Erlotinib in Treating Patients With a History of Stage I, Stage II, or Stage III Colorectal Cancer or Adenoma

OBJECTIVES:

Primary

- To determine if erlotinib hydrochloride at low dose (25 mg) will decrease aberrant
crypt foci (ACF) pERK levels in patients with curatively treated colorectal cancer or
adenoma.

Secondary

- To determine if additional EGF-inducible biomarkers (i.e., phosphorylated ERBB2,
phosphorylated AKT, total EGFR, Ki-67, and cyclin D1) will decrease from baseline
(pre-) to post-treatment in patients treated with erlotinib hydrochloride at different
dose levels.

- To determine the mean decrease from baseline of the ACF to normal mucosa pERK ratio
from pre- to 30 days post-treatment.

- To determine erlotinib hydrochloride concentration in plasma and colorectal tissue at
different dose levels after 30 days of therapy.

- To determine the incidence of rash, diarrhea, and other side effects in patients
treated with low-dose erlotinib hydrochloride.

OUTLINE: This is a multicenter study. Patients are stratified according to NSAID use (≥ 10
days/month vs < 10 days/month). Patients are randomized to 1 of 3 treatment arms.

- Arm I: Patients receive 100 mg of oral erlotinib hydrochloride and oral placebo once
daily.

- Arm II: Patients receive 50 mg of oral erlotinib hydrochloride and oral placebo once
daily.

- Arm III: Patients receive 25 mg of oral erlotinib hydrochloride and oral placebo once
daily.

In all arms, treatment continues for 30 days in the absence of disease progression or
unacceptable toxicity.

Biopsies are collected at baseline and post-treatment. Western blot analysis is used for
quantification of pERK and for protein biomarkers. Blood samples are collected and examined
for erlotinib hydrochloride levels.

After completion of study therapy, patients are followed for 4 to 9 weeks.

DISEASE CHARACTERISTICS:

- Diagnosis of stage I-III colorectal cancer (CRC) or adenoma meeting all of the
following criteria:

- Measuring ≥ 1 cm

- Curatively treated disease

- Must have ≥ 4 aberrant crypt foci on baseline colonoscopy

- No active CRC

PATIENT CHARACTERISTICS:

- ECOG performance status 0-1

- WBC > 3,000/mm³

- Platelet count > 100,000/mm³

- Hemoglobin > 10 g/dL

- Plasma creatinine < 1.6 mg/dL

- Total bilirubin < 1.5 times upper limit of normal (ULN)

- Serum AST and ALT < 1.5 times ULN

- Not pregnant or nursing

- Negative pregnancy test

- Fertile patients must use effective contraception

- No history of inflammatory bowel disease

- No active keratoconjunctivitis within the past 3 weeks

- No history of interstitial lung disease or chronic lung disease, or active smoking

- No patient with increased bleeding risk from biopsy protocol (i.e., renal failure,
decompensated cirrhosis, or blood dyscrasias)

- No uncontrollable diarrhea of any cause

- No history of sensitivity to EGFR antagonists

- No history of sensitivity to erlotinib hydrochloride, Iressa, or Erbitux (e.g., rash
uncontrollable by topical steroids or antibiotics)

- No history of active smoking

- No medical or psychosocial condition that would preclude study participation

PRIOR CONCURRENT THERAPY:

- See Disease Characteristics

- No chemotherapy for CRC within the past 6 months

- No prior rectal radiation in patients with rectal cancer

- No prior pelvic radiation

- No corneal surgery within the past 3 weeks

- No CRC-related surgery within the past 6 months

- No other concurrent EGFR antagonists

- No other concurrent investigational pharmaceutical agents

- No concurrent, significant CYP3A4 inducers (e.g., phenytoin, carbamazepine, Hypericum
perforatum [St. John's wort], or rifampin)

- No concurrent warfarin

- No concurrent medication known to be inhibitors or metabolized by CYP3A4 (e.g.,
erythromycin, clarithromycin, or ketoconazole)