A renal mass in the setting of a nonrenal malignancy†
When is a renal tumor biopsy appropriate?
Presented at the Third Annual Meeting of the Society of Urologic Oncology, Bethesda, Maryland, December 13–14, 2002.
Abstract
BACKGROUND
Frequently, a renal mass is identified when patients with cancer undergo studies for staging or surveillance. In determining whether it represents a metastasis, patients are frequently subjected to percutaneous renal biopsies. The authors evaluated their experience with this dilemma to formulate management guidelines.
METHODS
The authors reviewed the medical records of 100 consecutive patients with nonrenal malignancies diagnosed with renal masses at presentation or follow-up. Renal mass histology was available for all patients after nephrectomy or biopsy. Clinical characteristics were assessed to identify factors predictive for a renal metastasis versus a primary renal neoplasm.
RESULTS
The only factors predictive of a metastasis to the kidney were progression of the nonrenal malignancy and lack of enhancement of the renal mass (P < 0.0001). Forty-six patients (46%) had evidence of progression of their nonrenal malignancy in addition to the renal mass. In these patients, the probability of a metastasis to the kidney was 86% (95% confidence interval [CI], 57.2–98.2%) without renal mass enhancement and 32% (95% CI, 14–55%) with enhancing renal masses. None of the 54 patients without signs of progression of their nonrenal malignancy proved to have metastases to the kidney, regardless of the imaging characteristics of the mass (zero probability; 95% CI, 0–7%; P < 0.001).
CONCLUSIONS
In patients presenting with renal masses and another clinically localized malignancy, renal mass biopsies were not indicated, as the mass rarely represented a metastasis. These patients may opt for close surveillance or extirpation based on the prognosis of their nonrenal malignancy. Cancer 2004. © 2004 American Cancer Society.
Although metastases to the kidney have been reported to occur in 7–20% of patients with cancer at autopsy,1-3 the diagnosis of metastases to the kidney in patients without evidence of a disseminated nonrenal malignancy is rare. Nevertheless, owing to the widespread use of cross-sectional imaging, an increasing number of patients with nonrenal malignancies have renal masses diagnosed during routine tumor staging or surveillance. In a report of 1000 patients with cancer who underwent abdominal computed tomography (CT) scans to determine whether metastases were present, Pagani4 identified 7 patients with unsuspected renal masses. In that series and others, the tumors that metastasized most commonly to the kidney were lymphoma and carcinomas of the lung, breast, stomach, pancreas, and colon.5-8 However, because of reports of tumor needle tract seeding,9-11 associated morbidity,12, 13 and false-negative results as high as 15%,13, 14 percutaneous biopsy of solid renal tumors is not recommended routinely.15 Exceptions are usually made for patients with other primary malignancies because the possibility of the renal mass being a metastasis could have a significant impact on clinical management.15, 16 We are not aware of studies establishing guidelines for performing a renal mass biopsy in patients with a history of a nonrenal malignancy. Therefore, in the current study, we reviewed our experience with this clinical problem.
MATERIALS AND METHODS
Study Patients
We performed a retrospective review of the medical records of 100 consecutive patients who were diagnosed between July 1994 and August 2002 with a nonrenal malignancy and a renal mass at presentation or during follow-up. These patients were identified from an institutional database of all consecutive patients with cancer who underwent a renal mass biopsy or nephrectomy at The University of Texas M. D. Anderson Cancer Center (MDACC; Houston, TX). Only patients with discrete solid parenchymal masses were included. Patients with classic radiographic findings suggestive of renal lymphoma such as diffuse renal enlargement or solid perinephric masses not involving the kidney were excluded. All patients had undergone a thorough history, physical examination, and radiographic staging with a bone scan, CT scan, or magnetic resonance imaging (MRI) scan of the chest, abdomen, and pelvis. Brain imaging was performed as indicated in patients with neurologic symptoms. All patients underwent a percutaneous renal biopsy or nephrectomy (partial or radical) for histologic diagnosis or treatment, respectively. Clinical and radiographic characteristics were reviewed to identify factors predictive of a renal metastasis or a primary renal neoplasm.
