Pharmacological characterization of MP‐412 (AV‐412), a dual epidermal growth factor receptor and ErbB2 tyrosine kinase inhibitor

Abstract

Epidermal growth factor receptor (EGFR) and ErbB2 are currently recognized as validated target molecules in cancer treatment strategies. MP‐412 (AV‐412) is a potent dual inhibitor of EGFR and ErbB2 tyrosine kinases, including the mutant EGFR^L858R,T790M, which is clinically resistant to the EGFR‐specific kinase inhibitors erlotinib and gefitinib. In an enzyme assay, MP‐412 inhibited the EGFR variants and ErbB2 in the nanomolar range with over 100‐fold selectivity compared with other kinases, apart from abl and flt‐1, which were both moderately sensitive to the compound. In cells, MP‐412 inhibited autophosphorylation of EGFR and ErbB2 with IC₅₀ of 43 and 282 nM, respectively. It also inhibited epidermal growth factor (EGF)‐dependent cell proliferation with an IC₅₀ of 100 nM. Moreover, MP‐412 abrogated EGFR signaling in the gefitinib‐resistant H1975 cell line, which harbors a double mutation of L858R and T790M in EGFR. In animal studies using cancer xenograft models, MP‐412 (30 mg/kg) demonstrated complete inhibition of tumor growth of the A431 and BT‐474 cell lines, which overexpress EGFR and ErbB2, respectively. MP‐412 suppressed autophosphorylation of EGFR and ErbB2 at the dose corresponding to its antitumor efficacy. When various dosing schedules were applied, MP‐412 showed significant effects with daily and every‐other‐day schedules, but not with a once‐weekly schedule, suggesting that frequent dosing is preferable for this compound. Furthermore, MP‐412 showed a significant antitumor effect on the ErbB2‐overexpressing breast cancer KPL‐4 cell line, which is resistant to gefitinib. These studies indicate that MP‐412 has potential as a therapeutic agent for the treatment of cancers expressing EGFR and ErbB2, especially those resistant to the first generation of small‐molecule inhibitors. (Cancer Sci 2007; 98: 1977–1984)

Epidermal growth factor receptor (EGFR) and ErbB2 are members of the ErbB receptor family of type I tyrosine kinases that are overexpressed in various human solid cancers, and are often associated with malignancy as well as poor clinical outcomes.^{( 1} , ^{2 )} Nowadays, both receptors have been recognized as being among the most attractive and common targets for cancer therapy. Regarding small‐molecule compounds, the selective EGFR tyrosine kinase inhibitors (TKI) erlotinib and gefitinib showed objective response rates of 9–18% in previously treated or untreated advanced non‐small cell lung cancer (NSCLC) in their clinical trials, and have been approved for clinical use.^{( 3} , ^{4 )}

Recently, somatic mutations in the EGFR gene were identified in a subset of NSCLC patients and were found to be associated with sensitivity to erlotinib and gefitinib.^{( 5} , ^{6 )} In particular, a deletion in exon 19 and a L858R substitution in exon 21 accounted for approximately 85% of reported sensitizing mutations. Importantly, a T790M point mutation in exon 20 was found in approximately 50% of patients with NSCLC who relapsed after treatment with EGFR TKI.^{( 7 )} Therefore, examining mutations seems useful for the response prediction of TKI, and at the same time, acquired resistance evokes potential limits of TKI efficacy. There are several reports demonstrating that the irreversible EGFR inhibitors CL‐387 785 and HKI‐272 are effective against the T790M mutant in vitro.^{( 8} , ^{9 )}

Epidermal growth factor receptor forms a heterodimer with ErbB2 to amplify its signal transduction.^{( 10 )} This crosstalk signaling after heterodimerization elicits synergistic mitogenicity and invasiveness, whereas EGFR homodimerization does not.^{( 11} , ¹² , ^{13 )} In addition to this biological implication, there is clinical evidence indicating that cancers overexpressing both receptors have a worse outcome than those overexpressing either receptor alone.^{( 14} , ¹⁵ , ¹⁶ , ^{17 )} These observations suggest that a dual inhibitor of EGFR and ErbB2 should provide benefits to patients suffering from cancers expressing both receptors rather than either one.

