A New Strategy in the War on Renal Cell Cancer: Title and subTitle BreakHitting Multiple Targets With Limited Collateral Damage

An estimated 38 890 individuals will be diagnosed as having renal cell carcinoma (RCC) in the United States in 2006 and approximately 12 840 patients will die from the disease.¹ RCC is the most common malignant lesion of the kidney and accounts for 85% of all renal neoplasms and 3% of all adult malignancies.² The overall incidence of RCC has increased over the past 20 years from 2% to 4% per year.³ While smoking, hypertension, and obesity have all been associated with RCC risk, smoking has emerged as the most significant risk factor. Cigarette smokers have double the risk of RCC than nonsmokers and pipe and cigar smokers also have an increased risk of the disease.³ RCC occurs most often in individuals aged 50 to 70 years and affects men almost twice as often as women.⁴ Many patients with this cancer are being diagnosed at an earlier stage due to the increased use of diagnostic testing such as computed tomography and magnetic resonance imaging. In fact, approximately half of RCC cases are now detected because a renal mass is incidentally identified during a radiology examination.⁵ Yet, one fourth of patients present with advanced disease, including locally invasive or metastatic RCC, and one third of patients with resected localized disease will have a recurrence.⁶

While hormonal manipulations, chemotherapeutic agents, and targeted therapies have improved the survival of patients with several common forms of metastatic cancer, RCC has remained stubbornly refractory to these agents.² Response rates to chemotherapy have rarely exceeded 6%.⁷ Higher response rates (>10%-15%) have only been achieved with cytokines such as interleukin 2 and interferon alfa.^{8 - 9} However, these responses are often short-lived and less than 10% of patients receiving high-dose interleukin 2 treatment achieve long-term disease-free survival.⁸ This underscores the need for novel and more effective therapies for this common disease.

Recognition of RCC as a distinct entity dates back to the 19th century. In 1883, in his doctoral thesis titled “Die Sogenannten Lipome der Niere” (“The So Called Lipomas of the Kidney”), Paul Grawitz provided the first thorough macroscopic and microscopic description of the disease, which had been until then considered a form of lipoma because of its similar color.¹⁰ First, he described the lesions as “pea-sized knots, which rarely exceed the diameter of a cherry but can on rare occasions be rounded tumors up to 4 to 5 cm in size.” He emphasized that “the margins of the tumor are usually sharply defined and confined within the renal capsule” and that “the color of these tumors is closer to that of the white matter of the brain than that of lipomas, which always have some degree of yellow color.” In contrast to the firm consistence of lipomas, he described these tumors as soft and brittle. His detailed drawings of the tumor's constitutive cells in sheets, cords, or as papillary fronds support his contention that they are of epithelial origin. Grawitz was the first author to point out the surprisingly abundant vasculature of these tumors, a feature not seen in lipomas and rarely seen in other tumors.¹⁰ Last but not least, he emphasized the tumor's aggressive nature through the description of “man-head size tumor extending into a patient's pelvis.” His description of tumor-associated lung metastasis was further proof of this tumor's malignant potential.

Although most cases of RCC seem to occur sporadically, there is evidence that 1% to 4% of cases are due to an inherited predisposition.¹¹ The von Hippel-Lindau (VHL) tumor-suppressor gene was the first gene identified that causes a proportion of hereditary RCC cases.¹² This syndrome is named after Eugen von Hippel, a German ophthalmologist who reported the occurrence of hereditary retinal angiomas,¹³ and Arvid Lindau, an ophthalmologist from Sweden who determined that retinal angiomas and hemangioblastomas of the central nervous system belong to the same familial syndrome.¹⁴ Melmon and Rosen¹⁵ subsequently established that RCC is a clinical manifestation of the VHL syndrome. The manifestations of the syndrome include bilateral renal carcinoma and cysts, retinal angiomas, hemangioblastomas of the cerebellum and spinal cord, pancreatic cysts and islet cell tumors, and epididymal cystadenomas. A subset of VHL families also has an increased risk of pheochromocytoma. VHL syndrome affects approximately 1 in 36 000 live births and is transmitted in an autosomal dominant manner with a penetrance of 80% to 90%, but with variable expressivity.¹⁶ In this syndrome, RCC develops when the remaining normal VHL allele becomes inactivated. Recent studies suggest that more than 60% of sporadic RCCs arise because of somatically acquired mutation or methylation of the VHL gene,¹⁷ thus establishing a tight link between the rare VHL syndrome and this relatively common malignancy.

Pheochromocytomas, hemangioblastomas, and clear cell renal carcinomas occasionally secrete erythropoietin, which leads to uncontrolled production of red blood cells (polycythemia).¹⁸ In addition, hemangioblastomas and renal carcinomas are both highly vascular tumors, a feature partly attributable to increased production of vascular endothelial growth factor. Hypoxia induces the secretion of both erythropoietin and vascular endothelial growth factor, which work in concert to counteract the negative effects of low circulating oxygen levels. These observations led to the conclusion that VHL-associated tumors behave as if they lack oxygen. They further led to the hypothesis that VHL might affect oxygen sensing pathways.^{19 - 20}

In support of this concept, several groups of investigators demonstrated that VHL-defective tumor cells exhibit hypoxic features, regardless of oxygen availability.²¹ The search for hypoxia-related messengers led to the identification of the hypoxia-inducible factor. A link between hypoxia-inducible factor and the VHL gene was formerly established in 1999 with the discovery that VHL-deficient cells activate hypoxia-inducible factor.²² Activated hypoxia-inducible factor stimulates the transcription of several genes involved in blood vessel growth and oxygen delivery, such as vascular endothelial growth factor and platelet-derived growth factor. More than a century after Grawitz’ observations, these findings have provided a rational explanation for the highly vascular nature of RCC and highlighted potential therapeutic targets.

In this issue of JAMA, Motzer et al²³ report the results of a multicenter open-label study assessing the effects of sunitinib, an orally administered small molecule inhibitor of the tyrosine kinase portion of the vascular endothelial growth factor and platelet-derived growth factor receptors. This trial represents another example of rational cancer therapy based on cancer-specific molecular alterations. Objective responses were observed in 34% of patients. Given the dismal track record of chemotherapy in the treatment of RCC, this is nothing short of remarkable. These results extend and confirm the results of a previous phase 2 study conducted on a smaller number of patients and establish sunitinib as a bona fide therapeutic agent in this disease.²⁴ In January 2006, the US Food and Drug Administration granted approval for sunitinib for the treatment of advanced RCC based on the partial response rates and duration of responses presented in these reports.^{23 - 24} It is of concern that the investigator assessment of the response rates was nearly 10% higher than the response rate assessed by an independent third-party core imaging laboratory. This discrepancy underscores the need for centralized and independent review of radiological images in clinical trials.

Targeting multiple kinases comes at a cost. Eight patients experienced a decline in left ventricular ejection fraction during treatment. Five of them had a decline greater than 20% from baseline and these changes may not be reversible. Furthermore, 40% of the patients developed neutropenia, 28% experienced grade 2 or grade 3 fatigue, and 16% developed hypertension during treatment. Hence, hitting multiple targets does not happen without collateral damage. Postapproval trials will be needed to demonstrate clinical benefit, such as increased survival or improvement in disease-related symptoms. The short median progression-free survival (8.3 months) highlights the fact that sunitinib is far from a magic bullet. Nonetheless, it represents a new class of drug with a promising future for the treatment of this deadly disease. In the war on cancer, it is a small victory against one of the most ferocious enemies.