Phase II Study of Erlotinib (OSI-774) in Patients With Advanced Hepatocellular Cancer
http://www.100md.com
《临床肿瘤学》
the Karmanos Cancer Institute, Wayne State University, Detroit, MI
University of Wisconsin Comprehensive Cancer Center, Madison, WI
Johns Hopkins University, Baltimore, MD
Mayo Clinic Jacksonville, Jacksonville, FL
Mayo Clinic Scottsdale, Scottsdale, AZ
Washington University School of Medicine, St Louis, MO
Mayo Clinic, Rochester, MN
ABSTRACT
PURPOSE: Epidermal growth factor receptor/human epidermal growth factor receptor 1 (EGFR/HER1) and ligand expression is frequently seen in hepatocellular cancers (HCCs). Erlotinib (Tarceva, OSI-774; OSI Pharmaceuticals, Melville, NY) is a receptor tyrosine kinase inhibitor with specificity for the EGFR/HER1.
METHODS: The primary objective of this study was to determine the proportion of patients with advanced HCC who were progression-free at 6 months. Patients with either unresectable or metastatic disease were studied. Only one prior systemic or locoregional therapy was allowed. Erlotinib was given continuously at a dose of 150 mg per day orally.
RESULTS: Thirty-eight patients with HCC were enrolled. Median age of the patients was 69 years (range, 27 to 83 years). A majority of patients (63%) had an Eastern Cooperative Oncology Group performance status of 1. Forty-seven percent of patients had received prior chemotherapy for advanced HCC. EGFR/HER1 expression was detected in 88% of the patients. Median number of cycles per patient was two (range, 1 to 26). Twelve (32%; CI 95%, 18 to 49) of the 38 patients with HCC were progression-free at 6 months. Three patients had partial radiologic responses of duration of 2, 10, and 11 months, respectively. Disease control was seen in 59% of the patients. Median overall survival time was 13 months. Ten patients (26%) had toxicity-related dose reductions of erlotinib. Grade 3/4 skin toxicity or diarrhea was encountered in five and three patients, respectively.
CONCLUSION: Results of this trial suggest a benefit for EGFR/HER1 blockade with erlotinib in patients with HCC manifested by disease control. Additional studies with erlotinib as a single agent or in combination with other agents are warranted.
INTRODUCTION
Hepatocellular cancer (HCC) is the most common primary neoplasm of the liver, accounting for almost half a million deaths annually worldwide. Although the incidence of HCC is particularly high in parts of Asia and Africa, recent studies have documented a clear rise in the number of cases in Japan, Western Europe, and the United States, largely because of the increase in hepatitis C–related liver disease.1 Unfortunately, most patients with HCC are not candidates for any curative treatments because of advanced disease at presentation and/or background chronic liver disease. Cytotoxic therapy of HCC has also been disappointing, without a significant impact on the natural history of this disease.2-8
The epidermal growth factor receptor/human epidermal growth factor receptor 1 (EGFR/HER1) and its ligands EGF and transforming growth factor-alpha (TGF-) are important in cell proliferation, as well as motility, adhesion, invasion, survival, and angiogenesis.9,10 It has been suggested that hypomethylation of the EGF receptor gene may be associated with the development of HCC.11 Studies have indicated that EGFR/HER1 is actively expressed in human hepatoma cells and that EGF may be one of the mitogens needed for the growth of hepatoma cell lines.12 In one study, expression of EGFR/HER1 was present in nine of 11 HCCs. Additional data indicate that EGF, the ligand for EGFR/HER1, may have different effects on different human hepatoma cell lines, and its role might be more important in poorly differentiated hepatoma cells than in the well-differentiated ones.13 Studies have supported the theory of an autocrine, paracrine, and endocrine mechanism of TGF- and EGFR/HER1 on the proliferation of human HCC.14 Furthermore, the detection of greater staining in more highly differentiated portions of the tumors suggested that increased expression of TGF- and EGFR/HER1 were linked to the early stages of human hepatocarcinogenesis.15
Erlotinib (Tarceva, OSI-774; OSI Pharmaceuticals, Melville, NY) is an orally active, potent, selective inhibitor of the EGFR/HER1-related tyrosine kinase enzyme. Erlotinib inhibits the tyrosine kinase activity with an IC50 of 2 nmol/L in an in vitro enzyme assay and reduces EGFR/HER1 autophosphorylation in intact tumor cells with an IC50 of 20 nmol/L. Erlotinib inhibits EGF-dependent proliferation of cancer cells at submicromolar concentrations and blocks cell-cycle progression in the G1 phase.
The lack of any clinically useful drug therapy for patients with advanced HCC and the suggestion of an antiproliferative and proapoptotic role for EGFR/HER1 blockade prompted this pilot trial in patients with HCC. We hypothesized that using EGFR/HER1 blockade in patients with HCC by erlotinib will reduce the growth of these tumors and delay the progression of the disease. The primary objective of this study was to determine the progression-free survival at 6 months of patients with HCC treated with erlotinib 150 mg per day. Secondary end points included objective response frequency and duration, adverse events rates, time to disease progression, and overall survival. We also evaluated correlation of EGFR/HER1 status with patient characteristics and outcome.
METHODS
Eligibility Criteria
Patients eligible for the study had histologically or cytologically confirmed HCC that was surgically unresectable or metastatic. A hypervascular mass on computed tomography (CT) with a serum alpha-fetoprotein level of more than 100 ng/mL was sufficient noninvasive diagnostic criteria for HCC. Measurable disease was required. Other eligibility criteria included an Eastern Cooperative Oncology Group performance status (PS) of 0, 1, or 2; an estimated life expectancy of at least 3 months; adequate bone marrow, hepatic, and renal function, indicated by an absolute neutrophil count of 1,500/μL, platelets of 75,000/μL, and total bilirubin of 2x upper limit of normal, respectively; serum AST or ALT levels of 3x upper limit of normal; serum creatinine of 2 mg/dL; serum albumin concentration of 2.5 g/dL; an international normalized ratio of 1.5 (unless on anticoagulation); and a Child-Pugh classification of A or B. For patients having prior cryotherapy, radiofrequency ablation, ethanol injection, or photodynamic therapy, the following criteria were required: more than 6 weeks had elapsed since that therapy; indicator lesion(s) was/were outside the area of prior treatment; or, if the only indicator lesion was inside the prior treatment area, there must have been clear evidence of disease progression associated with that lesion. In addition, edges of the indicator lesion must have been clearly distinct on CT scanning. Patients with more than one prior systemic anticancer therapy were excluded. Other exclusion criteria included a history of other malignancy within the previous 3 years (except for adequately treated basal cell cancer, squamous cell skin cancer, or cervical cancer). Patients were also excluded on the basis of known abnormalities of the cornea, CNS metastases, and HIV infection.