Statistical Analysis
Univariate analysis was performed to evaluate the relation between a final renal mass diagnosis and clinical variables such as the status of the primary nonrenal malignancy, timing of diagnosis (synchronous or metachronous), patient age and gender, tumor location (unilateral or bilateral), tumor size, number of renal tumors, presence of renal mass enhancement by CT or MRI scans, body mass index (BMI), history of smoking or hypertension, or family history of cancer. These variables were included because of the known association between renal carcinoma and genetic predisposition, smoking, obesity, and hypertension.17 The probability of a metastasis to the kidney was estimated with a 95% confidence interval (CI) in each category with the combinations of renal mass enhancement as observed on CT or MRI scans (present or absent) and nonrenal malignancy progression (yes or no). Progression of the nonrenal malignancy was defined as the presence of systemic disease as documented by a biopsy, imaging scans, laboratory studies, or disease outside the primary organ site (one patient with lung carcinoma had chest wall invasion). The Fisher exact test was used to evaluate the dependence between the two categoric variables. The probability of an event (along with a 95% CI) was estimated using the StatXact-3 statistical package (Cytel Software Corporation, Cambridge, MA). Ten patients with nondiagnostic renal biopsies were excluded from the analysis.
RESULTS
Patients and Clinical Presentation
The mean age of the cohort was 60.7 years (range, 31–91 years). The mean renal mass size was 3.8 cm (range, 1–11 cm). Clinical presentation was incidental in 95% of patients, whereas 2% and 3% had had flank pain and hematuria, respectively. In 49% of patients, the presentation of the renal mass was synchronous with the diagnosis and staging of the nonrenal malignancy. Most of the patients were men (61%). Racial or ethnic groups included Caucasian (82%), Hispanic (9%), African-American (6%), Middle Eastern (2%), and Asian (1%) patients. Renal masses were located in the right kidney (54%), left kidney (40%), both kidneys (5%), and in the horseshoe kidney isthmus (1%).
Treatment
Twenty-six patients without progression of their nonrenal malignancy were managed by surgical excision without a preoperative biopsy of the renal mass. All 26 patients had primary renal tumors (Fig. 1). In 74 patients, a biopsy of the renal mass had been performed because of a history of a nonrenal malignancy or concurrent metastatic disease elsewhere. Biopsy results were metastasis in 18 patients, primary renal neoplasms in 41 patients, and 15 patients had a nondiagnostic biopsy. Of 18 patients (24.3%) who had metastatic disease as revealed in biopsy results, all had other clinical or radiographic evidence of progression from their primary malignancy. Fifty-five percent (41 of 74) of patients who underwent a renal mass biopsy had a diagnosis of renal cell carcinoma (RCC), oncocytic neoplasm (n = 6), or angiomyolipoma (n = 1). Of these, 24 underwent extirpative surgery during which the diagnosis was confirmed, and 17 were observed because of the poor prognosis associated with their nonrenal carcinoma. Fifteen patients had a negative or nondiagnostic biopsy. Ten of these who did not develop subsequent renal mass enlargement were followed with surveillance, and five underwent surgical extirpation. Four of these five had primary renal tumors and one had a metastasis from lung carcinoma.
Distribution of patients based on diagnosis and treatment. RCC: renal cell carcinoma; AML: renal angiomyolipoma.
Nonrenal Malignancies
A detailed description of the primary malignancy sites is presented in Table 1. Patients with the highest proportion of metastases to the kidney included those with primary tumors of the lung (30.8%), lymphoma (19%), esophagus (33.3%), and head and neck (30%). Of the 4 most common primary malignancies (lymphoma, breast, lung, and colon carcinoma), the patients with the lowest incidence of renal metastases in our series were those with breast (7.7%) and colon carcinomas (9.1%). With regards to their nonrenal malignancy, 54 patients had no clinical or radiographic evidence of progression or metastases except for the renal mass. None of these patients were found to have metastases to the kidney (Fig. 2). Forty-six patients (of the 100 consecutive patients) had clinical progression or radiographic evidence of other metastases from their primary malignancy in addition to the renal mass. In 19 of these 46 patients (41%; 95% CI, 27–56.8%), the renal mass was found to be a metastasis from their nonrenal malignancy.
