan, R. C., Salmon, S. E., Blumenstein, B. A., Bearman, S. I., Roy, V., McGrath, P. C. et al: Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med, 345: 1655, 2001