Jarmo Kulmala

Signaling via ErbB2 and ErbB3 Associates with Resistance and Epidermal Growth Factor Receptor (EGFR) Amplification with Sensitivity to EGFR Inhibitor Gefitinib in Head and Neck Squamous Cell Carcinoma Cells

Purpose: The epidermal growth factor receptor (EGFR) inhibitor gefitinib (Iressa) has shown antitumor activity in clinical trials against cancers, such as non–small cell lung cancer and head and neck squamous cell carcinoma (HNSCC). Research on non–small cell lung cancer has elucidated factors that may predict response to gefitinib. Less is known about molecular markers that may predict response to gefitinib in HNSCC patients. Experimental Design: We analyzed possible associations of responsiveness to gefitinib with molecular markers of the EGFR/ErbB receptor family signaling pathway using 10 established HNSCC lines in vitro. IC50 of gefitinib sensitivity was determined using clonogenic survival assays. ErbB signaling was assessed by Western and real-time reverse transcription-PCR analyses of EGFR, ErbB2, ErbB3, and ErbB4 expression levels as well as by phosphorylation analysis of pEGFR, pErbB2, pErbB3, pAkt, and pErk. EGFR sequences encoding kinase domain and EGFR gene copy numbers were determined by cDNA sequencing and real-time PCR, respectively. Finally, responsiveness to gefitinib was compared with responsiveness to the anti-EGFR antibody cetuximab (Erbitux). Results: Expression levels of pErbB2 (P = 0.02) and total ErbB3 protein (P = 0.02) associated with resistance to gefitinib. Combining gefitinib with pertuzumab (Omnitarg), an antibody targeting ErbB2 heterodimerization, provided additional growth-inhibitory effect over gefitinib alone on relatively gefitinib-resistant HNSCC cell lines. The same markers did not predict resistance to cetuximab. In contrast, a similar trend suggesting association between EGFR gene copy number and drug sensitivity was observed for both gefitinib (P = 0.0498) and cetuximab (P = 0.053). No activating EGFR mutations were identified. Conclusions:EGFR amplification may predict sensitivity to gefitinib in HNSCC. However, other EGFR/ErbB receptor family members than EGFR may contribute to resistance to gefitinib. ErbB2 and ErbB3 may have potential as predictive markers and as therapeutic targets for combination therapy in treatment of HNSCC with gefitinib.

Radiotherapy treatment planning and long-term follow-up with [11C]methionine PET in patients with low-grade astrocytoma

PURPOSE To evaluate the feasibility of [(11)C]-methionine positron emission tomography (MET PET) in radiotherapy (RT) treatment planning and long-term follow-up in patients with low-grade glioma. PATIENTS Thirteen patients with low-grade astrocytoma and 1 with anaplastic astrocytoma underwent sequential MET PET and magnetic resonance imaging (MRI) before and 3, 6, 12, and 21-39 months after RT, respectively. Ten patients were studied after initial debulking surgery or biopsy and 4 in the recurrence phase. METHODS A total of 58 PET scans were performed. After transmission scanning, a median dose of 425 MBq of MET was injected intravenously and emission data was acquired 20 min after injection for 20 min. The uptake of MET in tumor area was measured as standardized uptake value (SUV) and tumor-to-contralateral brain SUV ratios were generated to assess irradiation effects on tumor metabolism. Functional imaging with PET was compared with concurrent MRI in designing the RT planning volumes and in assessment of response to RT during a median follow-up time of 33 months. RESULTS In 12 patients (86%), tumor area was clearly discernible in the baseline PET study. In the remaining 2 patients with a suspected residual tumor in MRI, PET showed only a diffuse uptake of MET interpreted as negative in the original tumor area. In the dose planning of RT, MET PET was helpful in outlining the gross tumor volume in 3 of 11 cases (27%), whereas PET findings either coincided with MRI (46%) or were less distinctive (27%) in other cases. In quantitative evaluation, patients with a low tumor SUV initially had significantly better prognosis than those with a high SUV. Tumor-to-contralateral brain uptake ratios of MET discriminated well patients remaining clinically stable from those who have since relapsed or died of disease. CONCLUSION Quantitative MET PET has prognostic value at the time of initial treatment planning of low-grade glioma. Some patients may benefit of RT volume definition with MET PET, which seems to disclose residual tumor better than MRI in selected cases. Stable or decreasing uptake of MET in tumor area after RT during follow-up seems to be a favorable sign.

Concurrent cetuximab, cisplatin, and radiation for squamous cell carcinoma of the head and neck in vitro.