Erlotinib Administration
All patients received erlotinib 150 mg orally once daily. Treatment was given on a continuous daily basis, and a treatment cycle was 28 days. The erlotinib dose was reduced according to prestudy-defined toxicity criteria to 100 or 50 mg per day. No erlotinib dose re-escalation was allowed after dose reduction resulting from toxicity. Patients requiring dose reductions to less than 50 mg per day were taken off of the study. Treatment was held up to 14 days for grade 3 or 4 toxicity until resolution of toxicity to grade 1 or higher. Erlotinib was then reinitiated at a reduced dose. Any of the following were allowed for the treatment of acneiform skin rash: minocycline, topical tetracycline, topical clindamycin, topical silver sulfadiazine, diphenhydramine, and oral prednisone.
Disease Assessment
Objective response to therapy was assessed by using the Response Evaluation Criteria in Solid Tumors criteria.16 Measurable disease was defined as at least one lesion whose longest diameter (LD) could be measured accurately as 2.0 cm. Lesions detected by the physical examination were only considered measurable when they were superficial (eg, skin nodules, palpable lymph nodes). Lesions on chest x-ray were considered measurable lesions if clearly defined and surrounded by aerated lung. However, CT or magnetic resonance imaging evaluation was the measurement of choice for such lesions. All other lesions (or sites of disease), including small lesions (LD, < 2.0 cm) were considered nonmeasurable lesions.
All measurable lesions ( 2.0 cm) up to a maximum of 10 lesions representative of all involved organs were identified as target lesions and recorded at baseline. A sum of the LD for all target lesions were calculated and reported as the baseline sum LD. The baseline sum LD was used as reference to characterize the objective tumor response to erlotinib. All nonmeasurable lesions were identified as nontarget lesions and were also recorded at baseline. Measurements were not required for these lesions, and they were followed as "present" or "absent."
All identified sites of disease at baseline were followed on re-evaluation. Total disappearance of target lesions constituted a complete response (CR), whereas a minimum of a 30% decrease in the sum of the LDs of the lesions was classified as a partial response (PR). New lesions or a 20% increase in the sum of the LDs of the target lesions was considered progressive disease (PD). Otherwise, patients were classified as having stable disease (SD). Patients were re-evaluated for disease status after at least 4 weeks of achieving a CR or PR to confirm the assessment. Similarly, an SD was reassessed at a minimum interval of 8 weeks. Nontarget lesions were also classified as CR, PD, or SD. A CR was identified if all nontarget lesions disappeared, an SD was noted if one or more nontarget lesions persisted, and a PD was identified if one or more new lesions appeared or if there was unequivocal progression of existing nontarget lesions. The overall objective status was noted as a CR if both target and nontarget lesions were noted as CR and no new lesions developed. The overall objective status was identified as a PR if (1) the target lesions were identified as a CR, the nontarget lesions were noted as an SD, and no new lesions developed or (2) if the target lesions were noted as a PR, the nontarget lesions weren't PD, and no new lesions developed. Finally, the overall objective status was identified as a PD if either a PD was noted for target or nontarget lesions or new lesions developed. Patients with global deterioration of health status that required discontinuation of treatment without objective evidence of disease progression at that time, and not related to study treatment or other medical conditions, were considered to have PD resulting from symptomatic deterioration.
Patients who met the SD criteria at least once after study entry at a minimum interval of 8 weeks were considered to be achieving disease stabilization. The duration of disease stabilization was calculated from the study registration date to the last evaluation. For the purpose of this analysis, patients with CR, PR, or SD were considered to have achieved disease control. The duration of disease control was calculated from the first treatment response or stabilization date to the date of the last evaluation. The 24-week progression-free rate was estimated by dividing the number of patients who have not progressed within 24 weeks by the total number of evaluable patients. Patients having died (or lost to follow-up) without progression were considered to have progressed at the date of death (or last contact) unless documentation proved otherwise, in which case they would be considered to have had no progression at the date of last tumor evaluation. Time to progression is calculated from study entry to disease progression. Time to death (ie, survival) is calculated from the date of study entry to death or last contact. Duration of response is calculated from the date of the patient's first best objective status of CR or PR to the date of progression. Patients were observed until death or a maximum of 3 years postregistration.
Determination of EGFR/HER1 Expression
Tissue samples from liver needle biopsies, when available, were collected prospectively on entry to the study. Immunostaining of unstained tissue sections was performed to semiquantify the expression of EGFR/HER1 in liver cancer cells. Immunohistochemistry using DAKO EGFR/HER1 pharmDxTM monoclonal antibody (DAKO Corp, Carpinteria, CA) that detects HER1 (EGFR/HER1) protein in EGFR/HER1-expressing cells was used. The presence and intensity of staining were scored. A scale of 0 to 3 was used to indicate the following: 0, no staining; 1, few stained cells; 2, more than 10% of cells stained; 3, majority of cells stained with intense color.
Statistical Considerations
The primary end point of this trial was to evaluate the proportion of patients who were progression-free at 24 weeks. All eligible patients who began treatment and had been followed for at least 24 weeks were considered evaluable for the primary end point. A one-stage Simon-Optimal design17 was used to evaluate the 24-week progression-free rate. Erlotinib was considered inactive if none of the initial 15 patients (ie, at the interim analysis) were progression-free at 24 weeks. Four or more of 35 evaluable patients that were progression-free at 24 weeks were required to declare promising activity. Using this design, the study had 92% power (at a 0.09 level of significance) to detect a 24-week progression-free rate of at least 20%. The CI reported for the estimate of the 24-week progression-free rate is calculated by using the method of Duffy and Santner.18
Secondary end points included overall response rate, response rate in EGFR/HER1-positive patients, toxicity, time to disease progression, survival, and association between EGFR/HER1 protein levels and clinical outcome. All toxicity was summarized in a tabular manner as a maximum grade for a given type of event. Kaplan-Meier methodology19 was used to describe the distribution of time to disease progression and survival. Cox proportional hazards20 modeling was used to identify factors (ie, age, sex, and PS) that were significantly associated with positive EGFR/HER1 protein levels. P values of at least .05 were considered statistically significant in these analyses. All analyses were performed by using SAS 8.0 (SAS Institute, Cary, NC).
RESULTS
Patient Characteristics
A total of 38 patients from seven Phase II Consortium–treating institutions (Mayo Clinic, Rochester; Mayo Clinic, Scottsdale; Mayo Clinic, Jacksonville; Johns Hopkins Hospital, Baltimore, MD; Karmanos Cancer Institute, Detroit, MI; University of Wisconsin, Madison, WI; and Washington University, St Louis, MO) were enrolled between March 2002 and September 2003. Table 1 summarizes the characteristics for these patients. All were white, and 32% (12 of 38) were female. At study entry, patients ranged in age from 27 to 83 years (median, 69 years) and 26% (10 of 38) had PSs of 0 (v 1 or 2). Fifty-eight percent of the patients had documented background chronic liver disease. The majority of the patients (27 of 38) had a Child-Pugh classification of A. Forty-seven percent (18 of 38) of the patients had received prior systemic therapy for hepatoma. Twenty-two (92%) of the 24 patients with evaluable tumor samples were EGFR/HER1-positive (ie, 1, 2 or 3). Unavailability of a sufficient quantity of tissue to perform immunostaining was the reason for the lack of information on EGFR/HER1 status on the remaining 14 patients.