| Primary malignancy type | No. of patients | Metastases (%) | Final diagnosis of renal mass (%) | Nondiagnostic (%) | ||
|---|---|---|---|---|---|---|
| RCC | AML | Oncocytoma | ||||
| Lymphoma | 21 | 4 (19.0) | 10 (47.6) | 3 (14.3) | 4 (19.0) | |
| Breast | 13 | 1 (7.7) | 10 (76.9) | 2 (15.4) | ||
| Lung | 13 | 4 (30.8) | 8 (61.5) | 1 (7.7) | ||
| Colon | 11 | 1 (9.1) | 7 (63.6) | 1 (9.1) | 2 (18.2) | |
| Melanoma | 8 | 7 (87.5) | 1 (12.5) | |||
| Prostate | 6 | 4 (66.7) | 2 (33.3) | |||
| Head and neck | 5 | 2 (40.0) | 3 (60.0) | |||
| Pancreas | 5 | 1 (20.0) | 2 (40.0) | 1 (20.0) | 1 (20.0) | |
| Esophagus | 3 | 1 (33.3) | 2 (66.7) | |||
| Ovary | 2 | 1 | 1 | |||
| Extremity sarcoma | 2 | 1 | 1 | |||
| Testis | 2 | 1 | 1 | |||
| Myelodysplasia | 2 | 2a | ||||
| Neuroblastoma | 1 | 1 | ||||
| Cervix | 1 | 1 | ||||
| Skin (squamous) | 1 | 1 | ||||
| Small bowel | 1 | 1 | ||||
| Biliary | 1 | 1 | ||||
| Bladder | 1 | 1 | ||||
| Breast and lymphoma | 1 | 1 | ||||
| Total | 100 | 19 | 59 | 2 | 10 | 10 |
- RCC: renal cell carcinoma; AML: angiomyolipoma.
- a Extramedullary hematopoiesis.
Renal mass diagnosis after nephrectomy or renal biopsy according to the status of the nonrenal malignancy.
Factors Associated with Metastases to the Kidney
The characteristics of the 90 patients with a diagnosis of a primary renal tumor (n = 71) or metastasis to the kidney (n = 19) are presented in Table 2. Ten patients with nondiagnostic renal biopsies were excluded from this portion of the analysis. In particular, although all patients (19 of 19 patients [100%]) with metastases to the kidney had evidence of progression of the nonrenal malignancy, only 23.9% of patients with primary renal tumors (17 of 71) had progression of their nonrenal malignancy (Fisher exact test: P < 0.0001). The absence of renal mass enhancement on CT or MRI scans was also highly associated with metastases to the kidney. Renal mass enhancement after contrast administration was present in only 36.8% of patients with metastases to the kidney, as compared with 91.5% of patients with primary renal tumors (Fisher exact test: P < 0.0001).
| Characteristic | Final diagnosisa | P value | |
|---|---|---|---|
| Renal metastasis (n = 19) (%) | Primary renal tumor (n = 71) (%) | ||
| Progression of nonrenal malignancy | < 0.0001 | ||
| Yes (n = 36) | 19 (100) | 17 (23.9) | |
| No (n = 54) | 0 | 54 (76.1) | |
| Renal mass enhancement | < 0.0001 | ||
| Yes (n = 72) | 7 (36.8) | 65 (91.5) | |
| No or indeterminate (n = 18) | 12 (63.2) | 6 (8.5) | |
| Age (yrs) | 0.43 | ||
| Median (interquartile range) | 59.5 (54–60) | 64 (54–65) | |
| Percentage female | 42.1 | 36.6 | 0.79 |
| Racial or ethnic group | 0.24 | ||
| Non-Hispanic white | (84.2) | (83.1) | |
| Non-Hispanic black | (10.5) | (7.0) | |
| Hispanic | (0) | (8.5) | |
| Other | (5.3) | (1.4) | |
| Renal mass presentation | 0.49 | ||
| Asymptomatic | 0 | 5 (7) | |
| Symptomatic | 19 (100) | 66 (93) | |
| 0.44 | |||
| Synchronous | (36.8) | (49.3) | |
| Metachronous | (63.2) | (50.7) | |
| Tumor size (cm) | 0.49 | ||
| Mean | 4.34 | 3.93 | |
| Median (interquartile range) | 4.0 (3.0–5.4) | 3.0 (2.0–5.5) | |
| No. of renal tumors | 0.86 | ||
| Median | 1 | 1 | |
| > 1 renal mass | 4 (21) | 5 (7) | 0.09 |
| Bilateral tumors | 2 (10.5) | 2 (2.8) | 0.20 |
| History of hypertension | 7 (36.8) | 32 (45.1) | 0.63 |
| History of smoking | 11 (57.9) | 40 (56.3) | 0.31 |
| Family history of malignancy | 13 (68.4) | 43 (60.6) | 0.54 |
| Family history of RCC | 0 | 4 (5.6) | 0.57 |
| Body mass index (kg/m2) | 0.11 | ||
| Median (interquartile range) | 21.7 (18.5–27.5) | 29.7 (23.9–35.1) | |
- RCC: renal cell carcinoma.