BACKGROUND AND PURPOSE For locoregionally advanced HNSCC, chemoradiotherapy with cisplatin or another platinum compound is considered as one of the standard treatment regimes. Cisplatin has improved the loco-regional control, but also increased especially the acute side effects. Cetuximab blocks ligand binding and receptor activation by binding to the extracellular domain of the EGFR. The blockade of EGFR signaling in combination with cytotoxic drugs or with radiotherapy could be a novel effective management with a relatively favourable toxicity for HNSCC. In the present study we have examined in vitro a potentially novel effective management for HNSCC, cetuximab combined with cisplatin and radiotherapy. MATERIALS AND METHODS Seven head and neck SCC cell lines were studied. Cetuximab concentrations of 0.22-8.20 nM and cisplatin concentrations of 0.038-0.220 microg/ml were used. In order to test the concurrent use of cetuximab, cisplatin and radiation, the cells were treated with the desired drug concentrations immediately after irradiation, plated into 96-well culture plates, and incubated for 4 weeks. The number of positive wells was counted. The PE was calculated and fraction survival data were fitted to the LQ model. AUC value was obtained with numerical integration. The types of interaction were analyzed. RESULTS Cetuximab and cisplatin constantly induced an additive or supra-additive effect when combined with irradiation in the seven HNSCC cell lines tested. CONCLUSIONS We evaluated concurrent cetuximab, cisplatin, and radiation for HNSCC cell lines. Preliminary efficacy results are encouraging, and further development of this targeted combined modality paradigm is warranted.

Squamous cell carcinoma is highly sensitive to taxol, a possible new radiation sensitizer.

We studied the sensitivity of seven cell lines established from laryngeal carcinoma to the cytotoxic drug paclitaxel (Taxol) in vitro. In all four cell lines tested for growth inhibition, paclitaxel reduced growth at low concentrations, and in two cell lines growth was completely inhibited at a paclitaxel concentration of only 1 x 10(-8) M. Flow cytometric data showed a G2/M block in all seven cell lines after exposure to paclitaxel for 24 h at a concentration of 1 x 10(-8) M. This concentration is about one-one hundredth fold smaller than those measured in serum after a single intravenous dose of about 200 mg/m2. The high in vitro sensitivity of laryngeal cancer cell lines to paclitaxel,and the G2/M block suggest that the drug may potentially be used in conjunction with radiotherapy.

Additive and supra-additive cytotoxicity of cisplatin-taxane combinations in ovarian carcinoma cell lines

SummaryThe purpose of this study was to compare the growth-inhibitory effect of cisplatin–paclitaxel with that obtained with a cisplatin–docetaxel combination and to assess the type of interaction. Concomitant use of taxanes and cisplatin was studied in seven human ovarian carcinoma cell lines, using the 96-well plate clonogenic assay. Chemosensitivity was expressed in terms of IC50 values, the drug concentration causing 50% inhibition of clonogenic survival. The type of interaction was studied using the area under the survival curve ratios (AUC ratios) obtained by numerical integration. Comparison of the AUC ratio and the surviving fraction (SF) value after taxane alone was made using Student’s t-test. The influence of the drug concentration was tested by one-way analysis of variance (Anova). A supra-additive or additive effect was seen when seven ovarian carcinoma cell lines were exposed to paclitaxel or docetaxel concomitantly with cisplatin. A supra-additive effect was found in four cell lines (UT-OC-3, UT-OC-4, UT-OC-5 and SK-OV-3) after simultaneous use of cisplatin with all docetaxel concentrations tested, and in two cell lines (UT-OC-4 and SK-OV-3) when cisplatin was used concomitantly with paclitaxel. A more pronounced supra-additive effect was seen with the combination of cisplatin and docetaxel. The degree of supra-additivity was dose dependent, with increasing synergy after a higher taxane dose. The data obtained in this study suggest that a supra-additive or additive effect can be achieved in ovarian carcinoma with the concomitant use of cisplatin and a taxane.

The EGFR inhibitor gefitinib suppresses recruitment of pericytes and bone marrow-derived perivascular cells into tumor vessels.

Drugs that target EGFR have established anti-tumor effect and are used in the clinic. Here we addressed whether inhibition of EGFR tyrosine kinase activity by gefitinib in tumor microenvironment affected tumor angiogenesis or vasculogenesis. A syngeneic tumor model of mice with grafted GFP-labeled bone marrow cells was used to analyze the effects of gefitinib on different cellular components of tumor vasculature. To characterize tumor cell-independent stromal effects of EGFR targeting, the mice were injected with B16 melanoma cells not expressing significant quantities of EGFR, and treated with gefitinib for seven days, a period not sufficient for significant reduction in total tumor volume. Numbers of vessels as well as cell surface areas covered by markers of endothelial, pericyte and bone marrow-derived progenitor cells were quantified by image analysis of tumor sections. Quantitative analysis of immunohistochemical data demonstrated that gefitinib decreased the coverage of small CD31-positive vessels with NG2-positive pericytes, as well as reduced the recruitment of perivascular GFP-positive bone marrow-derived progenitor cells within the tumor tissue. These results suggest that inhibition of EGFR activity in tumors has vascular effects in the absence of direct effect on tumor cells. EGFR targeting may lead to suppressed mobilization of pericytes needed for vessel stabilization, as well as of bone marrow-derived perivascular progenitor cells. These findings introduce novel cellular mechanisms by which EGFR targeted drugs may suppress tumor growth.