We have developed a new dual EGFR and ErbB2 inhibitor, MP‐412 (AV‐412), possessing activity against EGFR^L858R,T790M in both enzyme assays and in vitro. This compound showed significant antitumor effects on both EGFR‐ and ErbB2‐overexpressing cancer xenografts, including cells resistant to gefitinib.

Materials and Methods

Chemicals.  MP‐412 and gefitinib, synthesized at Mitsubishi Pharma (Yokohama, Japan), were dissolved in dimethylsulfoxide for the in vitro assays or suspended in 0.5% aqueous gum tragacanth solution for the in vivo studies.

Kinase assays.  Recombinant intracellular kinase domains of EGFR, EGFR^L858R, EGFR^L858R,T790M, EGFR^T790M, and purified EGFR from A431 cell membranes were used. Kinase reactions were carried out in 8 mM MOPS (pH 7.0), 0.2 mM ethylenediaminetetraacetic acid (EDTA), 10 mM MnCl₂, 10 mM Mg acetate, 0.1 mg/mL poly(Glu, Tyr) 4 : 1, [γ³³P‐ATP], and 5–10 mU of enzyme, except that 250 µM of the GGMEDIYFEFMGGKKK peptide substrate was used for EGFR^T790M. Phosphorylation was initiated by the addition of ATP and was allowed to proceed for 40 min at room temperature. The reaction was stopped by the addition of 3% phosphoric acid, then aliquots of the reaction mixture were spotted onto a filtermat. After rinsing to remove peptides bound non‐specifically, the filter was scintillation counted.

The activity of the recombinant intracellular kinase domain of ErbB2 was measured using time‐resolved fluorometry (DELFIA; Perkin‐Elmer Life Sciences, Boston, MA, USA). Kinase reactions were carried out in 60 mM HEPES (pH 7.4), 5 mM MnCl₂, 5 mM MgCl₂, 1.2 mM dithiothreitol, 50 µg/mL polyethylene glycol (PEG)_20 000, 3 µM Na₃VO₄, 1 µM biotinylated poly(Glu, Ala, Tyr), 50 µM ATP, and 150 ng of the enzyme. Phosphorylation was initiated by the addition of ATP and was allowed to proceed for 60 min at room temperature. The reaction was stopped by the addition of EDTA solution, and aliquots of the reaction mixture were transferred to a streptavidin‐coated plate. Phosphorylation was detected using Europium‐labeled antiphosphotyrosine antibody (Cell Signaling Technology, Beverly, MA, USA). After washing of the plate and the enhancement steps, carried out in accordance with the manufacturer's recommendations, the signal was detected using an ARVO sx fluorescence reader (Wallac/Perkin Elmer, Boston, MA, USA).

The activities of the recombinant kinase domains of Abl, Flt‐1, and Src were assayed using the fluorescence polarization method as described previously with poly(Glu, Tyr) peptide as a substrate.^{( 18 )} Phosphorylation was carried out at 22–30°C for 10–20 min.

The activities of the recombinant kinase domains of insulin receptor kinase (IRK) and mitogen‐activated kinase (MAP) kinase kinase (MEK1), as well as those of purified protein kinase A (PKA) from bovine heart and protein kinase C (PKC) prepared from rat brain, were determined using [γ³³P‐ATP] as described above. Substrates used were poly(Glu, Tyr) (IRK), inactivated MEK1, and histone H1 (PKA, PKC). Phosphorylation was carried out at 22–30°C for 20–45 min.

Cell lines.  The human epidermoid cancer cell line A431, the human esophageal cancer cell line TE‐8, and A4 cells that were transfected with c‐erbB2^V659E to mouse embryonal fibroblast NIH/3T3 were obtained from the Cell Resource Center for Biomedical Research (Tohoku University, Sendai, Japan) and cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). The following cell lines were obtained from American Type Culture Collection (Rockville, MD, USA): human breast cancer cell lines BT‐474 and T‐47D, human NSCLC cell lines NCI‐H1650 and NCI‐H1975 (designated H1650 and H1975, respectively), and rat smooth‐muscle cell line A7r5 from thoracic aorta. Cells were cultured in RPMI‐1640 or DMEM–Ham's F12 medium supplemented with 10% FBS. The human breast cancer cell line KPL‐4 was kindly provided by Dr Junichi Kurebayashi (Kawasaki Medical University, Kurashiki, Japan) and cultured in DMEM–Ham's F12 medium supplemented with 10% FBS.

