Quality-of-life benefits and evidence of antitumour activity for patients with brain metastases treated with gefitinib

Brain metastases are a common complication of non-small-cell lung cancer (NSCLC). The role of chemotherapy in the treatment of brain metastases has not been clearly defined. Emerging case reports of patients with recurrent NSCLC treated as part of the Expanded Access Programme reveal that gefitinib (‘Iressa’, ZD1839) has clinical activity in some patients with brain metastases. Here, we describe a number of case studies documenting the response of patients with brain metastases to treatment with gefitinib. Many of these patients had quality-of-life benefits with improvement of neurological and systemic symptoms; some had a partial response of their brain metastases and even complete responses have been seen in a few patients. One case report also describes a durable long-term response with concurrent treatment with gefitinib and radiotherapy. Such results call for larger trials designed to evaluate and define the role of gefitinib in the treatment of brain metastases in NSCLC patients, either as a single agent or in combination with radiation therapy.

Brain metastases are a common complication for patients with non-small-cell lung cancer (NSCLC) and a significant cause of morbidity and mortality. Radiotherapy remains the mainstay of treatment for brain metastases; however, patients treated with whole-brain radiation still have a poor prognosis with a median survival of only 3 -4 months (Rodrigus et al, 2001). The role of chemotherapy in the treatment of patients with brain metastases has not been clearly defined. Many of these tumours are relatively resistant to chemotherapy and, traditionally, it has been assumed that the blood -brain barrier prevented chemotherapeutic agents from entering the central nervous system (CNS). However, the blood -brain barrier is disrupted in malignant brain tumours and recent evidence suggests that the chemosensitivity of the primary tumour is the major determinant of the response to systemic treatment of brain metastases, rather than the ability of the agent to penetrate an intact blood -brain barrier.
Gefitinib ('Iressa', ZD1839) is an orally active, epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor that blocks signal transduction pathways implicated in the proliferation and survival of cancer cells. High levels of EGFR are frequently found in malignant primary brain tumours (Ekstrand et al, 1992) and in tumours that often metastasise to the CNS, such as NSCLC and breast cancer (Wikstrand et al, 1995). Large Phase II studies have recently shown that gefitinib has clinical activity in advanced NSCLC and is well tolerated (Fukuoka et al, 2003;Kris et al, 2003). A preclinical study in mice showed that gefitinib retains therapeutic activity on intracranial tumours despite the blood -brain barrier (Heimberger et al, 2002). Oral administration of gefitinib at 50 or 100 mg kg À1 day À1 for 3 weeks in athymic mice with established intracerebral A431 human epidermoid carcinoma expressing EGFR increased median survival by 88% (P ¼ 0.009) and 105% (Po0.001), respectively. The preliminary results of a Phase II study also show modest activity of gefitinib in patients with recurrent glioblastoma (Peery et al, 2003).
Recently, a number of case reports of patients with recurrent NSCLC treated with gefitinib as part of the Expanded Access Programme (EAP) have described promising activity in patients with brain metastases. Here, we accumulate these results and discuss the possible implications and future directions.

GEFITINIB IN PATIENTS WITH BRAIN METASTASES: RESULTS FROM THE EAP
The case reports of patients with brain metastases secondary to NSCLC treated with gefitinib as part of the compassionate-use programme have recently been reported as published case studies (Cappuzzo et al, 2003a, b;Fujiwara et al, 2003;Knuti et al, 2003;Villano et al, 2003), at the World Conference on Lung Cancer 2003 (Chiu et al, 2003;Pino et al, 2003), at the American Society of Clinical Oncology 2003 conference (Ceresoli et al, 2003)  suggest that gefitinib has activity in NSCLC brain metastases. The results of published case series are shown in Table 1.
Most notable are the cases with a complete remission of brain metastases. In the largest case series that included 27 patients who had asymptomatic or symptomatic brain metastases, two of 20 evaluable patients had a response in the brain, with complete remission in one patient (Ceresoli et al, 2003). In another case series described in two separate publications by Cappuzzo et al, eight patients with NSCLC and brain metastases were treated with gefitinib (Cappuzzo et al, 2003a, b; ICE abs). One patient had a complete response and seven had partial responses in the brain, all within 3 months of treatment. The complete response in the brain was observed 6 weeks after treatment with gefitinib started in a patient who had completed whole-brain radiotherapy 3 months prior to the beginning of gefitinib ( Figure 1). All eight patients experienced improvements in neurological and systemic symptoms. Fujiwara et al (2003) described the case report of a patient who had achieved a complete response in the brain following treatment with gefitinib. This particular patient was diagnosed in July 1999 with stage IV adenocarcinoma of the left lung. In March 2001, after treatment with three different chemotherapy regimens, asymptomatic brain metastases were detected and subsequently treated with whole-brain radiotherapy. However, by October 2001 the patient's condition had deteriorated, with progressive brain metastases, and the patient was readmitted to hospital. Treatment with gefitinib started in November 2001 after which computed tomography scans showed that brain metastases had disappeared and that there was a marked regression of the lung tumour. The patient was then discharged in December 2001 and was able to resume full-time work for nearly 8 months. Two other published case reports also describe patients whose brain metastases responded to treatment (Knuti et al, 2003;Villano et al, 2003). In all three of these cases, patients had worsening neurological symptoms and deteriorating performance status before treatment with gefitinib. Following treatment, a significant clinical improvement was seen in all three patients: brain metastases disappeared in the patient already described and the sizes of the other two patients' lesions decreased. Quality of life was improved in all three patients.
One patient described at the ICE meeting showed good tolerance of concurrent gefitinib and radiotherapy (van Zandwijk [b], ICE abs, personal communication). In this case study, a 55-year-old male exsmoker with stage IV NSCLC and multiple pulmonary (lymphangitic) metastases developed multiple brain metastases (approximately 20 small lesions) after receiving treatment with gefitinib for 41 year (Figure 2). Radiotherapy to the brain (10 Â 3 Gy) and continued treatment with gefitinib resulted in disease stabilisation and symptom improvement and the patient was able to return to work. Delayed imaging 9 months later showed almost complete resolution of the brain metastases ( Figure 3). Thus, although gefitinib did not prevent the development of brain metastases, concurrent treatment with radiotherapy and gefitinib resulted in a durable clinical response.
In total, 25 case reports of gefitinib use in patients with brain metastases and NSCLC were described at the ICE meeting (including the eight patients described by Cappuzzo Table 1 Published case series of patients with brain metastases and non-small-cell lung cancer treated with gefitinib as part of the Expanded Access Programme