Erlotinib Administration
Patients received a total of 197 cycles of treatment (mean, five cycles per patient; range, 1-26). Sixty-eight percent of patients received between two and four cycles of treatment. Ten patients (26%) had up to two dose reductions. The reasons for dose reductions were skin rash (five patients), gastrointestinal symptoms (four patients), anemia (one patient), elevated bilirubin (one patient), and elevated liver-function test (one patient). Treatment delays were seen in 16 patients for a total of 20 cycles, occurring most frequently on cycles two and three. The most common reasons for treatment delays were skin rash in nine patients and gastrointestinal symptoms in five patients (including one bowel obstruction and one gastrointestinal bleeding). At the time of the analysis, two patients are still receiving treatment. Reasons for discontinuation of treatment included: 27 patients (71%) had PD, five patients (13%) experienced adverse reactions, 1 patient (2.6%) went to alternate therapy, and three patients (8%) discontinued treatment for other reasons (one had a car accident, one had heart failure, and one had a family emergency).
Toxicity
All patients were evaluable for toxicity. Sixty-one percent of the patients experienced grade 3 or 4 adverse events. Overall, leukopenia, anemia, and thrombocytopenia were minimal (ie, grade 2 or higher). Thirteen (34%) patients experienced a grade 3 or higher nonhematologic toxicity (ie, at least possibly related to erlotinib). Most frequent grade 3 to 4 toxicities were skin rash (13%), diarrhea (8%), and fatigue (8%; Table 2). We explored the correlation between severe (grade 3 or higher) toxicity and Child-Pugh classification and found that only 22% of the Child-Pugh A patients experienced severe toxicity compared to 70% of Child-Pugh B patients (P = .02).
Efficacy
All patients were considered evaluable for the primary end point. Twelve patients (32%) were progression-free after 24 weeks (95% CI, 18% to 49%). At the time of this analysis, 84% (32 of 38) of the patients have experienced disease progression, with sites including: liver (20 patients), lymph nodes (five patients), abdomen (two patients), lung (two patients), adrenal gland (two patients), ribs (one patient), spine (one patient), and kidney (one patient).
Thirty-four patients (89%) were evaluable for objective tumor response, having at least one postbaseline tumor assessment. Three PRs were observed, lasting 2, 10, and 11 months (Table 3). Table 4 lists the clinicopathologic features of the responders. The estimated overall confirmed response rate was 9% (3 of 34; 95% CI, 2% to 24%). Seventeen patients (50%) achieved stabilization of disease (95% CI, 32% to 68%) for a median of 3.8 months (range, 2 to 19 months).
Twenty-four patients have died, with a median follow-up of 11.2 months on surviving patients (range, 1.6 to 22.3 months). The distributions of time to progression and survival are depicted in Figure 1. The median overall survival time was 13 months (95% CI, 5 to 19), with a probability of 33% of patients alive at 18 months from entry into the study (Table 3).
EGFR/HER1 Expression
Twenty-five patients (66%) had sufficient tissue for EGFR/HER1 determination, three of which lacked EGFR/HER1 expression (ie, EGFR/HER1-negative). Twenty-three percent (5 of 22) of the EGFR/HER1-positive cases were progression-free at 24 weeks, whereas two (67%) of three EGFR/HER1-negative cases were progression-free at 24 weeks. Of the three patients responding to erlotinib, only one had sufficient tissue for EGFR/HER1 determination (Table 4). Therefore, correlations between response and EGFR/HER1 status could not be explored. EGFR/HER1 expression was marginally associated with grade 2 or higher skin reactions (P = .09); however, these results should be confirmed in larger studies. EGFR/HER1 status was not significantly associated with outcome regardless of whether we evaluated EGFR/HER1 as a continuous (0, 1, 2, or 3) or grouped (+ or –) variable (P > .05).
DISCUSSION
HCC represents a major unmet need area for new therapies. The problems of treating this disease are the relative chemoresistance of HCC and the background chronic liver disease that limits systemic therapies that require hepatic metabolism. The rationale for this study was the demonstration of a role for the EGFR/HER1 signaling pathway in the carcinogenesis of liver cancer.
Results of this trial revealed activity of erlotinib as a single agent in treating patients with advanced HCC that was similar to that seen with EGFR/HER1 blockade in other EGFR/HER1-expressing tumor types.21 The study population was selected with a lower threshold for certain abnormal parameters such as the platelet count. Approximately half of the treated patients had prior systemic therapy with cytotoxic drugs. Given the difficulties in accurately assessing disease response with objective measurements, the primary end point of the study was disease stabilization at 6 months. This was assumed to represent an efficacy parameter of clinical relevance that may serve as a surrogate for drug activity in this disease. The prestudy threshold for clinical interest was set at 20% of the study population for progression-free status at 6 months. The results of the study showed that 32% of the patients (95% CI, 18% to 49%) had disease stabilization at 6 months. Of the 34 patients eligible for objective response assessment, three had a PR. However, it was noteworthy that tumor-size reduction using the CT criteria was not evident until at least 195 days on therapy. This finding has an impact on the design of future studies to assess the activity of similar and other novel agents in this and probably other diseases. The role of functional imaging as an early indicator for biochemical response may need to be investigated also.
The toxicity to erlotinib was predominantly cutaneous and similar in profile to other drugs that target the EGFR/HER1-related tyrosine kinase activity. Twenty-four percent of the patients required dose reductions because of toxicity. The frequency of dose modifications was greater compared to a parallel cohort of patients with biliary cancer treated by a similar dose and schedule of erlotinib.22 In this study, at least 58% of the patients had documented chronic liver disease. Of note, patients with Child-Pugh classification B were more likely to experience grade 3 or 4 toxicity than those with the A classification. The resultant clinical or subclinical impairment of liver function may have interfered with the metabolic clearance of erlotinib and resulted in higher plasma and tissue levels. Future randomized trials in HCC should have Child-Pugh classification as a stratification factor. Additional pharmacokinetic studies of erlotinib in patients with liver dysfunction could resolve this issue. All three patients who had PR to therapy had some degree of skin toxicity.
In a study of EGFR/HER1 expression in seven of 17 HCCs, no significant difference was observed between EGFR/HER1-positive and EGFR/HER1-negative cases in tumor size, alpha-fetoprotein, macroscopic and histopathologic classification, capsular invasion, and portal invasion. However, in EGFR/HER1-positive cases, the number of DNA polymerase-alpha–positive cells and recurrence rate were higher than that in EGFR/HER1-negative cases. These results suggest that EGFR/HER1 may play an important role in the development and progression of HCC.23 Although the presence of EGFR/HER1 was found to be necessary, but not sufficient, for in vitro response to EGFR/HER1 inhibitors, the degree of EGFR/HER1 expression may not be as important for antitumor effect as the reliance on this pathway for cell proliferation and survival.24,25 Clinical trials of EGFR/HER1 blockers in other cancers have been conducted in patients with EGFR/HER1-expressing tumors. However, the development of EGFR/HER1-blocking therapies has not been based on the expression of EGFR/HER1 protein by the tumor. The frequency of positive staining for the EGFR/HER1 protein in this study was 88%. The low frequency of objective responses and the small proportion of the patients with negative expression of the EGFR/HER1 preclude drawing any conclusions on the relationship between EGFR/HER1 status and response to EGFR/HER1 blockade with erlotinib. Evidence from other tumors treated with EGFR/HER1 blockers has indicated a lack of correlation between intensity of EGFR/HER1 positivity using immunohistochemistry and response to therapy.26 There is also emerging evidence that response to an EGFR/HER1 blocker may occur even in the absence of EGFR/HER1 positivity in colorectal cancer.21 At this time there is no evidence to support the development of erlotinib in only patients with EGFR/HER1-positive tumors.
