The precise management of a gefitinib responder having severe adverse events remains unknown. There are only two case reports on readministration of gefitinib in responders following treatment discontinuation due to severe gefitinib-related hepatotoxicity [6, 7]. Although the resumption of gefitinib (250 mg/day) and concurrent steroid therapy  failed to control hepatotoxicity, intermittent schedule of gefitinib administration (250 mg/day once every 5 days)  not only successfully reduced hepatotoxicity but also induced disease regression.
To the best of our knowledge, this is the second report of readministration of gefitinib in a patient who had once developed gefinitib-related ILD. Yano et al  previously reported the readministration of gefitinib in a responder after discontinuation owing to ILD(alveolar hemorrhage). Starting with 250 mg/day, gefitinib was given every other day after blepharitis developed at the time of initial treatment. This intermittent schedule of gefitinib administration was effective both for the initial treatment and the resumed treatment.
The currently approved dose of gefitinib was determined based on the data from two large phase II trials (IDEAL 1 and 2). Similar efficacy but higher toxicity was observed with the 500 mg dose. Consequently, 250 mg has been the recommended dose [9, 10]. However, anti-tumor activity was not necessarily dependent on the dose of gefitinib in previous trials [1, 2, 9, 10]. Even when the dose of gefitinib was less than 250 mg/day, several responders were documented [1, 2]. However, the development of adverse effects other than ILD is usually dependent on the dose of gefitinib [1, 2]. Therefore, re-treatment by a reduced dose may be effective for patients who had previously responded to standard dose of gefitinib but had discontinued treatment due to the occurrence of severe adverse events other than ILD. Due to the low incidence of gefinitib-related ILD, it is difficult to show whether the same hypothesis fit to the gefinitib-related ILD based on the observational studies. Although there are no data indicating the dose-dependency of ILD, we could successfully manage a patient who had once developed ILD with a half dose of gefitinib.
Numerous studies suggest that gefitinib has fairly effective anti-tumor activity, especially for tumors with EGFR gene mutations [11, 12]. It is well-known that EGFR-mutated tumors respond dramatically to lower dose of EGFR-TKIs such as gefitinib in experimental studies [13, 14]. These findings may suggest that a reduced dose of gefitinib is sufficient for tumors with these molecular characteristics, and that the currently approved dose is excessively potent in some of these patients and results in adverse events.
The prognosis of patients with lymphangitis carcinomatosa is extremely poor, with approximately 50% of the patients dying within three months of their first respiratory symptoms . Fortunately, our patient was able to spend the rest of his life at home without the relapse of ILD for 16 months after readministration of gefitinib. Therefore, a resumption of gefitinib therapy would be clinically beneficial for him. However, secondary treatment with gefitinib for a patient who once developed gefinitib-related ILD is usually considered very risky. Therefore, at this time, alternative therapeutic modalities should be chosen, if available.