The patient is a 43-year-old man who was initially evaluated at an outside institution for unexplained anemia and who was found to have a large right kidney mass. He underwent a radical nephrectomy for a 19-cm large-cell, poorly differentiated neoplasm, consistent with pleomorphic, epithelioid angiomyolipoma (EAML) with extensive necrosis and cytologic atypia. A surveillance abdominal ultrasound was performed 6 months after surgery; this showed no evidence of disease. However, a second ultrasound performed 12 months after nephrectomy showed extensive involvement of his liver, with multiple large lesions. Fine-needle biopsy of one of the liver lesions confirmed the presence of epithelioid, pleomorphic cells consistent with EAML. Staging computed tomography (CT) scans of the chest, abdomen, and pelvis were also notable for several 2- to 4-mm pulmonary nodules. A CT scan of the head with contrast showed no intracranial lesions.
The patient presented to our cancer center 14 months after his nephrectomy for consultation and management of his recurrent metastatic disease. His main complaints at this time were fatigue, dyspnea, nausea, anorexia, and occasional right upper quadrant abdominal pain. The patient’s medical history was notable for hypercholesterolemia and for a Brugada-like syndrome, for which he had undergone prophylactic implantable cardioverter defibrillator placement. His family history was notable for a brother who died of sudden cardiac death at age 46, and a father who died of pancreatic cancer at age 72. On physical examination, he appeared fatigued and had palpable hepatomegaly and a hepatic mass just below the right costal margin. His lungs were clear, and he had no rashes or skin lesions. Laboratory evaluation revealed the following levels: creatinine, 1.09 mg/dL; calcium, 8.4 mg/dL; albumin, 2.5 g/dL; aspartate aminotransferase, 46 U/L; alanine aminotransferase, 23 U/L; total bilirubin, 2.0 mg/dL; direct bilirubin, 0.5 mg/dL; alkaline phosphatase, 169 U/L; lactate dehydrogenase, 449 U/L; and thyroid stimulating hormone, 1.58 μIU/mL. White blood cell count was 12.6 × 103/μL; hemoglobin level, 9.2 g/dL; platelet count, 313 × 103/μL; prothrombin time, 16.1 sec; international normalized ratio, 1.3; and activated partial thromboplastin time, 31.9 sec.
DR. ELAINE T. LAM, medical oncologist: Dr. La Rosa, would you review the patient's pathology?
DR. FRANCISCO G. LA ROSA, pathologist: We reviewed the original slides (from an outside institution) of the right kidney nephrectomy specimen. The sections showed a tumor composed of medium- to large-sized polygonal epithelioid cells with eosinophilic cytoplasm, moderately pleomorphic nuclei, prominent nucleoli, and occasional mitotic figures. The tumor showed focal areas of hemorrhage and necrosis and was admixed with clusters of foamy macrophages, some with golden brown pigments consistent with hemosiderin deposits (Figure 1A and 1B). Special stains showed that the tumor cells were positive for Melan-A103 (Figure 1C), HMB-45 (MAA, Figure 1D), and CD68; weakly positive for desmin and CD117; and negative for pancytokeratin, epithelial membrane antigen (EMA), S100 antigen, carbonic anhydrase IX (CA-9), and alpha-methylacyl-CoA racemase (AMACR). Nonspecific staining was present with CK7, 34BE12, and CD10.
We also reviewed the outside slides of the fine-needle aspiration specimen from one of the lesions in the right lobe of the liver. The cytology smears and histological sections from the cell block showed pleomorphic epithelioid cells very similar to those described in the renal tumor, which were consistent with metastatic EAML (Figure 2A and 2B). Results of immunohistochemical staining performed at the outside institution were reported as positive for HMB-45 and desmin (focal) and negative for pancytokeratin, S100, Melan-A103, and smooth muscle actin (SMA).
HMB-45 and Melan-A103 are positive in EAML and negative in renal cell carcinomas. Thus, this specimen’s histopathologic and cytologic features and the tumor marker profile were consistent with the diagnosis of EAML. However, other than extrarenal fat tissue, some components (such as smooth muscle and mature adipose tissue) were not readily observed in the slides provided.[1,2]
DR. LAM: How is the EAML variant different from classic benign angiomyolipomas? Are there other factors that may predict for malignant transformation?
DR. LA ROSA: EAML consists of a predominant, exclusive population of pleomorphic epithelioid cells. Some reports have described these tumors as “atypical angiomyolipomas.” Strict criteria for diagnosing malignant EAML on morphologic grounds have not been established; however, a potential for malignant behavior should be anticipated for EAMLs that are highly pleomorphic, that are mitotically active, and that contain areas of hemorrhage and necrosis—as in this case.
DR. L. MICHAEL GLODÉ, medical oncologist: How is hepatic angiomyolipoma (AML) different from renal AML? Can you tell in this case whether these lesions developed de novo in the liver, or whether they represented true metastasis from his renal epithelioid AML?
DR. LA ROSA: Because these tumors are so rare, and the majority of AMLs are of a benign nature, it is difficult to answer this question with absolute certainty. Some case reports mention the simultaneous presence of hepatic and renal EAML, and it is not very clear whether these lesions represent just a multifocal appearance of the AML tumors or real metastases. In our case, the hepatic tumor was discovered several months after the renal tumor, which strongly suggests a metastatic nature for the hepatic tumor. In addition, many case reports mention the presence of real metastatic liver and lung tumors several months after resection of the primary renal tumor.
DR. LAM: Dr. Suby-Long, what are the sensitivities of abdominal ultrasound and CT for surveillance of patients with resected renal lesions?
DR. THOMAS SUBY-LONG, radiologist: The sensitivities of both abdominal ultrasound and CT are high for surveillance. Both have advantages and disadvantages. The primary advantage of ultrasound over CT is the lack of ionizing radiation. The primary disadvantage of ultrasound is that it is operator-dependent.
DR. LAM: Would you review the patient’s radiographic findings?
DR. SUBY-LONG: Review of the initial outside CT revealed an 18-cm right renal mass with extensive heterogeneity and no demonstrable macroscopic fat. There were no liver lesions on the initial CT (Figure 3A). An ultrasound obtained within 10 days of the initial CT revealed an 18-cm solid, heterogeneous, moderately vascular mass with central necrosis (Figure 3B).
DR. LAM: Are there characteristic abnormalities seen with mesenchymal tumors such as AML? How does an AML differ from a primary renal cell or hepatocellular carcinoma?
DR. SUBY-LONG: Typically AMLs, whether in the kidney or the liver, are markedly echogenic on ultrasound and demonstrate macroscopic fat density (-30 to -100 HU) on CT. Approximately 5% of AMLs do not have detectable fat on CT and so cannot be distinguished from other solid masses, such as renal cell or hepatocellular carcinoma.
DR. LAM: Are there radiographic characteristics that would predict the risk for rupture of an AML lesion?
DR. SUBY-LONG: Size, multifocality, and vascular abnormalities are the main risk factors for bleeding. Typically we treat AMLs that are symptomatic or larger than 4 cm. Among AMLs of 4 cm or larger, 46% to 94% are symptomatic and 50% to 60% bleed spontaneously.