Future improvements in the therapy of HCC will continue to be based on identifying molecular markers that select patients who respond to a given therapy. It is also noteworthy that HCC represents a heterogeneous disease with multiple etiologies that include viral and metabolic insults to the liver. The role of pathways such as EGFR/HER1 may consequently vary in the different subgroups with resultant differences in the efficacy of these agents. For example, the hepatitis C–causative virus (HCV) genotype 1a and 3 core proteins activate MEK1 and Erk1/2 mitogen-activated protein kinases, and the constitutive expression of the HCV core results in a high basal activity of Raf1 and mitogen-activated protein kinase kinase. Moreover, the activation of both Erk1/2 and the downstream transcription factor Elk-1 in response to the mitogenic stimulus EGF is significantly prolonged.27 It is noteworthy that the patient with the best response to therapy had chronic liver disease caused by hepatitis C infection.
In conclusion, results of this trial encourage the development of a strategy of EGFR/HER1 blockade as a treatment modality in HCC. Selection of patients who will respond to EGFR/HER1 inhibitors must also be investigated. Mutations in the EGFR/HER1 gene have been suggested to influence the responsiveness of non–small-cell lung cancers to anti-EGFR/HER1 agents.28 Similar work is needed to identify patients who will respond to erlotinib. Given the redundancy of signaling pathways in established cancers and the multiple genetic abnormalities that lead to the malignant phenotype, combinations of EGFR/HER1 blockade with other novel agents should be considered also.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by NO-1 grant CM17104 from the National Institutes of Health.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
El-Serag HB, Mason AC: Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 340:745-750, 1999
Ihde DC, Kane RC, Cohen MH, et al: Adriamycin therapy in American patients with hepatocellular carcinoma. Cancer Treat Rep 61:1385-1387, 1977
Chlebowski RT, Brzechwa Adjukiewicz A, Cowden A, et al: Doxorubicin (75 mg/m2) for hepatocellular carcinoma: Clinical and pharmacokinetic results. Cancer Treat Rep 68:487-491, 1984
Sciarrino E, Simonetti RG, LeMoli S, et al: Adriamycin treatment for hepatocellular carcinoma: Experience with 109 patients. Cancer 56:2751-2755, 1985
Lee Y-T: Systemic and regional treatment of primary carcinoma of the liver. Cancer Treat Rev 4:195-212, 1977
Nerenstone SR, Ihde DC, Freidman MA: Clinical trials in primary hepatocellular carcinoma: Current status and future directions. Cancer Treat Rev 15:1-31, 1988
Ahlgren JD: Neoplasms of the hepatobiliary system, in P Calbresi, P Schein (eds): Medical Oncology (2nd ed). New York, NY, McGraw Hill, 1996, pp 713-739
Izumi R, Shimizu K, Li T, et al: Prognostic factors of hepatocellular carcinoma undergoing hepatic resection. Gastroenterology 106:720-727, 1994
Woodburn JR: The epidermal growth factor receptor and its inhibition in cancer therapy. Pharmacol Ther 82:241-250, 1999
Gibbs JB: Anticancer drug targets: Growth factors and growth factor signaling. J Clin Invest 105:9-13, 2000
Kaneko Y, Shibuya M, Nakayama T, et al: Hypomethylation of c-myc and epidermal growth factor receptor genes in human hepatocellular carcinoma and fetal liver. Jpn J Cancer Res 76:1136-1140, 1985
Xu YH, Jiang WL, Peng SF: EGFR/HER1 expression and EGF stimulation of proliferation in human liver carcinoma cells [in Chinese]. Shi Yan Sheng Wu Xue Bao 22:445-453, 1989
Hung WC, Chuang LY, Tsai JH, et al: Effects of epidermal growth factor on growth control and signal transduction pathways in different human hepatoma cell lines. Biochem Mol Biol Int 30:319-328, 1993
Yamaguchi K, Carr BI, Nalesnik MA: Concomitant and isolated expression of TGF-alpha and EGF-R in human hepatoma cells supports the hypothesis of autocrine, paracrine, and endocrine growth of human hepatoma. J Surg Oncol 58:240-245, 1995
Morimitsu Y, Hsia CC, Kojiro M, et al: Nodules of less-differentiated tumor within or adjacent to hepatocellular carcinoma: Relative expression of transforming growth factor-alpha and its receptor in the different areas of tumor. Hum Pathol 26:1126-1132, 1995
Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92:205-216, 2000
Simon R: Optimal two-stage designs for phase II clinical trials. Control Clin Trials 10:1-10, 1989
Duffy DE, Santner TJ: Confidence intervals for a binomial parameter based on multistage tests. Biometrics 43:81-93, 1987
Kaplan E, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression models and life tables (with discussion). J R Stat Soc (B) 74:187-220, 1972
Lenz HJ, Mayer RJ, Gold PJ, et al: Activity of cetuximab in patients with colorectal cancer refractory to both irinotecan and oxaliplatin. J Clin Oncol 23:248, 2004 (abstr 3510)
Philip PA, Mahoney M, Thomas J, et al: Phase II trial of erlotinib (OSI-774) in patients with hepatocellular or biliary cancer. J Clin Oncol 23:318, 2004 (abstr 4025)
Tanaka H: Immunohistochemical studies on epidermal growth factor receptor in hepatocellular carcinoma. [in Japanese]. Nippon Shokakibyo Gakkai Zasshi 88:138-144, 1991
Baguley BC, Marshall ES, Holdaway KM, et al: Inhibition of growth of primary human tumour cell cultures by a 4-anilinoquinazoline inhibitor of the epidermal growth factor receptor family of tyrosine kinases. Eur J Cancer 34:1086-1090, 1998
Sirotnak FM, Zakowsky MF, Miller VA, et al: Potentiation of cytotoxic agents against human tumors in mice by ZD1839 (IRESSA), an inhibitor of EGFR/HER1 tyrosine kinase, does not require high levels of expression of EGFR/HER1. Proc Annu Meet Am Assoc Cancer Res 41:482, 2000 (abstr 3076)
Cunningham D, Humblet Y, Siena S, et al: Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 351:337-345, 2004
Giambartolomei S, Covone F, Levrero M, et al: Sustained activation of the Raf/MEK/Erk pathway in response to EGF in stable cell lines expressing the hepatitis C virus (HCV) core protein. Oncogene 20:2606-2610, 2001
Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med, Volume 350:2129-2139, 2004(Philip A. Philip, Michell)
University of Wisconsin Comprehensive Cancer Center, Madison, WI
Johns Hopkins University, Baltimore, MD
Mayo Clinic Jacksonville, Jacksonville, FL
Mayo Clinic Scottsdale, Scottsdale, AZ
Washington University School of Medicine, St Louis, MO
Mayo Clinic, Rochester, MN
ABSTRACT
PURPOSE: Epidermal growth factor receptor/human epidermal growth factor receptor 1 (EGFR/HER1) and ligand expression is frequently seen in hepatocellular cancers (HCCs). Erlotinib (Tarceva, OSI-774; OSI Pharmaceuticals, Melville, NY) is a receptor tyrosine kinase inhibitor with specificity for the EGFR/HER1.