Patients Previous treatment Results
Pino et al (2003) n ¼ 63, of whom 11 patients One prior chemotherapy regimen: 18 Of 11 patients with brain metastases, had brain metastases Two prior chemotherapy regimens: 42 two had a response in the brain Chiu et al (2003) n ¼ 57, of whom 16 had brain metastases One prior chemotherapy regimen: 1 Of 16 patients with brain metastases, X2 prior chemotherapy regimens: 15 two had a partial response Ceresoli et al (2003) n ¼ 27, all with brain metastases Platinum-based chemotherapy: 20 Two patients had a response in the brain, Brain radiotherapy: 11 one with complete remission (both had brain radiotherapy 43 months before starting gefitinib) Cappuzzo et al (2003a) n ¼ 4, all with brain metastases X2 prior chemotherapy regimens: 4 Whole-brain radiotherapy: 3 One patient had a complete response and three patients had a partial response of brain metastases Response duration: 3+, 5, 7, 12+ months Cappuzzo et al (2003b) n ¼ 4, all with brain metastases One prior chemotherapy regimen: 2 X2 prior chemotherapy regimens: 2 All four patients had a partial response of brain metastases Whole-brain radiotherapy: 2 Response ICE abs; Kowalski, ICE abs; Martín-Algarra, ICE abs; Pavlakis, ICE abs). Of these, three patients had a complete response, 10 a partial response, eight had stable disease and four patients had progression of their brain metastases. In all of these case reports, treatment with gefitinib appears to have been well tolerated. Any reported adverse events are consistent with the adverse-event profile of gefitinib established in Phase II trials in patients with advanced NSCLC (Fukuoka et al, 2003;Kris et al, 2003). There was also no unexpected or cumulative toxicity described for the patient who received concurrent gefitinib and radiotherapy (van Zandwijk [b], ICE abs).

DISCUSSION
A number of case reports suggest that gefitinib has clinical activity against brain metastases of patients with NSCLC. These preliminary data indicate that patients with brain metastases should not be excluded from treatment with gefitinib (as they were in some of the early clinical trials of gefitinib), an opinion also expressed by Cappuzzo et al (2003b). There is a need for larger trials to followup these positive preliminary findings in the hope of developing a treatment option for these patients who have a very poor prognosis. Combination of radiotherapy and gefitinib administered either concurrently or sequentially warrants further investigation.
A consideration for such future studies in patients with brain metastases is the optimum dose of gefitinib if dexamethasone is also being administered. Dexamethasone is an inducer of cytochrome P450 3A isoenzymes, including CYP 3A4, which has been identified as the isoenzyme primarily involved in the metabolism of gefitinib. Therefore, coadministration of dexamethasone may lead to increased metabolism of gefitinib, with consequently lower plasma concentrations and potentially reduced efficacy. In addition, such studies would need to use standardised patient staging to identify and monitor asymptomatic as well as symptomatic brain metastases.
A number of preclinical studies support the combined use of radiotherapy and gefitinib as a strategy for cancer treatment (Bianco et al, 2002;Huang et al, 2002;Magne et al, 2002;Williams et al, 2002;Solomon et al, 2003). Coadministration of gefitinib potentiated the antitumour effect of radiation on A431 cells  (human vulvar squamous-cell carcinoma cells that express high levels of EGFR) both in vitro and in vivo (Solomon et al, 2003). Another preclinical study showed a cooperative antiproliferative effect when gefitinib was combined with radiation in colon, ovarian, NSCLC and breast cancer cell lines (Bianco et al, 2002). Further, the combination of gefitinib and radiation resulted in long-term tumour growth regressions of established human colon cancer GEO and lung adenocarcinoma A549 xenografts (Bianco et al, 2002). Radiotherapy treatment may also disrupt the bloodbrain barrier (van Vulpen et al, 2002) that would allow improved access of gefitinib to brain tumour tissue.
In summary, a number of case reports suggest activity of gefitinib against brain metastases secondary to advanced NSCLC. Larger trials are needed to validate these findings and further investigate the potential of combined gefitinib and radiotherapy treatment.