Cell proliferation assay.  To test the effects of MP‐412 on growth factor‐dependent cell proliferation, A431 and A7r5 cells were cultured for 24 h at 37°C in the presence of 1 ng/mL epidermal growth factor (EGF; Invitrogen, Carlsbad, CA, USA) and 50 ng/mL platelet‐derived growth factor (PDGF)‐BB (Sigma, St Louis, MO, USA), respectively. The ³H‐thymidine incorporation during this period was measured as described previously.^{( 19} , ^{20 )}

Western blotting.  For analysis of EGFR autophosphorylation and its downstream signaling, cells were serum starved overnight and incubated with various concentrations of compounds for 2 h, then stimulated with 20–100 ng/mL EGF (BioSource International, Camarillo, CA, USA) for 5–10 min. For evaluation of the reversibility of inhibition, culture medium containing the compound was removed after 2 h incubation and replaced with fresh medium lacking the compound. After further incubation for 6 or 24 h, cells were stimulated with 100 ng/mL EGF for 5 min. For analysis of ErbB2 autophosphorylation, A4 cells were incubated with various concentrations of compounds for 2 h in the presence of 10% FBS.

After treatment, cells were washed twice in PBS and lysed in RIPA lysis buffer (50 mM Tris‐HCl [pH 7.4], 150 mM NaCl, 0.25% deoxycholic acid, 1% nonidet P‐40, 1 mM EDTA) containing 1 mM Na₃VO₄, 1 mM NaF, and protease inhibitor cocktail (Roche, Mannheim, Germany). The amount of protein in cell lysates was determined using a BCA protein assay kit (Pierce, Rockford, IL, USA). The proteins were separated by sodium dodecylsulfate–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes. After blocking, blots were incubated with antibody solutions followed by washing and incubation with secondary antibodies, then developed using chemiluminescence reagents. Blots were scanned and quantified using LAS‐1000 Plus lumino‐image analyzer (Fuji Photo Film, Tokyo, Japan). The following antibodies were used: phospho‐EGFR Y1068 (BioSource International), EGFR (Upstate Biotechnologies, Lake Placid, NY, USA), phosphotyrosine PY20 (BD Biosciences, San Jose, CA, USA), c‐erbB‐2 Ab‐15 (NeoMarkers, Fremont, CA, USA), phospho‐ErbB2 Y1221, Akt, phospho‐Akt S473, p44/42 MAP kinase, and phospho‐p44/42 MAP kinase T202/Y204 (Cell Signaling Technology).

Tumor xenograft models.  BALB/c nu/nu (athymic) mice and ICR strain severe combined immunodeficiency (ICR‐SCID) mice were purchased from CLEA Japan (Tokyo, Japan). The care and treatment of experimental animals were in accordance with institutional guidelines. Tumor cells were implanted subcutaneously into the right flanks of mice. Treatment was initiated after the tumor volume had reached 100–300 mm³, followed by allocation of the animals to equalize the average of and variations in tumor volume across the treatment groups. Tumor volume ([length × width²]/2) and bodyweight were measured two times per week. The compounds or vehicle were administered once daily by oral gavage for 14 days. For studies examining the dosing schedule in relation to efficacy against TE‐8 tumors, MP‐412 was administered either once daily, every other day, or once per week for 2 weeks. Mice were killed 1 day after the final treatment, and the tumors were dissected and weighed. For evaluation of tumor phosphorylation, tumor‐bearing mice were given a single administration of MP‐412 and tumors were dissected 4 h later. Tumor lysates were prepared in a similar manner to that described above, except that tumors were homogenized in RIPA lysis buffer using a grinder with a plastic pestle.

Statistical analysis.  To evaluate the antitumor effects of the compounds, a Dunnett's multiple‐comparison test or t‐test was carried out to compare the treated groups with the controls. To estimate the IC₅₀ values and their 95% confidence intervals, non‐linear regression analysis was carried out. All analyses were done using the SAS system, and tests were two sided with a significance level of <0.05.