METHODS: The primary objective of this study was to determine the proportion of patients with advanced HCC who were progression-free at 6 months. Patients with either unresectable or metastatic disease were studied. Only one prior systemic or locoregional therapy was allowed. Erlotinib was given continuously at a dose of 150 mg per day orally.
RESULTS: Thirty-eight patients with HCC were enrolled. Median age of the patients was 69 years (range, 27 to 83 years). A majority of patients (63%) had an Eastern Cooperative Oncology Group performance status of 1. Forty-seven percent of patients had received prior chemotherapy for advanced HCC. EGFR/HER1 expression was detected in 88% of the patients. Median number of cycles per patient was two (range, 1 to 26). Twelve (32%; CI 95%, 18 to 49) of the 38 patients with HCC were progression-free at 6 months. Three patients had partial radiologic responses of duration of 2, 10, and 11 months, respectively. Disease control was seen in 59% of the patients. Median overall survival time was 13 months. Ten patients (26%) had toxicity-related dose reductions of erlotinib. Grade 3/4 skin toxicity or diarrhea was encountered in five and three patients, respectively.
CONCLUSION: Results of this trial suggest a benefit for EGFR/HER1 blockade with erlotinib in patients with HCC manifested by disease control. Additional studies with erlotinib as a single agent or in combination with other agents are warranted.
INTRODUCTION
Hepatocellular cancer (HCC) is the most common primary neoplasm of the liver, accounting for almost half a million deaths annually worldwide. Although the incidence of HCC is particularly high in parts of Asia and Africa, recent studies have documented a clear rise in the number of cases in Japan, Western Europe, and the United States, largely because of the increase in hepatitis C–related liver disease.1 Unfortunately, most patients with HCC are not candidates for any curative treatments because of advanced disease at presentation and/or background chronic liver disease. Cytotoxic therapy of HCC has also been disappointing, without a significant impact on the natural history of this disease.2-8
The epidermal growth factor receptor/human epidermal growth factor receptor 1 (EGFR/HER1) and its ligands EGF and transforming growth factor-alpha (TGF-) are important in cell proliferation, as well as motility, adhesion, invasion, survival, and angiogenesis.9,10 It has been suggested that hypomethylation of the EGF receptor gene may be associated with the development of HCC.11 Studies have indicated that EGFR/HER1 is actively expressed in human hepatoma cells and that EGF may be one of the mitogens needed for the growth of hepatoma cell lines.12 In one study, expression of EGFR/HER1 was present in nine of 11 HCCs. Additional data indicate that EGF, the ligand for EGFR/HER1, may have different effects on different human hepatoma cell lines, and its role might be more important in poorly differentiated hepatoma cells than in the well-differentiated ones.13 Studies have supported the theory of an autocrine, paracrine, and endocrine mechanism of TGF- and EGFR/HER1 on the proliferation of human HCC.14 Furthermore, the detection of greater staining in more highly differentiated portions of the tumors suggested that increased expression of TGF- and EGFR/HER1 were linked to the early stages of human hepatocarcinogenesis.15
Erlotinib (Tarceva, OSI-774; OSI Pharmaceuticals, Melville, NY) is an orally active, potent, selective inhibitor of the EGFR/HER1-related tyrosine kinase enzyme. Erlotinib inhibits the tyrosine kinase activity with an IC50 of 2 nmol/L in an in vitro enzyme assay and reduces EGFR/HER1 autophosphorylation in intact tumor cells with an IC50 of 20 nmol/L. Erlotinib inhibits EGF-dependent proliferation of cancer cells at submicromolar concentrations and blocks cell-cycle progression in the G1 phase.
The lack of any clinically useful drug therapy for patients with advanced HCC and the suggestion of an antiproliferative and proapoptotic role for EGFR/HER1 blockade prompted this pilot trial in patients with HCC. We hypothesized that using EGFR/HER1 blockade in patients with HCC by erlotinib will reduce the growth of these tumors and delay the progression of the disease. The primary objective of this study was to determine the progression-free survival at 6 months of patients with HCC treated with erlotinib 150 mg per day. Secondary end points included objective response frequency and duration, adverse events rates, time to disease progression, and overall survival. We also evaluated correlation of EGFR/HER1 status with patient characteristics and outcome.
METHODS
Eligibility Criteria
Patients eligible for the study had histologically or cytologically confirmed HCC that was surgically unresectable or metastatic. A hypervascular mass on computed tomography (CT) with a serum alpha-fetoprotein level of more than 100 ng/mL was sufficient noninvasive diagnostic criteria for HCC. Measurable disease was required. Other eligibility criteria included an Eastern Cooperative Oncology Group performance status (PS) of 0, 1, or 2; an estimated life expectancy of at least 3 months; adequate bone marrow, hepatic, and renal function, indicated by an absolute neutrophil count of 1,500/μL, platelets of 75,000/μL, and total bilirubin of 2x upper limit of normal, respectively; serum AST or ALT levels of 3x upper limit of normal; serum creatinine of 2 mg/dL; serum albumin concentration of 2.5 g/dL; an international normalized ratio of 1.5 (unless on anticoagulation); and a Child-Pugh classification of A or B. For patients having prior cryotherapy, radiofrequency ablation, ethanol injection, or photodynamic therapy, the following criteria were required: more than 6 weeks had elapsed since that therapy; indicator lesion(s) was/were outside the area of prior treatment; or, if the only indicator lesion was inside the prior treatment area, there must have been clear evidence of disease progression associated with that lesion. In addition, edges of the indicator lesion must have been clearly distinct on CT scanning. Patients with more than one prior systemic anticancer therapy were excluded. Other exclusion criteria included a history of other malignancy within the previous 3 years (except for adequately treated basal cell cancer, squamous cell skin cancer, or cervical cancer). Patients were also excluded on the basis of known abnormalities of the cornea, CNS metastases, and HIV infection.
Erlotinib Administration
All patients received erlotinib 150 mg orally once daily. Treatment was given on a continuous daily basis, and a treatment cycle was 28 days. The erlotinib dose was reduced according to prestudy-defined toxicity criteria to 100 or 50 mg per day. No erlotinib dose re-escalation was allowed after dose reduction resulting from toxicity. Patients requiring dose reductions to less than 50 mg per day were taken off of the study. Treatment was held up to 14 days for grade 3 or 4 toxicity until resolution of toxicity to grade 1 or higher. Erlotinib was then reinitiated at a reduced dose. Any of the following were allowed for the treatment of acneiform skin rash: minocycline, topical tetracycline, topical clindamycin, topical silver sulfadiazine, diphenhydramine, and oral prednisone.