- a Diagnosis was based on renal biopsy or partial or radical nephrectomy. Ten patients with nondiagnostic renal biopsy data were excluded.
We also found that these two variables (progression of the nonrenal malignancy and renal mass enhancement after contrast administration on imaging scans) were highly correlated (Fisher exact test: P < 0.001; Table 3). In particular, the probability of a metastasis to the kidney was estimated to be 86% (95% CI, 57–98%) for patients with nonrenal malignancy progression and no evidence of renal mass enhancement, and 32% (95% CI, 14–55%) for patients with nonrenal malignancy progression and renal mass enhancement. The probability of metastases to the kidney was estimated to be zero for patients without progression of their nonrenal malignancy, regardless of whether the renal mass was enhancing (95% CI, 0–7%) or nonenhancing (95% CI, 0–60%).
| Nonrenal malignancy progression | Renal mass enhancement | |||
|---|---|---|---|---|
| Yes | No | |||
| Probability | 95% CI | Probability | 95% CI | |
| Yes | 32% (7/22) | 14–55% | 86% (12/14) | 57–98% |
| No | 0% (0/50) | 0–7% | 0% (0/4) | 0–60% |
- CI: confidence interval.
- a Ten patients with nondiagnostic renal biopsies were excluded. Fisher exact test P < 0.001.
There were no statistically significant differences between patients with primary renal tumors and those with metastases to the kidney with regards to any other clinical or radiographic factors (Table 2). There was a tendency for patients with metastases to the kidney to have bilateral masses, a higher number of renal masses, and a lower mean BMI than patients with primary renal tumors.
Imaging and Pathologic Characteristics
In the current study of 100 patients, 91% of all renal masses were solid tumors and 9% were solid tumors with a cystic component. A higher proportion of pure solid masses were metastatic tumors to the kidney (20%; 95% CI, 12–30%), compared with solid masses with a cystic component (11%; 95% CI, 0.3–48%). However, the difference was not statistically significant. All patients with a final diagnosis of a metastasis to the kidney (n = 19) had undergone contrast-enhanced imaging with CT or MRI scans. Although only 37% (95% CI, 16–62%) of the renal masses that proved to be metastases were enhanced after the administration of contrast material, 92% [95% CI, 83–97%) of masses in patients with a final diagnosis of a primary renal tumor demonstrated enhancement on imaging studies (Fisher exact test: P < 0.001; Table 2).
The majority of all renal masses were primary renal tumors (59 RCCs, 10 oncocytomas, and 2 angiomyolipomas). Of the 46 patients with RCC who were treated with partial (n = 20) or radical nephrectomy (n = 26), renal mass histology was conventional in 80.4%, papillary in 10.9%, unclassified in 6.5%, and chromophobe in 2.2%. Tumors were of Fuhrman Grades 2 (45.7%), 3 (45.7%), and 4 (8.6%). Pathologic stages were pT1 (76%), pT2 (4.3%), pT3a (11%), pT3b (0%), and pT4 (8.6%). Only one patient had evidence of lymph node involvement (N2) from RCC, and two other patients had distant metastases from RCC at presentation.
DISCUSSION
This retrospective study of 100 consecutive patients demonstrates that most patients with discrete, parenchymal renal masses diagnosed in the setting of a nonrenal malignancy will harbor either malignant (59%) or benign (12%) primary renal tumors. Of 19 patients who were found to have metastases to the kidney, all had evidence of clinical progression or radiographic evidence of other metastases from their nonrenal malignancy. On the basis of these data, a renal mass biopsy does not seem warranted in patients with a discrete renal mass diagnosed in the setting of an otherwise clinically localized nonrenal malignancy. Indications to perform a biopsy for a renal mass in the setting of a metastatic nonrenal malignancy should be based on the prognosis of the patient and the importance of differentiating the presence of an additional metastastic organ site versus another primary malignancy in that setting.