Results

Kinase specificity of MP‐412.  The kinase inhibition profiles of MP‐412 (Fig. 1) against a panel of tyrosine and serine/threonine kinases are shown in Table 1. MP‐412 inhibited all variants of EGFR kinase (IC₅₀ 0.5–2 nM) as well as ErbB2 and Abl (IC₅₀ 19 and 41 nM, respectively), and weakly inhibited Flt‐1 and src (IC₅₀ 920 and 2000 nM, respectively). MP‐412 did not affect IRK, MEK1, PKC, or PKA up to 10 µM. When the kinase specificity of this compound was further investigated across 147 different protein kinases at 100 nM, it was found that only four enzymes, ErbB4, Blk, Lyn, and EphB4, other than the aforementioned enzymes were susceptible to MP‐412 with inhibition exceeding 50%.^{( 21 )} One of the characteristics of the chemical structure of MP‐412 is a reactive acrylamide functional group that potentially forms a covalent bond with a cysteine residue within the ATP pocket in EGFR and ErbB2. Based on our docking study, MP‐412 is expected to bind to Cys797 in EGFR and Cys808 in ErbB2, both located near the gate of the ATP pockets (Suppl. Fig. S1). The docking model also showed that MP‐412 occupied a similar space to gefitinib in the kinase domain of EGFR (Suppl. Fig. S2).

Figure 1.

The chemical structure of MP‐412 (2 tosylate salt).

Table 1.

Kinase inhibition profile of MP‐412

Kinase	IC50 (µM)	95% confidence interval
EGFR (from A431 membrane)	0.00075	0.00054–0.0011
EGFR (recombinant)	0.0014	0.0011–0.0018
EGFR L858R	0.00051	0.00040–0.00066
EGFR T790M	0.00079	0.00072–0.00086
EGFR L858R, T790M	0.0023	0.0020–0.0026
ErbB2	0.019	0.012–0.028
Abl	0.041	0.035–0.047
Flt‐1	0.92	0.64–1.3
Src	2.0	1.5–2.7
IRK	>10	–
MEK1	>10	–
PKA	>10	–
PKC	>10	–

EGFR, epidermal growth factor receptor; IRK, insulin receptor kinase; MEK, mitogen activated protein (MAP) kinase kinase; PKA, protein kinase A; PKC, protein kinase C.

The effect of MP‐412 on cell proliferation stimulated by EGF or PDGF was examined by [³H]thymidine incorporation for 24 h (Table 2). MP‐412 inhibited EGF‐induced mitogenesis of A431 cells with an IC₅₀ of 0.1 µM. In contrast, a 40‐times‐higher concentration was required for inhibition of PDGF‐induced mitogenesis of A7r5 rat smooth‐muscle cells with an IC₅₀ of 4.1 µM. When the effects of MP‐412 on A431 cell growth in the presence of 10% FBS for 3 days was examined, the IC₅₀ was 0.69 µM, which was still lower than that of the above‐mentioned PDGF‐induced mitogenesis.

Table 2.

Effects of MP‐412 on growth factor‐dependent cell proliferation

Cell	IC50 (µM)	95% confidence interval
EGF‐stimulated A431	0.10	0.05–0.20
PDGF‐stimulated A7r5	4.1	2.0–8.7

EGF, epidermal growth factor; PDGF, platelet‐derived growth factor.

Epidermal growth factor receptor and ErbB2 dual inhibition in cells.  Expression of EGFR and ErbB2 was analyzed by western blotting for all cancer cell lines used in the present study (Suppl. Fig. S3). When A431 cells, which overexpress EGFR, were treated with MP‐412, it produced a dose‐dependent inhibition of EGFR autophosphorylation with an IC₅₀ of 43 nM, and gefitinib showed similar inhibitory activity with an IC₅₀ of 33 nM (Fig. 2a). For evaluation of ErbB2 autophosphorylation, A4 cells were used because the expression of EGFR was not detected in the parental NIH/3T3 cell line,^{( 22 )} and the influence of trans‐phosphorylation between EGFR and ErbB2 was considered to be negligible. When A4 cells were treated with MP‐412, it produced a dose‐dependent inhibition of ErbB2 autophosphorylation with an IC₅₀ of 282 nM and complete inhibition at 1000 nM. In contrast, gefitinib showed less‐potent activity with an IC₅₀ of 1370 nM. In addition, the inhibition curve of gefitinib formed a blunted slope, which was distinct from that of MP‐412 (Fig. 2b). The duration of inhibition of EGFR autophosphorylation was investigated by washout studies using A431 cells. Only a 2 h exposure of MP‐412 at 1 µM kept blocking EGFR autophosphorylation for 24 h after washout from the medium (Fig. 2c).