Disease Assessment
Objective response to therapy was assessed by using the Response Evaluation Criteria in Solid Tumors criteria.16 Measurable disease was defined as at least one lesion whose longest diameter (LD) could be measured accurately as 2.0 cm. Lesions detected by the physical examination were only considered measurable when they were superficial (eg, skin nodules, palpable lymph nodes). Lesions on chest x-ray were considered measurable lesions if clearly defined and surrounded by aerated lung. However, CT or magnetic resonance imaging evaluation was the measurement of choice for such lesions. All other lesions (or sites of disease), including small lesions (LD, < 2.0 cm) were considered nonmeasurable lesions.
All measurable lesions ( 2.0 cm) up to a maximum of 10 lesions representative of all involved organs were identified as target lesions and recorded at baseline. A sum of the LD for all target lesions were calculated and reported as the baseline sum LD. The baseline sum LD was used as reference to characterize the objective tumor response to erlotinib. All nonmeasurable lesions were identified as nontarget lesions and were also recorded at baseline. Measurements were not required for these lesions, and they were followed as "present" or "absent."
All identified sites of disease at baseline were followed on re-evaluation. Total disappearance of target lesions constituted a complete response (CR), whereas a minimum of a 30% decrease in the sum of the LDs of the lesions was classified as a partial response (PR). New lesions or a 20% increase in the sum of the LDs of the target lesions was considered progressive disease (PD). Otherwise, patients were classified as having stable disease (SD). Patients were re-evaluated for disease status after at least 4 weeks of achieving a CR or PR to confirm the assessment. Similarly, an SD was reassessed at a minimum interval of 8 weeks. Nontarget lesions were also classified as CR, PD, or SD. A CR was identified if all nontarget lesions disappeared, an SD was noted if one or more nontarget lesions persisted, and a PD was identified if one or more new lesions appeared or if there was unequivocal progression of existing nontarget lesions. The overall objective status was noted as a CR if both target and nontarget lesions were noted as CR and no new lesions developed. The overall objective status was identified as a PR if (1) the target lesions were identified as a CR, the nontarget lesions were noted as an SD, and no new lesions developed or (2) if the target lesions were noted as a PR, the nontarget lesions weren't PD, and no new lesions developed. Finally, the overall objective status was identified as a PD if either a PD was noted for target or nontarget lesions or new lesions developed. Patients with global deterioration of health status that required discontinuation of treatment without objective evidence of disease progression at that time, and not related to study treatment or other medical conditions, were considered to have PD resulting from symptomatic deterioration.
Patients who met the SD criteria at least once after study entry at a minimum interval of 8 weeks were considered to be achieving disease stabilization. The duration of disease stabilization was calculated from the study registration date to the last evaluation. For the purpose of this analysis, patients with CR, PR, or SD were considered to have achieved disease control. The duration of disease control was calculated from the first treatment response or stabilization date to the date of the last evaluation. The 24-week progression-free rate was estimated by dividing the number of patients who have not progressed within 24 weeks by the total number of evaluable patients. Patients having died (or lost to follow-up) without progression were considered to have progressed at the date of death (or last contact) unless documentation proved otherwise, in which case they would be considered to have had no progression at the date of last tumor evaluation. Time to progression is calculated from study entry to disease progression. Time to death (ie, survival) is calculated from the date of study entry to death or last contact. Duration of response is calculated from the date of the patient's first best objective status of CR or PR to the date of progression. Patients were observed until death or a maximum of 3 years postregistration.
Determination of EGFR/HER1 Expression
Tissue samples from liver needle biopsies, when available, were collected prospectively on entry to the study. Immunostaining of unstained tissue sections was performed to semiquantify the expression of EGFR/HER1 in liver cancer cells. Immunohistochemistry using DAKO EGFR/HER1 pharmDxTM monoclonal antibody (DAKO Corp, Carpinteria, CA) that detects HER1 (EGFR/HER1) protein in EGFR/HER1-expressing cells was used. The presence and intensity of staining were scored. A scale of 0 to 3 was used to indicate the following: 0, no staining; 1, few stained cells; 2, more than 10% of cells stained; 3, majority of cells stained with intense color.
Statistical Considerations
The primary end point of this trial was to evaluate the proportion of patients who were progression-free at 24 weeks. All eligible patients who began treatment and had been followed for at least 24 weeks were considered evaluable for the primary end point. A one-stage Simon-Optimal design17 was used to evaluate the 24-week progression-free rate. Erlotinib was considered inactive if none of the initial 15 patients (ie, at the interim analysis) were progression-free at 24 weeks. Four or more of 35 evaluable patients that were progression-free at 24 weeks were required to declare promising activity. Using this design, the study had 92% power (at a 0.09 level of significance) to detect a 24-week progression-free rate of at least 20%. The CI reported for the estimate of the 24-week progression-free rate is calculated by using the method of Duffy and Santner.18
Secondary end points included overall response rate, response rate in EGFR/HER1-positive patients, toxicity, time to disease progression, survival, and association between EGFR/HER1 protein levels and clinical outcome. All toxicity was summarized in a tabular manner as a maximum grade for a given type of event. Kaplan-Meier methodology19 was used to describe the distribution of time to disease progression and survival. Cox proportional hazards20 modeling was used to identify factors (ie, age, sex, and PS) that were significantly associated with positive EGFR/HER1 protein levels. P values of at least .05 were considered statistically significant in these analyses. All analyses were performed by using SAS 8.0 (SAS Institute, Cary, NC).
RESULTS
Patient Characteristics
A total of 38 patients from seven Phase II Consortium–treating institutions (Mayo Clinic, Rochester; Mayo Clinic, Scottsdale; Mayo Clinic, Jacksonville; Johns Hopkins Hospital, Baltimore, MD; Karmanos Cancer Institute, Detroit, MI; University of Wisconsin, Madison, WI; and Washington University, St Louis, MO) were enrolled between March 2002 and September 2003. Table 1 summarizes the characteristics for these patients. All were white, and 32% (12 of 38) were female. At study entry, patients ranged in age from 27 to 83 years (median, 69 years) and 26% (10 of 38) had PSs of 0 (v 1 or 2). Fifty-eight percent of the patients had documented background chronic liver disease. The majority of the patients (27 of 38) had a Child-Pugh classification of A. Forty-seven percent (18 of 38) of the patients had received prior systemic therapy for hepatoma. Twenty-two (92%) of the 24 patients with evaluable tumor samples were EGFR/HER1-positive (ie, 1, 2 or 3). Unavailability of a sufficient quantity of tissue to perform immunostaining was the reason for the lack of information on EGFR/HER1 status on the remaining 14 patients.