Few previous studies have evaluated the clinical, pathologic, and radiographic characteristics of renal masses found in patients with a history of other primary malignancies. Among 1000 patients with cancer who underwent cross-sectional imaging to determine the presence of metastases, Pagani4 identified 7 patients with unsuspected renal masses. The primary malignancies included carcinomas of the breast (n = 3), lung (n = 2), colon (n = 1), and lymphoma (n = 1). Renal mass histology was confirmed to be RCC in five patients and metastatic disease to the kidney in two patients. Similar to results from our series, both patients with metastatic disease to the kidney in the Pagani series had evidence of other metastases (colon carcinoma with liver metastases) or progression from their nonrenal malignancy (diffuse lymphoma) at the time of diagnosis of the renal mass. In addition, neither patient had evidence of enhancement of the renal mass by diagnostic imaging. In another series18 of four patients with lung carcinoma who developed metastases to the kidney, three patients had evidence of radiographic progression of their primary lung tumor. Comparable to our experience, the most common malignancies metastatic to the kidney in this autopsy series were lymphoma and lung carcinoma.18
Although several studies have evaluated the utility of a renal mass biopsy to evaluate patients with an indeterminate renal mass,19, 20 we are not aware of any studies specifically aimed at establishing guidelines for renal biopsies for patients with a history of a previous malignancy. Wood et al.20 reviewed the results of 79 renal biopsies performed under ultrasound or CT scan guidance for 73 patients with indeterminate renal masses. Although biopsies were performed in 24 patients with a previous malignancy, the type and status of the nonrenal malignancy were not discussed. Similarly, in studies by Gattuso et al.16 and Johnson et al.21, in which 58 biopsies were performed in patients with nonrenal malignancies, no information was provided on the status of the other malignancy at the time the biopsy was performed. Furthermore, no guidelines or recommendations for performing a renal biopsy were presented. Mydlo and Gerstein22 reported on 5 patients diagnosed with RCC and other nonrenal malignancies over a 6-year period, including colon carcinoma (n = 3), breast carcinoma (n = 1), and uterine carcinoma (n = 1). However, only one of the renal tumors was detected during staging of the primary tumor (colon carcinoma), and the other four renal masses were detected during a hematuria evaluation.
In a series from Memorial Sloan-Kettering Cancer Center, patients with papillary RCC were at increased risk of developing both bladder and prostate carcinoma.23 We did not find an association between RCC tumor histology and the type of coexistent nonrenal primary malignancy in our patients. Our 5 patients who underwent surgery that revealed papillary RCC had a history of lymphoma (n = 2) and carcinomas of the colon, pancreas, and head and neck.
Traditionally, a renal biopsy is recommended for patients with a history of lymphoma or leukemia who develop a renal mass, because the preferred treatment involves systemic chemotherapy with or without radiotherapy. However, our data do not support a renal biopsy for all patients with a history of lymphoma who have a renal mass, particularly if the patient has a discrete, enhancing, parenchymal renal mass and no other evidence of lymphoma progression. Of 24 patients with lymphoma or myelodysplasia and a renal mass, the variable most predictive of renal mass histology was the clinical status of the nonrenal malignancy at the time of diagnosis of the renal mass. In fact, for 12 patients with a history of lymphoma in remission, the histology of the renal mass never proved to be lymphoma. This potential association between lymphoma and RCC has been studied previously24-27 and remains controversial. At MDACC, Anderson et al.24 identified 41 patients over a 40-year period with a simultaneous diagnosis of RCC and non-Hodgkin lymphoma. As confirmed by two other hospital-based studies,26, 27 the observed rates of RCC diagnosed in patients with a history of lymphoma were greater than expected in the general population. This was also true for the risk of lymphoma developing in patients with a history of RCC. This association, however, has been refuted by a retrospective epidemiologic study based on Surveillance, Epidemiology, and End Results data. In this study, Rabbani and Russo25 found no increased risk of non-Hodgkin lymphoma after RCC or vice versa when the first year of follow-up was excluded. Whether the perceived association between one form of cancer and the other is due to methodologic differences in study design remains to be evaluated.
In conclusion, the majority of patients with renal masses diagnosed in the setting of a clinically localized nonrenal malignancy will not have metastatic disease to the kidney. Given that all patients who were found to have metastases to the kidney had evidence of clinical progression or radiographic evidence of other metastases from their nonrenal malignancy, a renal mass biopsy may not be indicated if the nonrenal malignancy is clinically localized. Patients with a nonrenal primary tumor with poor prognosis and a small renal mass may choose careful surveillance of the renal mass, whereas patients with a long life expectancy are more likely to benefit from surgical extirpation of their renal tumor.
