Association between gain-of-function mutations in PIK3CA and resistance to HER2-targeted agents in HER2-amplified breast cancer cell lines.

BACKGROUND: The mechanism of resistance to human epidermal growth factor receptor 2 (HER2)-targeted agents has not been fully understood. We investigated the influence of PIK3CA mutations on sensitivity to HER2-targeted agents in naturally derived breast cancer cells. MATERIALS AND METHODS: We examined the effects of Calbiochem (CL)-387,785, HER2 tyrosine kinase inhibitor, and trastuzumab on cell growth and HER2 signaling in eight breast cancer cell lines showing HER2 amplification and trastuzumab-conditioned BT474 (BT474-TR). RESULTS: Four cell lines with PIK3CA mutations (E545K and H1047R) were more resistant to trastuzumab than the remaining four without mutations (mean percentage of control with 10 microg/ml trastuzumab: 58% versus 92%; P = 0.010). While PIK3CA-mutant cells were more resistant to CL-387,785 than PIK3CA-wild-type cells (mean percentage of control with 1 microM CL-387,785: 21% versus 77%; P = 0.001), CL-387,785 retained activity against BT474-TR. Growth inhibition by trastuzumab and CL-387,785 was more closely correlated with changes in phosphorylation of S6K (correlation coefficient, 0.811) than those of HER2, Akt, or ERK1/2. Growth of most HER2-amplified cells was inhibited by LY294002, regardless of PIK3CA genotype. CONCLUSIONS: PIK3CA mutations are associated with resistance to HER2-targeted agents. PI3K inhibitors are potentially effective in overcoming trastuzumab resistance caused by PIK3CA mutations. S6K phosphorylation is a possibly useful pharmacodynamic marker in HER2-targeted therapy.

In vitro thermo- and thermochemo-sensitivity of retinoblastoma cells from surgical specimens.

The sensitivity to heat and chemical modification of human retinoblastoma cells obtained from patients with primary retinoblastoma was studied in vitro by the human tumour colony assay established by Hamburger and Salmon in 1977. Retinoblastoma cells showed moderate sensitivity to 1 h of hyperthermia at 42 degrees C; the median T/C% (ratio of the colony number in treated vs. control dishes, x 100) under hyperthermia was 47.0% for 46 tumours studied. When tumours were treated with melphalan, cis-diamminedichloroplatinum (II), adriamycin, etoposide and teniposide at 37 degrees C and 42 degrees C, the median T/C% for each chemical agent was decreased significantly by concomitant hyperthermia. One-hour exposure of 38 tumours to melphalan, cis-diamminedichloroplatinum (II) (30 tumours), adriamycin (27 tumours), teniposide (22 tumours) and etoposide (20 tumours) at 37 degrees C gave median T/C%s of 9.5, 33.5, 16.0, 3.8 and 38.0%, respectively, while exposure at 42 degrees C gave values of 2.4, 8.2, 5.6, 1.0 and 6.6%, respectively. Combination of heat and chemical treatment with melphalan, cis-diamminedichloroplatinum (II) and etoposide appeared to be synergistic with median T/C%s that were significantly lower than the median T/C%s expected from a simple sum of their individual effects. These in vitro results suggest that combining the treatment modalities of hyperthermia and chemotherapy for primary retinoblastoma would be advantageous.

Ectopic p16(ink4) expression enhances CPT-11-induced apoptosis through increased delay in S-phase progression in human non-small-cell-lung-cancer cells.