Figure 2.

Inhibition of epidermal growth factor receptor (EGFR) and ErbB2 autophosphorylation in cells. (a) Serum‐starved A431 cells were incubated with the compounds for 2 h, then stimulated with 20 ng/mL epidermal growth factor (EGF) for 5 min. (b) A4 cells were incubated with the compounds for 2 h in the presence of 10% fetal bovine serum. Tyrosine phosphorylation of EGFR and ErbB2 was analyzed by western blotting and quantitated by densitometry. Each data point represents the mean ± SEM of three independent experiments. The curve‐fitting of the plot was generated using non‐linear regression analysis. (c) Serum‐starved A431 cells were incubated with 0.1 (open column) or 1 µM (closed column) MP‐412 for 2 h, then washed twice with fresh medium, and incubated for a further 6 or 24 h in the absence of the compound. After stimulation with 100 ng/mL EGF for 5 min, protein extracts were analyzed by western blotting and phosphorylation was quantitated by densitometry. The results are representative of two independent experiments.

Next we examined the ability of MP‐412 and gefitinib to inhibit EGFR signaling in NSCLC cell lines harboring either the delE746‐A750 or L858R + T790M mutations using H1650 and H1975 cell lines, respectively. MP‐412 inhibited EGFR autophosphorylation and the consequent activation of Erk1/2 and Akt in both cell lines (Fig. 3). In contrast, gefitinib showed comparable activity to MP‐412 only in H1650 cells, and marginal activity in H1975 cells up to 10 µM (Fig. 3). In a study using H1975 cells, phosphorylation of Akt was partially suppressed after treatment with either compound, regardless of their concentrations. We also examined the effects of MP‐412 on downstream signaling of ErbB2 in T‐47D cells stimulated with heregulin‐β1. MP‐412 inhibited ErbB2 autophosphorylation at 0.3 µM or higher, with decreased phosphorylation of Erk1/2 and Akt (Suppl. Fig. S4).

Figure 3.

Inhibition of epidermal growth factor receptor (EGFR) signaling in H1650 and H1975 cells. Serum‐starved cells were incubated with the compounds for 2 h, then stimulated with 100 ng/mL epidermal growth factor (EGF) for 10 min. Protein extracts were analyzed by western blotting with the antibodies indicated on the left.

Potent inhibition of tumor growth by MP‐412.  The antitumor effects of MP‐412 were evaluated using both EGFR‐overexpressing A431 and ErbB2‐overexpressing BT‐474 tumor xenograft models. In both cases, MP‐412 showed complete inhibition of tumor growth at 30 mg/kg (Fig. 4a,b), whereas there was <10% bodyweight loss in animals treated at these doses (data not shown). The antitumor effects of MP‐412 on the other EGFR‐overexpressing cancer cell line, TE‐8, were investigated using various dosing frequencies, with the same total dose divided into one, three, or seven administrations per week. Significant inhibition of tumor growth was observed with every‐other‐day and once‐daily treatment. In contrast, when the schedule was less frequent (i.e. given as a weekly regimen) the compound failed to show a significant antitumor effect (Fig. 4c). These results suggest that once‐daily administration is the optimal schedule for MP‐412, and that an every‐other‐day schedule is also effective.

Figure 4.

Inhibition of tumor growth by MP‐412 in animal models. (a) A431 cells (5 × 10⁶) were inoculated subcutaneously into athymic mice. Four days later (day 0), MP‐412 was administered by oral gavage once daily for 14 days (n = 6). (b) BT‐474 cells (1 × 10⁷) were inoculated subcutaneously into SCID mice. Fifteen days later (day 0), mice (n = 4) were treated with MP‐412 as described in (a). (c) TE‐8 cells (5 × 10⁶) were inoculated subcutaneously into athymic mice. Five days later (day 0), MP‐412 was administered by oral gavage either once daily at 10 mg/kg, three times weekly at 23 mg/kg, or once weekly at 70 mg/kg for 14 days (n = 6). Arrows indicate administration. **P < 0.01 versus vehicle control by Dunnett's multiple‐comparison test.

