Bevacizumab protects against sinusoidal obstruction syndrome and does not increase response rate in neoadjuvant XELOX/FOLFOX therapy of colorectal cancer liver metastases

Abstract

Aim

In patients suffering from colorectal cancer liver metastases, 5-fluorouracil-based chemotherapy plus oxaliplatin ensures superior response rates at the cost of hepatic injury. Knowledge about the consequences of bevacizumab on chemotherapy-induced hepatic injury and tumor response is limited.

Methods

Resected liver specimens from patients of two prospective, non-randomized trials (5-fluorouracil/oxaliplatin ± bevacizumab) were analyzed retrospectively. Hepatotoxicity to the non-tumor bearing liver was evaluated for sinusoidal obstruction syndrome, hepatic steatosis and fibrosis. Tumor response under chemotherapy was assessed according to Response Evaluation Criteria in Solid Tumors (RECIST).

Results

Bevacizumab decreased the severity of the sinusoidal obstruction syndrome. Bevacizumab had no impact on hepatic steatosis and fibrosis. The addition of bevacizumab to chemotherapy had no effect on tumor response compared to combination chemotherapy alone.

Conclusions

This analysis shows that bevacizumab protects against the sinusoidal obstruction syndrome and thus provides the histological explanation of the safe use of bevacizumab prior to liver resection. Furthermore, we show that bevacizumab does not improve tumor response according to RECIST.

Introduction

Bevacizumab (Avastin^®), a monoclonal humanized antibody directed against vascular endothelial growth factor (VEGF), is being increasingly used in the treatment of mCRC. In a large randomized phase III clinical trial bevacizumab given in combination with conventional chemotherapy significantly improved survival of patients with mCRC¹ and has been approved by the FDA and EMEA as a first-line therapy for colorectal cancer. Bevacizumab is considered to be only effective in conjunction with cytotoxic drugs and may be given in combination with 5-fluorouracil- based chemotherapy plus oxaliplatin (5-FU/oxaliplatin).²

5-FU/oxaliplatin is being administered prior to liver resection in mCRC patients and prolongs recurrence-free survival.³ The use of chemotherapy prior to liver surgery offers the possibility to convert a patient from unresectable disease to resectable disease and to treat undetected micrometastases.4, 5 Furthermore, as reported recently in a prospective non-randomized trial in our institution, tumor response to 5-FU/oxaliplatin is qualified to predict benefit from liver resection for colorectal liver metastases with respect to recurrence-free survival.⁶ Disadvantages of neoadjuvant chemotherapy are an increased, albeit still low postoperative morbidity rate.³ However, postoperative morbidity after liver resection is associated with the number of chemotherapy cycles administered before surgery, independently of the scheme applied.⁷ Therefore, one of the rationales of combining chemotherapy and bevacizumab is to shorten the length of chemotherapy while offering the same (or even better) quantity of response to the patient. We showed recently in a prospective non-randomized trial that bevacizumab can be administered before liver resection safely and with a low postoperative complication rate.⁸

The neoadjuvant use of oxaliplatin is associated with the occurrence of a distinctive type of hepatic injury, namely the sinusoidal obstruction syndrome.9, 10 For 5-fluorouracil and irinotecan, an association with hepatic steatosis has been shown.¹¹ For this type of hepatic steatosis, which belongs to the family of non-alcoholic fatty liver disease (NAFLD), the term CASH (chemotherapy-associated steatohepatitis) has been introduced.¹² Cytotoxic chemotherapy primarily affects the tumor itself with tumor necrosis and tumor fibrosis as regression correlates.¹³

Bevacizumab neutralizes free VEGF and thus inhibits VEGF-mediated endothelial cell proliferation, survival and migration in vitro.¹⁴ On the other hand, bevacizumab induces apoptosis in hypoxia susceptible tumor cell lines.¹⁵ Bevacizumab is able to normalize tumor vasculature, thereby decreasing intratumoral fluid pressure and increasing delivery of chemotherapy to the tumor cells.¹⁶ Thus, it is a priori not predictable how bevacizumab affects the tumor and the non-tumor bearing parenchyma when combined with chemotherapy. In this retrospective exploratory analysis of two non-randomized prospective trials of our institution,6, 8 we compared non-tumor bearing liver histology and tumor response between patients treated with 5-FU/oxaliplatin and patients treated with 5-FU/oxaliplatin plus bevacizumab. We addressed the following questions: (1) which effect, if any, does bevacizumab have on chemotherapy induced hepatic injury, and furthermore, (2) does bevacizumab have an impact on the radiologic response, according to RECIST?