A tumor-suppressor gene, p16(INK4), which is deleted or mutated in tumors, regulates cell-cycle progression through a G(1)-S restriction point by inhibiting CDK4(CDK6)/cyclin-D-mediated phosphorylation of pRb. We have found that ectopic p16(INK4) expression increased cellular sensitivity of human non-small-cell-lung-cancer (NSCLC) A549 cells to a selective growth-inhibitory effect induced by the topoisomerase-I inhibitor 11, 7-ethyl-10-[4-(1-piperidino)-1-piperidino] carbonyloxy camptothecin (CPT-11) in vitro. In this study, we observed enhanced apoptosis characterized by DNA fragmentation in A549 cells transfected with p16(INK4) cDNA (A549/p16-1) and treated with CPT-11. This apoptosis was suppressed by the inhibitor of interleukin-1beta-converting enzyme (ICE/caspase-1) or ICE-like proteases, Z-Asp-CH2-DCB, as determined by DNA fragmentation and proteolytic cleavage of poly(ADP-ribose) polymerase, a natural substrate for CPP32/caspase-3. In A549/p16-1 cells, cytosolic peptidase activities that cleaved Z-DEVD-7-amino-4-trifluoromethylcoumarin increased during CPT-11-induced apoptosis and were suppressed by a highly specific caspase-3 and caspase-3-like inhibitor, Z-DEVD-fluoromethylketone. These findings indicate that p16(INK) is positively involved in the activation pathway of the caspase-3 induced by CPT-11. The increased delay in S-phase progression and subsequent induction of apoptosis were observed in CPT-11-treated A549/p16-1 cells on the basis of DNA histograms. Specific down-regulation of the cyclin-A protein level in A549/p16-1 cells was observed after CPT-11-treatment, whereas cyclin B, cdk2, and cdc2 protein levels were unaffected. These results suggest that ectopic p16(INK4) expression inappropriately decreases cyclin A and thereby terminates CPT-11-induced G(2)/M accumulation, which is followed by increased apoptosis in p16(INK4)-expressing A549 cells.

An active efflux system for heavy metals in cisplatin-resistant human KB carcinoma cells.

The mechanism for cisplatin resistance in cisplatin-resistant KCP-4 cells was studied. Although multidrug resistance-associated protein (MRP) was not detected in KCP-4 cells, the cells were more resistant to heavy metals than multidrug-resistant C-A120 cells that overexpressed MRP. KCP-4 cells expressed metallothionein, but it was scarcely involved in cisplatin resistance in these cells. KCP-4 cells did not express canalicular multispecific organic anion transporter (cMOAT). The glutathione (GSH) level was 4.7-fold higher in KCP-4 cells than in KB-3-1 cells. When the GSH level in KCP-4 cells was decreased by treating the cells with buthionine sulfoximine and nitrofurantoin, the accumulation of and sensitivity to cisplatin in the cells were increased. C-A120 cells were only 3.0-fold more resistant to cisplatin than KB-3-1 cells and this resistance was not affected by the increased glutathione level. The accumulation of platinum in C-A120 and KCP-4 cells was 68.5 and 20.4% of that in KB-3-1 cells, respectively, while the intracellular levels of antimony potassium tartrate in C-A120 and KCP-4 cells were 13.2 and 9.9% of that in KB-3-1 cells, respectively. The ATP-dependent efflux of antimony was enhanced in both C-A120 and KCP-4 cells. These results, taken together, suggest an efflux pump for heavy metals different from MRP and cMOAT is involved in cisplatin resistance in KCP-4 cells.

Characterization of a human small-cell lung cancer cell line resistant to a new water-soluble camptothecin derivative, DX-8951f.

DX-8951f, a water-soluble and non-pro-drug analogue of camptothecin, exhibits a strong inhibitory action on DNA topoisomerase I (Topo I) and in vitro cytotoxicity against various human cancer cell lines. In order to elucidate the mechanisms of its cytotoxicity, we established a DX-8951f-resistant cell line, SBC-3/DXCL1, from human small cell lung cancer cells (SBC-3) by stepwise exposure to DX-8951f. SBC-3/DXCL1 cells were approximately 400 times more resistant to DX-8951f than parent cells. The SBC-3/DXCL1 cells showed a high degree of cross-resistance to other Topo I inhibitors such as CPT-11, SN-38 and camptothecin, but not to non-Topo I targeting agents such as cisplatin, adriamycin, etoposide, and vincristine. The mechanisms of resistance of SBC-3/DXCL1 cells to DX-8951f were examined. Intracellular accumulation of DX-8951f by SBC-3 and SBC-3/DXCL1 cells did not differ significantly. Although the Topo I activity of nuclear extracts obtained from SBC-3/DXCL1 cells was the same as that of the parent cells, the Topo I of SBC-3/DXCL1 cells was resistant to the inhibitory effects of DX8951f and SN-38. Immunoblotting using anti-Topo I antibody demonstrated similar protein levels of Topo I in SBC-3 and SBC-3/DXCL1 cells. The active Topo I protein of SBC-3/DXCL1 was eluted by a high concentration of NaCl (0.4 N) compared with that of SBC-3 (0.3 N). DX-8951f stabilized the DNA-Topo I cleavable complex from SBC-3 cells, as measured by Topo I-mediated cleavage assay. In SBC-3/DXCL1 cells, DX-8951f also stabilized the DNA-Topo I complex, but with a 10-fold lower efficiency. These results suggest that a qualitative change in Topo I contributes, at least partially, to the resistance to DX-8951f in SBC-3/DXCL1 cells. Therefore, SBC-3/DXCL1 cells may have a unique mechanism of resistance to Topo I-directed antitumor drugs.