We observed spontaneous autophosphorylation of EGFR in HSC‐3 tumors and of ErbB2 in KPL‐4 tumors, but not in A431 and BT‐474. When in vivo target inhibition was examined using those tumors, autophosphorylation of EGFR or ErbB2 was suppressed by single administration of MP‐412 at a dose corresponding to its antitumor efficacy (Fig. 5). Next we examined whether MP‐412 has an advantage over gefitinib by using the ErbB2‐overexpressing KPL‐4 xenograft model. Both compounds were administered at the maximum tolerated dose (MTD) in order to see their maximum efficacy. After administration for 14 days, mice were killed and the tumors were dissected and weighed. MP‐412 showed significant antitumor effects at both 100 and 150 mg/kg, whereas gefitinib was ineffective even at the MTD, 225 mg/kg (Fig. 5). When the MTD was defined as ‘non‐lethal and not producing a bodyweight loss exceeding 20% during the treatment’, the MTD for MP‐412 and gefitinib reached 150 and 225 mg/kg, respectively. Therefore, MP‐412 demonstrated an advantage over the EGFR‐specific inhibitor gefitinib against the ErbB2‐overexpressing cancer KPL‐4.

Figure 5.

Inhibition of tumor growth (upper) and decreased receptor autophosphorylation in tumors (lower) in mice. (a) HSC‐3 cells (5 × 10⁶) were inoculated subcutaneously into athymic mice. Five days later (day 0), MP‐412 was administered for 14 days (n = 6). Mice were killed on day 14, and the dissected tumors were weighed. To assess in vivo epidermal growth factor receptor autophosphorylation in a separate experiment, tumor‐bearing mice were treated with a single administration of MP‐412 (n = 3) and 4 h later tumors were dissected. Homogenates were analyzed by western blotting. (b) KPL‐4 cells (5 × 10⁶) were inoculated subcutaneously into athymic mice. Seven days later (day 0), MP‐412 or gefitinib was administered for 14 days (n = 8). Mice were killed on day 14, and the dissected tumors were weighed. Evaluation of ErbB2 autophosphorylation in KPL‐4 tumors in mice (n = 3) was carried out as described in (a). *P < 0.05, **P < 0.01 versus vehicle control.

Discussion

We have developed a new anilinoquinazoline class of compound, MP‐412, which inhibits both EGFR and ErbB2 kinases at nanomole concentrations. Interestingly, MP‐412 showed equipotent activity against several variants of EGFR kinase, including L858R and T790M double mutants known to be resistant to erlotinib and gefitinib. This potency and specificity to EGFR and ErbB2 might be partly explained by a docking study using structural modeling. MP‐412 has an acrylamide functional group in the side chain, which is expected to approach a close position capable of forming a covalent bond with a thiol moiety of Cys797 in EGFR or Cys805 in ErbB2 in the ATP pocket of their kinase domains (Suppl. Fig. S1). The position of these cysteines is unique for EGFR and ErbB2, because there is no cysteines at equivalent positions in other tyrosine kinases such as src, platelet‐derived growth factor receptor (PDGFR), insulin receptor, or even ErbB3.^{( 23 )} Nonetheless, Fry et al. demonstrated that the irreversible ErbB inhibitor PD168393 formed covalent bond with Cys773 instead of Cys797 in EGFR, based on their analyses of peptide mapping and mass spectroscopy.^{( 24 )} Moreover, unlike gefitinib, MP‐412 is expected to show irreversible inhibition, even in T790M mutant EGFR, possibly by forming a covalent bond with cysteine, although the existence of steric hindrance for MP‐412 caused by methionine is predicted in this mutant. Therefore, further determination of the structure of the T790M mutant complexed with MP‐412 will be needed to reveal which position in the reactive side chain of MP‐412 really forms a covalent bond. Very recently, Yun et al. elucidated crystal structures of the L858R and G719S mutants of EGFR complexed with TKI and demonstrated that these kinases were 50‐ and 10‐fold more active than wild type, respectively, and that gefitinib bound 20 times more tightly to the L858R mutant than to wild type.^{( 25 )} Their results indicated that mutations in the kinase domain dramatically affected the enzyme conformation and its activity and also its binding affinity to inhibitors. However, there is so far no report revealing the crystal structure of EGFR containing the T790M mutation, so that the precise mechanism of inhibition by MP‐412 remains unclear. In the present report, MP‐412 inhibited autophosphorylation of EGFR in H1975 cells harboring EGFR^L858R,T790M; in contrast, gefitinib was ineffective up to 10 µM (Fig. 3). We are currently investigating whether MP‐412 demonstrates an advantage over other EGFR and ErbB2 TKI by using a xenograft model of this cell line.

