We planned this study to investigate the efficacy and safety of XELOX (capecitabine and oxaliplatin) plus lapatinib treatment in HER2-positive gastric cancer patients with liver metastasis.

Gastric cancer is the leading cause of cancer death worldwide with the incidence of 18.9/100,000 per year and the mortality rate of 14.7/100,000 per year [1] and is the most common malignancy in Korea[2]. Metastatic gastric cancer remains a therapeutic challenge for medical oncologists due to poor prognosis. Several randomized phase III trials comparing combination chemotherapy such as 5-fluorouracil (5-FU), doxorubicin, and mitomycin (FAM), or 5-FU, doxorubicin, and high-dose methotrexate (FAMTX) with best supportive care have demonstrated significantly prolonged overall survival (8 - 10 months) for chemotherapy group as compared to best supportive care alone (3 - 5 months)[3, 4].

In quest of a novel therapeutic target for gastric cancer, HER2 overexpression has been tested and was reported in 6-35% of stomach and gastroesophageal tumors [5]. Trastuzumab, a humanized monoclonal antibody which selectively targets HER2, has shown survival benefit in patients with HER2(+) metastatic breast cancer [6-8]. The ToGA trial is the first randomized, prospective, multicenter, phase III trial to study the efficacy and safety of trastuzumab in HER2(+) GC [9]. Of 3,807 tumor samples screened for Her2 status, 22.1% were Her2 positive and 594 patients were randomized to receive chemotherapy alone or chemotherapy + trastuzumab.

The ToGA trial demonstrated a significant survival benefit in the transtuzumab +chemotherapy when compared with chemotherapy alone arm: 13.5 vs. 11.1 months, respectively (p=0.0048; HR 0.74; 95% CI 0.60, 0.91). ORR was 47.3% in the trastuzumab + 5-FU/CDDP (or capecitabine/CDDP) arm and 34.5% in the chemotherapy alone arm (p=0.0017). The ToGA trial is the first phase III trial to demonstrate survival benefit from molecularly targeted agent in gastric cancer.

Of note, 70 - 80% of patients HER2 overexpressing breast cancer do not respond to trastuzumab due to either primary or acquired resistance. One of the important mechanisms for trastuzumab resistance is the accumulation of truncated forms of the HER2 receptor which lack the extracellular trastuzumab-binding domain (Figure 2). P95HER2, an amino terminally truncated carboxyl terminal fragments of HER2, is frequently found in HER2(+) breast cancer cell lines and tumor specimens (~20%)[11]. Intriguingly, recent study showed that p95HER2 (+) breast cancer cells were resistant to trastuzumab but remained sensitive to the antiproliferative effects of the tyrosine kinase inhibitor lapatinib, both in vitro and in vivo[11]. In addition, patients with p95HER2(+)breast cancer were resistant to trastuzumab with significantly lower response rate when compared with full-length HER2(+) breast cancer (11.1% vs 51.4%, respectively; P = 0.029).

We have surveyed the incidence of p95HER2 expression in fresh frozen tissues from gastric cancer and found that > 60% of HER2 (3+) GC demonstrated p95HER2. Hence, the role of lapatinib may be more promising than trastuzumab in GC HER2(+) patients with truncation.

Inclusion Criteria:

1. Histologically proven gastric cancer with metastatic lesion(s) that is (are) unresectable

   • locally advanced gastric cancer that are NOT resectable
   • distant metastases limited to abdominal lymph node, liver only :Patients with liver metastasis : Number of liver metastasis between 2 and 5 or maximal diameter should be under 5 cm (2 = liver mets = 5 or maximal diameter = 5cm) No LN metastasis within group 3 and no bulky N2 metastasis Clinically no distant metastasis (lung metastasis, mediastinal LN metastasis, neck LN metastasis, bone metastasis, brain metastasis, and peritoneal seeding in abdominal and pelvis CT; in cases of suspicious peritoneal seeding in imaging without any evidence of ascites and/or peritoneal enhancement will be allowed to enter the study based on investigators' decision)
   • chemo-naïve (adjuvant treatment will be allowed if the last date of treatment is ≥ 6 months from the study entry date

2. Age ≥ 18

3. ECOG performance 0 - 1

4. Adequate organ function (AST and ALT ≤2x upper limit of normal, bilirubin ≤1.5 x upper limit of normal, and creatinine < 1.5x upper limit of normal, platelet > 100,000/ul, absolute neutrophil count ≥ 1,500/ul)

5. At least one measurable lesion by RECIST 1.1 criteria

6. HER 2 (+) by HercepTest(IHC 3+ alone, or IHC 2+ with FISH amplification)

7. Written informed consent

Exclusion Criteria:

1. Prior therapy for metastatic disease
2. Pregnant or lactating women
3. Uncontrolled medical illnesses including medically uncontrolled infection, uncontrolled hypertension, unstable angina, symptomatic congestive heart failure, myocardial infarction within 6 months
4. Any comorbidities which are not suitable for general anesthesia and surgical resection
5. Distant metastases other than liver or abdominal lymph nodes (As outlined in inclusion criteria, and peritoneal seeding in abdominal and pelvis CT; in cases of suspicious peritoneal seeding in imaging without any evidence of ascites and/or peritoneal enhancement will be allowed to enter the study based on investigators' decision)
6. Known immediate or delayed hypersensitivity reaction to lapatinib ,capecitabine, oxaliplatin or any other platinum compounds, any recipients.

8) Subjects with DPD deficiency 9) Subjects who have current active hepatic or biliary disease (with exception of patients with Gilbert's syndrome, asymptomatic gallstones, liver metastases or stable chronic liver disease per investigator assessment) 10) Pre-existing hand and foot syndrome and peripheral neuropathy of grade 2 or greater 11) Subjects unsuitable to resection or general anesthesia