Chronopharmacology of etoposide given by low dose prolonged infusion in lung cancer patients.

We examined the chronopharmacology of prolonged etoposide administration by intravenous infusion over 14 days in 9 patients with lung cancer. Blood samples were obtained every 4 hours between 24 hours and 48 hours after initiation of the infusion. Using the unpaired t-test, the percentage plasma etoposide concentration at 09:00 hours calculated from the 24 hour average value, was significantly higher than that at 21:00 hours (p = 0.024). However, neither ANOVA nor cosinor analysis revealed any significant effect of sampling time (ANOVA: p = 0.29, cosinor analysis: p = 0.46).

Detection of proteins that recognize platinum-modified DNA using gel mobility shift assay.

Using a gel mobility shift assay, we have identified proteins, in the nuclear extracts of a human lung cancer cell line, that recognize cis-diamminedichloroplatinum(II) (cis-DDP, CDDP)-modified DNA. A 158-base-pair double-stranded DNA fragment, derived from pBR322 plasmid DNA, was modified by either CDDP, tetrachloro(dl-trans)-1,2-diaminocyclohexaneplatinum(IV) (tetraplatin) or trans-DDP (the stereoisomer of CDDP and clinically ineffective). These platinum drug-modified probes were incubated with nuclear extracts and analyzed by gel mobility shift assay. Proteins in the extracts selectively recognized the clinically active platinum-modified DNA fragment. No binding to the trans-DDP-modified DNA fragment was observed. These proteins may play a role in the cytotoxicity or in a DNA repair process.

In vitro colony inhibition of carboplatin against stomach and lung cancer cell lines in comparison with cisplatin.

The effects of carboplatin and cisplatin on colony formation in stomach and lung cancer cell lines were examined and compared. The colony-inhibitory activity of carboplatin against stomach and lung cancer cell lines was similar to that of cisplatin when one-tenth of the peak plasma concentration of each drug was used (r = 0.80). One of the four stomach cancer cell lines was sensitive to carboplatin although all the stomach cancer cell lines were resistant to cisplatin. Of the three small cell lung cancer cell lines tested, two were sensitive to both carboplatin and cisplatin, and only one cell line (N857) was resistant to cisplatin; all the non-small cell cancer cell lines tested were resistant to both drugs. On the basis of these preliminary results, we suggest that carboplatin has potential therapeutic activity against stomach cancer and should be evaluated carefully from this aspect.

[Fundamental study and clinical testing of human tumor clonogenic assay (HTCA)].

Through our retrospective analysis of HTCA and clinical effects as well as fundamental studies, the following results were obtained: In this retrospective study, HTCA for lung cancer showed a predictive accuracy of 71%, a true positive rate of 50% and a true negative rate of 77%. To obtain good predictive accuracy, HTCA should be modified to provide conditions comparable to those in vivo with regard to drug concentration and drug exposure time. More precise analysis of the pharmacokinetics of anticancer agents might yield methodological improvement. A decrease in the chemosensitivity spectrum in vitro was observed after chemotherapy. This might be related to evidence that patients with prior chemotherapy exhibited a poor response rate to chemotherapy. There were no active anticancer agents against specimens with aplating efficiency of more than 0.04%. More extensive prospective trials will be necessary to determine the clinical value of HTCA. HTCA could be a superior assay for detecting the anti-tumor activity of new agents and a useful method for in vitro phase II study.