MP‐412 showed antitumor effects on various cancer‐cell lines overexpressing either EGFR or ErbB2 with inhibition of receptor autophosphorylation in tumors at correlated dosages. When the pharmacokinetic profile of MP‐412 in athymic mice was examined, the plasma C_max reached 5 µM after a single administration at 30 mg/kg, and then the plasma level maintained its cellular IC₅₀ of EGFR autophosphorylation for 16 h (data not shown). This indicated that MP‐412 was significantly effective with a once‐daily schedule, despite the limited duration of drug exposure. Moreover, an intermittent dosing of MP‐412 at 23 mg/kg, given three times a week, also showed a significant antitumor effect against TE‐8. It is likely that sustained inhibition of receptor phosphorylation, as shown in Fig. 2c, may contribute to its antitumor effects. MP‐412 showed significant antitumor effects on two ErbB2‐overexpressing breast cancer cell lines, BT‐474 and KPL‐4, although 100 mg/kg was necessary in the KPL‐4 model, compared with 30 mg/kg in BT‐474. The reason for this is uncertain but it might be related to the nature of the cell lines. There are several mechanisms proposed for the development of resistance to ErbB‐targeted therapies (except for resistant mutations within kinase domains). For example, constitutive activation of phosphoinositide 3‐kinase (PI3K)–Akt signaling caused by phosphatase and tensin homolog (PTEN) deficiency contributes to resistance to gefitinib but not lapatinib.^{( 26} , ^{27 )} Also, activation of estrogen receptor signaling is related to lapatinib resistance.^{( 28 )} Furthermore, coexpression of insulin‐like growth factor (IGF)‐1 receptor in ErbB2 overexpression contributes to trastuzumab resistance.^{( 29 )} These results suggest that the sensitivity to MP‐412 might also be affected by some molecular events, whereas both the BT‐474 and KPL‐4 cell lines overexpress ErbB2. Moreover, in contrast to MP‐412, gefitinib was ineffective against KPL‐4 even at its MTD, suggesting the advantage of MP‐412 over the first generation of EGFR TKI. However, identification of the mechanism or biomarkers of MP‐412 will be important not only to understand how it works but also to predict the target population in clinical trials.

There are several dual TKI of EGFR and ErbB2 in clinical development,^{( 30 )} and one reversible‐type inhibitor, lapatinib, was very recently approved by the Food and Drug Administration for the treatment of patients with advanced or metastatic breast cancer whose tumors overexpress ErbB2. The sensitivity to lapatinib was predicted by overexpression of ErbB2 but not EGFR in a recent phase II study,^{( 31 )} suggesting that ErbB2 is more important than EGFR for the activity of lapatinib. BIBW‐2992 and HKI‐272 are irreversible‐type inhibitors in phase II studies. Dose‐limiting toxicities for BIBW‐2992 were diarrhea and skin rash,^{( 32 )} whereas the most frequent severe adverse events for HKI‐272 were diarrhea and asthenia, surprisingly lacking skin rash.^{( 33 )} Canertinib is also an irreversible‐type inhibitor and effective in preclinical cancer models,^{( 34 )} but phase II studies were eventually disappointing, and further development was precluded.^{( 30 )} In the present report, we focused on the differentiation of MP‐412 from the EGFR‐specific TKI gefitinib by using H1975 cells harboring EGFR^L858R,T790M and a xenograft model of KPL4 cells overexpressing ErbB2. The differences among existing dual TKI should also be addressed by comparing efficacies in preclinical studies, especially using cancer cells harboring resistant mutations in EGFR or ErbB2. But more importantly, we should also consider comparing the safety profiles of the compounds in cancer patients in the future and then identify who will receive maximum benefit from the therapy of MP‐412 coupled with consideration of the mutation status of EGFR and ErbB2. In conclusion, MP‐412 is a potent antitumor agent in a number of EGFR‐ and ErbB2‐overexpressing preclinical cancer models. On the basis of this favorable preclinical profile, this compound has recently entered phase I clinical trials in oncology.