Identification of predictive factors for response to neo-adjuvant Sunitinib in patients with resectable hepatocellular carcinoma

Hepatocellular carcinoma(HCC) is the fifth most common malignancy in the world.
The incidence of HCC is still rising. The only cure for HCC is surgical
resection or livertransplantation. Unfortunately most patients more than 80%
of the patients present with unresectabel or advanced disease, with the
majority of patients dying within 1 year after diagnosis.

Partial liver resection can be performed in patients with a preserved liver
function. This is often not the case in patients with HCC with cirrhotic
livers. Radiofrequency Ablation (RFA) is a treatment modality which has emerged
as a technique applicable in patients in whom partial liver resection is not
possible or is contraindicated because of an expected high operative risk.
Also, in patients in whom a liver transplantation is scheduled, RFA is used a
bridge to transplantation.

To improve outcome after RFA or partial liver resection neoadjuvant therapie
might be an option. In metastatic HCC tyrosine kinase inhibitors have showed
improvement in overall survival. Sunitinib is an orally administered, small
molecule inhibitor of multiple receptor tyrosine kinases implicated in tumor
growth, angiogenesis, and metastatic progression.

The problem with response assessment in HCC with conventional imaging is to
differentiate between benign or malignant disease, especially in patients with
cirrhosis who have regenerating nodules and fibrosis. With new agents like
tyrosine kinase inhibitors the problem with conventional imaging is that
usually responses involve central necroses that does not involve the outside
borders for some period of time. Therefore in patients with GIST and kidney
cancer who are treated with tyrosine kinase inhibitors response evaluation is
best done using 18F-FDG PET scanning. In contrast with the tumors mentioned
above, in patients with HCC FDG PET scans are positive in only 60% of cases.
This may be related to a more aggressive feature of the tumor compared to
negative scans due to inherent differences or stage of the tumor. Because of
these considerations evaluation of sunitinib response in HCC patients with FDG
PET scans next to conventional imaging seems useful.

Over-expression of VEGF occurs in many human tumor types, also in
hepatocellular carcinoma. VEGF production by the tumour will result in abundant
presence of VEGF in the micro-environment of the tumour. Non-invasive
assessment of VEGF-levels in the micro-environment of the tumour could guide in
the development and follow up of anti-VEGF targeted therapy. Currently, we are
evaluating the pharmacokinetics, biodistribution and tumour uptake of
111In-bevacizumab in patients. If scanning with 111In-bevacizumab leads to
adequate imaging of tumours and assessment of VEGF levels in the tumour, this
technique may be used to assess tumour response in the future of
anti-angiogenic therapy.

To identify predictive factors for response to sunintib treatment in patients
with HCC. By discovering changes in pathways involved in HCC carcinogenesis due
to sunitinib treatment in tissue/biomarkers and to correlate these changes to
FDG-PET scan and 111In- bevacizumab scan results before and after sunitinib
treatment.

Patients who are planned for RFA or partial liver resection will be treated
neo-adjuvant with sunitinib.

Before start and between day 18-21 a FDG-PET scan will be performed.

Four weeks before surgical resection, patients receive an i.v. injection of
111In-bevacizumab. Gamma-camera imaging will be performed on day 0, 2 , 4 and 7
post injection (The optimal time to scan the patients will be determined after
two or three patients. Hereafter patients will only be scanned once or twice
after tracer injection, most likely on day 0 and 4 after tracer injection.)
After the first series of scans, patients will be treated with sunitinib One
week before RFA or surgical resection, patients receive another dose
111In-bevacizumab (100mBq). Gamma-camera imaging will be performed on day 0, 2,
4 and 7 post injection (The optimal time to scan the patients will be
determined after two or three patients. Hereafter patients will only be scanned
once or twice after tracer injection, most likely on day 0 and 4 after tracer
injection) and before surgical resection.

After neoadjuvant treatment with sunitinib, the tumor is biopsied either during
open surgery (resection or RFA) or during CT guided percutaneous RFA.

The tumour activity, anti-angiogenic and apoptotic response of sunitinib will
be determined by evaluating histological, haematological parameters and FDG-
PET and 111In- bevacizumab respons.

Patients will be treated with a daily dose of 37.5 mg sunitinib for 21 days
until 48 hours before RFA or partial liver resection.

Any toxicity will be scored according to the Common Toxicity Criteria version
3.0.
Sunitinib
Sunitinib administration will start 23 days before partial liver resection or
RFA and will be cessitated 2 days before partial liver resection or RFA. In
case of grade 3/4 toxicity treatment is withheld until grade 2 toxicity. Dose
reduction of 12.5 mg is recommended based on individual safety and
tolerability.A dose reduction for sunitinib should be considered if sunitinib
must be co-administered with a strong CYP3A4 inhibitor. Although Sunitinib is a
FDA/EMEA approved compound for metastatic renal cell carcinoma and GIST its use
in combination with RFA has never been tested. However major operations seem to
be safe when performed 48 hours after cessitation of Sunitinib. However since
we do not know the exact effect of Sunitinib on wound healing and liver
regeneration in case of partial liver resection or RFA procedure we have chosen
a cautious study design and will include only one patient at a time in the
study.

Radiolabeled 111In-bevacizumab
In this study bevacizumab is administered in a tracer-dose. The risk of any
additional side effects by the administration of a tracer dose (10 mg) will be
very low, compared to therapeutic administration of bevacizumab (375-600 mg).
Administration of 111In-bevacizumab entails radiation load from the
participating patients. It has been calculated that a dose of 100 Mbq will lead
to a radiation load of 18mSv. In this study protocol this will lead to 36 mSv.
For comparison, a CT-abdomen will lead to a radiation dose of 10-15 mSv. The
poor prognosis of thos patient group ( 5 year survival: in case of RFA before
liver transplantation 70%, in case liver transplantation is not possible, but
only RFA or partial liver resection 40 %) and the potential new information
given by the study makes this radiation loas acceptable.

FDG PET scan
There will be an additional radiation dose of 7 mSv due to the FDG PET scan.

Liver biopsy
The risk of performing a biopsy of an HCC is considered to be small, with an
overall reported complication rate of less than 0.2%. Potential risks are
bleeding, arterioportal shunts and needle tract seeding. The risk of needle
tract seeding is estimated to be between 0.6 and 5.1%. In two studies no needle
tract seedings were found in 433 and 101 tumor biopsies respectively. The use
of a guiding needle introducer (as is current standard in the UMCG) which is
not in contact with the tumor itself protects the tract from tumor cell
deposition, even if the cutting needle (which enters the tumor) contains tumor
cells at its outer side.