The in vitro effect of gefitinib ('Iressa') alone and in combination with cytotoxic chemotherapy on human solid tumours

Abstract

Background: Activation of the epidermal growth factor receptor (EGFR) triggers downstream signaling pathways that regulate many cellular processes involved in tumour survival and growth. Gefitinib ('Iressa') is an orally active tyrosine kinase inhibitor (TKI) targeted to the ATP-binding domain of EGFR (HER1; erbB1).

Methods: In this study we have used a standardised ATP-based tumour chemosensitivity assay (ATP-TCA) to measure the activity of gefitinib alone or in combination with different cytotoxic drugs (cisplatin, gemcitabine, oxaliplatin and treosulfan) against a variety of solid tumours (n = 86), including breast, colorectal, oesophageal and ovarian cancer, carcinoma of unknown primary site, cutaneous and uveal melanoma, non-small cell lung cancer (NSCLC) and sarcoma. The IC50 and IC90 were calculated for each single agent or combination. To allow comparison between samples the IndexSUM was calculated based on the percentage tumour growth inhibition (TGI) at each test drug concentration (TDC). Gefitinib was tested at concentrations ranging from 0.0625-2 microM (TDC = 0.446 microg/ml). This study represents the first use of a TKI in the assay.

Results: There was heterogeneity in the degree of TGI observed when tumours were tested against single agent gefitinib. 7% (6/86) of tumours exhibited considerable inhibition, but most showed a more modest response resulting in a low TGI. The median IC50 value for single agent gefitinib in all tumours tested was 3.98 microM. Interestingly, gefitinib had both positive and negative effects when used in combination with different cytotoxics. In 59% (45/76) of tumours tested, the addition of gefitinib appeared to potentiate the effect of the cytotoxic agent or combination (of these, 11% (5/45) had a >50% decrease in their IndexSUM). In 38% of tumours (29/76), the TGI was decreased when the combination of gefitinib + cytotoxic was used in comparison to the cytotoxic alone. In the remaining 3% (2/76) there was no change observed.

Conclusion: The in vitro model suggests that gefitinib may have differential effects in response to concomitant cytotoxic chemotherapy with the agents tested during this study. The mechanism involved may relate to the effect of TKIs on growth rate versus their effect on the ability of the cell to survive the stimulus to apoptosis produced by chemotherapy.

Figure 1

Median effect of gefitinib on tumour-derived cells compared with a sensitive and non-sensitive colorectal tumour. Error bars show 25^th and 75^th inter-quartile range.

Figure 2

Frequency histogram showing heterogeneity of the Index_SUM for gefitinib alone in all tumours tested (n = 86).

Figure 3

The effect of gefitinib on tumour-derived cells from recurrent ovarian cancer (n = 13). Legend Some tumours show increased sensitivity (lower Index_SUM), while others show enhanced resistance (higher Index_SUM)

Figure 4

Tumour growth inhibition by gefitinib vs. treosulfan + gefitinib in 2 different tumour types. Legend Figure 4a shows the positive effect of combining treosulfan plus gefitinib in cells derived from a skin melanoma compared with figure 4b in which the combination of treosulfan plus gefitinib has a negative effect in cells derived from an ovarian tumour.

Figure 5

Tumour growth inhibition by gefitinib vs. gefitinib + oxaliplatin in a NSCLC. Legend The combination of oxaliplatin + gefitinib has a positive effect compared with single agent oxaliplatin which is inactive.