It has now been more than one-quarter of a century since the concept of antiangiogenic therapy as a treatment for cancer was proposed.1 Although it can be debated whether this approach has revolutionized cancer therapy, it has clearly influenced therapy and practice for patients with a variety of cancers. One notable example is hepatocellular carcinoma (HCC), which is a highly vascularized tumor that needs an intense angiogenic activity to develop and progress. In fact, one of the most effective therapies for HCC palliative therapy is chemoembolization, which can be seen as the ultimate antiangiogenic therapy. Recent studies have also delineated a beneficial effect of sorafenib in patients with HCC.2, 3 Sorafenib is a receptor tyrosine kinase inhibitor with multiple targets on multiple cell types. Signaling pathways affected by sorafenib include raf, platelet-derived growth factor (PDGF), c-kit, and vascular endothelial growth factor (VEGF), but it is unknown which of these is the dominant mechanistic target for benefit in hepatocellular cancer. Likely, the efficacy is related to a balanced action on several cell systems that results in a safe and effective therapeutic intervention. It is important to be aware of the potential relevance of this fine balance because drugs with apparent pharmacologic similarities may have a different profile in vivo, both in terms of antitumoral efficacy and in terms of safety.
Receptor tyrosine kinase inhibitors have also begun to receive greater attention as a potential therapy in the treatment of portal hypertension and cirrhosis. Indeed, in preclinical studies, a number of receptor tyrosine kinase inhibitors, including imatinib and sunitinib, and now sorafenib, have been shown to improve portal hypertension.4, 5 The mechanistic basis for benefit is not fully elucidated; however, there is increasing and compelling evidence suggesting an intimate link between angiogenesis, fibrosis, and portal hypertension and, thus, it has been proposed that a part of the mechanism of action of these agents may be through an antiangiogenic mechanism.6–9 In the current issue of HEPATOLOGY, the article by Mejias et al.10 demonstrates a beneficial effect of the receptor tyrosine kinase inhibitor, sorafenib, on splanchnic, intrahepatic, and portocollateral circulations in cirrhotic rats with portal hypertension; these observations are reminiscent of recent work with sunitinib in this model.4 This work is exciting in many ways, especially given that sorafenib is presently approved and used in patients with existing cirrhosis and HCC. This highlights the opportunities that are available to test this agent in humans with cirrhosis and portal hypertension. In the study by Mejias and colleagues, sorafenib was administered orally once a day for 2 weeks in rats with cirrhosis and portal hypertension, and it was shown to inhibit signaling through VEGF targeted at endothelial cells, PDGF targeted at hepatic stellate cells, smooth muscle cells, and pericytes and RAF. Interestingly, the effects were observed not only in the intrahepatic circulation, but also in the systemic and collateral circulations, suggesting this therapy may have multiple benefits in portal hypertension. Indeed, the researchers observed a 25% reduction in portal pressure and a prominent improvement in liver injury, fibrosis, and angiogenesis.
These results raise the compelling question of whether angiogenic therapy has an adequate rationale to be considered for therapy in patients with cirrhosis and portal hypertension. Certainly, there are some histologic similarities in the sinusoidal and parenchymal structures in the liver in cirrhosis and cancer. For example, as seen in Fig. 1, many would be hard pressed to distinguish whether this micrograph depicts a cirrhotic liver or a liver with metastasis and desmoplastic reaction. Both conditions evidence a prominent stroma with myofibroblast activation and prominent changes in the sinusoidal endothelial cells and vasculature with increased “scar vessels” traversing through the dense matrix network. So the question arises whether antiangiogenesis therapy is ready for prime time evaluation in patients with cirrhosis and portal hypertension, especially with clinical evidence that patients with cirrhosis can receive sorafenib without severe hepatic decompensation and that this agent has prominent beneficial effects in experimental models of cirrhosis and portal hypertension. Certainly, some of the analysis from histologic sections of tissue obtained during the cancer trials of sorafenib may provide some clues, especially if some antifibrotic effect can be demonstrated in the fibrotic tissues adjacent to the tumor.
Evaluation of the potential efficacy of these drugs in cirrhosis will need a very careful approach. Short-term studies surely are underway, but there are several issues that will need detailed assessment. First is the question of optimal dosage to maintain efficacy (assuming there is efficacy) and to what extent the chronic administration will be tolerable and safe. It has to be stressed that the dose in patients with cancer was defined according to maximal tolerability; likely the dose needed to properly modulate fibrosis and hepatic hemodynamics may be less. Presumably, the tolerance to side effects in patients without cancer will be less than in patients with advanced malignancy. Sorafenib has quite a safe profile with manageable toxicity as compared to conventional chemotherapy, but even so, the adverse effects are significant and induce treatment interruptions and dose reduction in more than 40% of the patients.2, 3 A major aspect to take into account is that angiogenesis is a critical aspect of cirrhosis and cancer, but at the same time this is of paramount importance in several other pathophysiological processes needing indemnity of vessels. In that sense, one of the major complications of antiangiogenic therapy in patients with liver cancer has been variceal bleeding.11, 12 Although in the SHARP (Sorafenib HCC Assessment Randomized Protocol) sorafenib trial the rate of bleeding due to varices was not different between treated and untreated patients, such a complication has been a major concern in patients treated with agents such as bevacizumab and sunitinib.11, 12 It may be hypothesized that the vasa vasorum needed to maintain the structure of varices may be damaged if the antiangiogenic effect is too intense, and this emphasizes the previous comment about the delicate balance of pathway inhibition that should be in place. The impact on cardiac function is also of major interest because myocardial pathobiology due to the hyperdynamic circulation may be further impaired by agents blocking several key mechanisms necessary to maintain cardiac tissue physiology.13
As a whole, it is obvious that a new avenue for pharmacologic intervention in patients with cirrhosis has emerged. The increasing knowledge of molecular biology has allowed the identification of new therapeutic targets to modulate cell biology, and this will affect the management of both malignant and nonmalignant diseases. In the specific case of cirrhosis and liver cancer, it may be that the same agent or family of agents will be used for the prevention of fibrosis progression leading to cirrhosis, which at a point is the main risk factor for liver cancer development, and for the treatment of liver cancer itself, when this unfortunate development takes place.
In summary, all these questions will eventually need to be addressed through different clinical studies with the proper design and primary endpoints. The challenge is there and it is time to move ahead.