Erlotinib Administration
Patients received a total of 197 cycles of treatment (mean, five cycles per patient; range, 1-26). Sixty-eight percent of patients received between two and four cycles of treatment. Ten patients (26%) had up to two dose reductions. The reasons for dose reductions were skin rash (five patients), gastrointestinal symptoms (four patients), anemia (one patient), elevated bilirubin (one patient), and elevated liver-function test (one patient). Treatment delays were seen in 16 patients for a total of 20 cycles, occurring most frequently on cycles two and three. The most common reasons for treatment delays were skin rash in nine patients and gastrointestinal symptoms in five patients (including one bowel obstruction and one gastrointestinal bleeding). At the time of the analysis, two patients are still receiving treatment. Reasons for discontinuation of treatment included: 27 patients (71%) had PD, five patients (13%) experienced adverse reactions, 1 patient (2.6%) went to alternate therapy, and three patients (8%) discontinued treatment for other reasons (one had a car accident, one had heart failure, and one had a family emergency).
Toxicity
All patients were evaluable for toxicity. Sixty-one percent of the patients experienced grade 3 or 4 adverse events. Overall, leukopenia, anemia, and thrombocytopenia were minimal (ie, grade 2 or higher). Thirteen (34%) patients experienced a grade 3 or higher nonhematologic toxicity (ie, at least possibly related to erlotinib). Most frequent grade 3 to 4 toxicities were skin rash (13%), diarrhea (8%), and fatigue (8%; Table 2). We explored the correlation between severe (grade 3 or higher) toxicity and Child-Pugh classification and found that only 22% of the Child-Pugh A patients experienced severe toxicity compared to 70% of Child-Pugh B patients (P = .02).
Efficacy
All patients were considered evaluable for the primary end point. Twelve patients (32%) were progression-free after 24 weeks (95% CI, 18% to 49%). At the time of this analysis, 84% (32 of 38) of the patients have experienced disease progression, with sites including: liver (20 patients), lymph nodes (five patients), abdomen (two patients), lung (two patients), adrenal gland (two patients), ribs (one patient), spine (one patient), and kidney (one patient).
Thirty-four patients (89%) were evaluable for objective tumor response, having at least one postbaseline tumor assessment. Three PRs were observed, lasting 2, 10, and 11 months (Table 3). Table 4 lists the clinicopathologic features of the responders. The estimated overall confirmed response rate was 9% (3 of 34; 95% CI, 2% to 24%). Seventeen patients (50%) achieved stabilization of disease (95% CI, 32% to 68%) for a median of 3.8 months (range, 2 to 19 months).
Twenty-four patients have died, with a median follow-up of 11.2 months on surviving patients (range, 1.6 to 22.3 months). The distributions of time to progression and survival are depicted in Figure 1. The median overall survival time was 13 months (95% CI, 5 to 19), with a probability of 33% of patients alive at 18 months from entry into the study (Table 3).
EGFR/HER1 Expression
Twenty-five patients (66%) had sufficient tissue for EGFR/HER1 determination, three of which lacked EGFR/HER1 expression (ie, EGFR/HER1-negative). Twenty-three percent (5 of 22) of the EGFR/HER1-positive cases were progression-free at 24 weeks, whereas two (67%) of three EGFR/HER1-negative cases were progression-free at 24 weeks. Of the three patients responding to erlotinib, only one had sufficient tissue for EGFR/HER1 determination (Table 4). Therefore, correlations between response and EGFR/HER1 status could not be explored. EGFR/HER1 expression was marginally associated with grade 2 or higher skin reactions (P = .09); however, these results should be confirmed in larger studies. EGFR/HER1 status was not significantly associated with outcome regardless of whether we evaluated EGFR/HER1 as a continuous (0, 1, 2, or 3) or grouped (+ or –) variable (P > .05).
DISCUSSION
HCC represents a major unmet need area for new therapies. The problems of treating this disease are the relative chemoresistance of HCC and the background chronic liver disease that limits systemic therapies that require hepatic metabolism. The rationale for this study was the demonstration of a role for the EGFR/HER1 signaling pathway in the carcinogenesis of liver cancer.
Results of this trial revealed activity of erlotinib as a single agent in treating patients with advanced HCC that was similar to that seen with EGFR/HER1 blockade in other EGFR/HER1-expressing tumor types.21 The study population was selected with a lower threshold for certain abnormal parameters such as the platelet count. Approximately half of the treated patients had prior systemic therapy with cytotoxic drugs. Given the difficulties in accurately assessing disease response with objective measurements, the primary end point of the study was disease stabilization at 6 months. This was assumed to represent an efficacy parameter of clinical relevance that may serve as a surrogate for drug activity in this disease. The prestudy threshold for clinical interest was set at 20% of the study population for progression-free status at 6 months. The results of the study showed that 32% of the patients (95% CI, 18% to 49%) had disease stabilization at 6 months. Of the 34 patients eligible for objective response assessment, three had a PR. However, it was noteworthy that tumor-size reduction using the CT criteria was not evident until at least 195 days on therapy. This finding has an impact on the design of future studies to assess the activity of similar and other novel agents in this and probably other diseases. The role of functional imaging as an early indicator for biochemical response may need to be investigated also.
The toxicity to erlotinib was predominantly cutaneous and similar in profile to other drugs that target the EGFR/HER1-related tyrosine kinase activity. Twenty-four percent of the patients required dose reductions because of toxicity. The frequency of dose modifications was greater compared to a parallel cohort of patients with biliary cancer treated by a similar dose and schedule of erlotinib.22 In this study, at least 58% of the patients had documented chronic liver disease. Of note, patients with Child-Pugh classification B were more likely to experience grade 3 or 4 toxicity than those with the A classification. The resultant clinical or subclinical impairment of liver function may have interfered with the metabolic clearance of erlotinib and resulted in higher plasma and tissue levels. Future randomized trials in HCC should have Child-Pugh classification as a stratification factor. Additional pharmacokinetic studies of erlotinib in patients with liver dysfunction could resolve this issue. All three patients who had PR to therapy had some degree of skin toxicity.
In a study of EGFR/HER1 expression in seven of 17 HCCs, no significant difference was observed between EGFR/HER1-positive and EGFR/HER1-negative cases in tumor size, alpha-fetoprotein, macroscopic and histopathologic classification, capsular invasion, and portal invasion. However, in EGFR/HER1-positive cases, the number of DNA polymerase-alpha–positive cells and recurrence rate were higher than that in EGFR/HER1-negative cases. These results suggest that EGFR/HER1 may play an important role in the development and progression of HCC.23 Although the presence of EGFR/HER1 was found to be necessary, but not sufficient, for in vitro response to EGFR/HER1 inhibitors, the degree of EGFR/HER1 expression may not be as important for antitumor effect as the reliance on this pathway for cell proliferation and survival.24,25 Clinical trials of EGFR/HER1 blockers in other cancers have been conducted in patients with EGFR/HER1-expressing tumors. However, the development of EGFR/HER1-blocking therapies has not been based on the expression of EGFR/HER1 protein by the tumor. The frequency of positive staining for the EGFR/HER1 protein in this study was 88%. The low frequency of objective responses and the small proportion of the patients with negative expression of the EGFR/HER1 preclude drawing any conclusions on the relationship between EGFR/HER1 status and response to EGFR/HER1 blockade with erlotinib. Evidence from other tumors treated with EGFR/HER1 blockers has indicated a lack of correlation between intensity of EGFR/HER1 positivity using immunohistochemistry and response to therapy.26 There is also emerging evidence that response to an EGFR/HER1 blocker may occur even in the absence of EGFR/HER1 positivity in colorectal cancer.21 At this time there is no evidence to support the development of erlotinib in only patients with EGFR/HER1-positive tumors.
Future improvements in the therapy of HCC will continue to be based on identifying molecular markers that select patients who respond to a given therapy. It is also noteworthy that HCC represents a heterogeneous disease with multiple etiologies that include viral and metabolic insults to the liver. The role of pathways such as EGFR/HER1 may consequently vary in the different subgroups with resultant differences in the efficacy of these agents. For example, the hepatitis C–causative virus (HCV) genotype 1a and 3 core proteins activate MEK1 and Erk1/2 mitogen-activated protein kinases, and the constitutive expression of the HCV core results in a high basal activity of Raf1 and mitogen-activated protein kinase kinase. Moreover, the activation of both Erk1/2 and the downstream transcription factor Elk-1 in response to the mitogenic stimulus EGF is significantly prolonged.27 It is noteworthy that the patient with the best response to therapy had chronic liver disease caused by hepatitis C infection.
In conclusion, results of this trial encourage the development of a strategy of EGFR/HER1 blockade as a treatment modality in HCC. Selection of patients who will respond to EGFR/HER1 inhibitors must also be investigated. Mutations in the EGFR/HER1 gene have been suggested to influence the responsiveness of non–small-cell lung cancers to anti-EGFR/HER1 agents.28 Similar work is needed to identify patients who will respond to erlotinib. Given the redundancy of signaling pathways in established cancers and the multiple genetic abnormalities that lead to the malignant phenotype, combinations of EGFR/HER1 blockade with other novel agents should be considered also.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by NO-1 grant CM17104 from the National Institutes of Health.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
El-Serag HB, Mason AC: Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 340:745-750, 1999
Ihde DC, Kane RC, Cohen MH, et al: Adriamycin therapy in American patients with hepatocellular carcinoma. Cancer Treat Rep 61:1385-1387, 1977
Chlebowski RT, Brzechwa Adjukiewicz A, Cowden A, et al: Doxorubicin (75 mg/m2) for hepatocellular carcinoma: Clinical and pharmacokinetic results. Cancer Treat Rep 68:487-491, 1984
Sciarrino E, Simonetti RG, LeMoli S, et al: Adriamycin treatment for hepatocellular carcinoma: Experience with 109 patients. Cancer 56:2751-2755, 1985
Lee Y-T: Systemic and regional treatment of primary carcinoma of the liver. Cancer Treat Rev 4:195-212, 1977
Nerenstone SR, Ihde DC, Freidman MA: Clinical trials in primary hepatocellular carcinoma: Current status and future directions. Cancer Treat Rev 15:1-31, 1988
Ahlgren JD: Neoplasms of the hepatobiliary system, in P Calbresi, P Schein (eds): Medical Oncology (2nd ed). New York, NY, McGraw Hill, 1996, pp 713-739
Izumi R, Shimizu K, Li T, et al: Prognostic factors of hepatocellular carcinoma undergoing hepatic resection. Gastroenterology 106:720-727, 1994
Woodburn JR: The epidermal growth factor receptor and its inhibition in cancer therapy. Pharmacol Ther 82:241-250, 1999
Gibbs JB: Anticancer drug targets: Growth factors and growth factor signaling. J Clin Invest 105:9-13, 2000
Kaneko Y, Shibuya M, Nakayama T, et al: Hypomethylation of c-myc and epidermal growth factor receptor genes in human hepatocellular carcinoma and fetal liver. Jpn J Cancer Res 76:1136-1140, 1985
Xu YH, Jiang WL, Peng SF: EGFR/HER1 expression and EGF stimulation of proliferation in human liver carcinoma cells [in Chinese]. Shi Yan Sheng Wu Xue Bao 22:445-453, 1989
Hung WC, Chuang LY, Tsai JH, et al: Effects of epidermal growth factor on growth control and signal transduction pathways in different human hepatoma cell lines. Biochem Mol Biol Int 30:319-328, 1993
Yamaguchi K, Carr BI, Nalesnik MA: Concomitant and isolated expression of TGF-alpha and EGF-R in human hepatoma cells supports the hypothesis of autocrine, paracrine, and endocrine growth of human hepatoma. J Surg Oncol 58:240-245, 1995
Morimitsu Y, Hsia CC, Kojiro M, et al: Nodules of less-differentiated tumor within or adjacent to hepatocellular carcinoma: Relative expression of transforming growth factor-alpha and its receptor in the different areas of tumor. Hum Pathol 26:1126-1132, 1995
Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92:205-216, 2000
Simon R: Optimal two-stage designs for phase II clinical trials. Control Clin Trials 10:1-10, 1989
Duffy DE, Santner TJ: Confidence intervals for a binomial parameter based on multistage tests. Biometrics 43:81-93, 1987
Kaplan E, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression models and life tables (with discussion). J R Stat Soc (B) 74:187-220, 1972
Lenz HJ, Mayer RJ, Gold PJ, et al: Activity of cetuximab in patients with colorectal cancer refractory to both irinotecan and oxaliplatin. J Clin Oncol 23:248, 2004 (abstr 3510)
Philip PA, Mahoney M, Thomas J, et al: Phase II trial of erlotinib (OSI-774) in patients with hepatocellular or biliary cancer. J Clin Oncol 23:318, 2004 (abstr 4025)
Tanaka H: Immunohistochemical studies on epidermal growth factor receptor in hepatocellular carcinoma. [in Japanese]. Nippon Shokakibyo Gakkai Zasshi 88:138-144, 1991
Baguley BC, Marshall ES, Holdaway KM, et al: Inhibition of growth of primary human tumour cell cultures by a 4-anilinoquinazoline inhibitor of the epidermal growth factor receptor family of tyrosine kinases. Eur J Cancer 34:1086-1090, 1998
Sirotnak FM, Zakowsky MF, Miller VA, et al: Potentiation of cytotoxic agents against human tumors in mice by ZD1839 (IRESSA), an inhibitor of EGFR/HER1 tyrosine kinase, does not require high levels of expression of EGFR/HER1. Proc Annu Meet Am Assoc Cancer Res 41:482, 2000 (abstr 3076)
Cunningham D, Humblet Y, Siena S, et al: Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 351:337-345, 2004
Giambartolomei S, Covone F, Levrero M, et al: Sustained activation of the Raf/MEK/Erk pathway in response to EGF in stable cell lines expressing the hepatitis C virus (HCV) core protein. Oncogene 20:2606-2610, 2001
Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med, Volume 350:2129-2139, 2004(Philip A. Philip, Michell)