A method to improve target dose homogeneity of craniospinal irradiation using dynamic split field IMRT

PURPOSE Craniospinal irradiation (CSI) is technically very challenging and field edge matching is needed because of the mechanical limitations of standard linear accelerators. We assessed the feasibility of intensity-modulated radiotherapy (IMRT) in CSI to overcome the standard feathering and dose inhomogeneities associated with the standard feathering technique in the junction areas. MATERIALS AND METHODS The use of IMRT in CSI was studied with five patients CT scanned in the supine position. Isocentric treatment plans of three dimensional conventional radiotherapy (3D-CRT) and split field IMRT (sfIMRT) with dynamic intrafractional feathering were created with the same field setup and the resulted dose distributions were compared. The effect of treatment inaccuracy was simulated with an intentional shift of +/-3mm with both treatment plans. Dosimetric verification of the sfIMRT treatment plan was performed with radiographic films placed in a phantom. RESULTS The sfIMRT treatment plans resulted in a better dose coverage and uniformity in the target volume. The +/-3mm shift had only a minor effect on the dose distribution of the sfIMRT treatment plan whereas with the 3D-CRT the shift resulted in an error of +/-38% of the calculated dose in the spinal cord. The measured dose distribution of the sfIMRT treatment plan correlated well with the calculations. CONCLUSIONS Improved dose homogeneity in the target volume was achieved with the sfIMRT compared to the conventional 3D-CRT treatment plan. With the sfIMRT technique only a single treatment plan is required to deliver the total treatment dose and the resulting dose distribution is also less volatile for technical uncertainties of the treatment.

Effects of radiation fractionation on four squamous cell carcinoma lines with dissimilar inherent radiation sensitivity.

SummaryThe effect of radiation fractionation was investigated using a new 96-well-plate clonogenic assay in four squamous cell carcinoma lines. Earlier experiments had shown that two of the cell lines (UT-SCC-1A and UM-SCC-14A) were inherently relatively sensitive to irradiation, and two (UM-SCC-1 and UM-SCV-1A) relatively resistant. All of the four carcinomas from which the cell lines were established had poor clinical outcome. The radiation doses were given as a single exposure, or split into two or three equal fractions with a 24-h interval. The two inherently sensitive cell lines showed enhanced survival after radiation fractionation as compared with a single dose, whereas the resistant cell lines did not. The result suggests that both the inherent resistance of cancer cells to irradiation and the repair of sublethal radiation damage may lead to treatment failure, and that shortening of the total irradiation time may overcome cancer cell recovery between fractions in some, but not in all carcinomas.

Comparison of cellular radiosensitivity between different localizations of head and neck squamous-cell carcinoma

The prognosis of carcinomas arising from various sites in the head and neck varies even when the stage of the disease is taken into consideration, e.g. laryngeal carcinoma has a more favourable prognosis compared to oral-cavity malignancies. The purpose of this study was to evaluate intrinsic cellular radiosensitivity as one possible explanation for the observed differences in the survival rates of different anatomical groups. The radiation survival curves were determined for well characterized cell lines derived from laryngeal carcinoma (n=14), pharyngeal carcinoma (n=6), carcinoma of the oral cavity (n=14) and the skin of the face (n=3). The intrinsic radiosensitivity was expressed as area under the survival curve (AUC) values, and this cellular parameter was compared with clinical data and survival of the patients. The intrinsic radiosensitivity in the whole group varied between 1.0 Gy and 2.8 Gy with an average of 1.9 Gy. The mean AUC values for the laryngeal cell lines were 2.0 Gy±0.2, for the oral cavity 1.8±0.3 Gy, for the pharynx 1.8±0.2 Gy and for cutaneous carcinoma 2.1±0.1 Gy. There was a slight difference between the groups of glottic and supraglottic cell lines (mean 1.8±0.2 Gy and 2.1±0.3 Gy, respectively), which is consistent with the differences in clinical curability of these cancers. Otherwise, the differences in cellular radiosensitivity of the carcinoma groups studied did not reach statistical significance. These results indicate that the intrinsic radiosensitivity of squamous-cell carcinoma (SCC) of the larynx does not significantly differ from that of SCC of other sites of the head and neck. Variations in the intrinsic radiosensitivity do not as such seem to explain the observed differences in radiocurability of SCC variously localized in the head and neck.

Supra-additive effect with concurrent paclitaxel and cisplatin in vulvar squamous cell carcinoma in vitro

The effect of concurrent paclitaxel and cisplatin was tested in vitro in 5 vulvar squamous cell carcinoma (SCC) cell lines (UM‐SCV‐1A, ‐2, ‐4 and ‐7 and UT‐SCV‐3). Chemosensitivity was tested using the 96‐well plate clonogenic assay. Paclitaxel concentrations used varied between 0.4 and 1.6 nM, and cisplatin concentrations varied between 0.1 and 0.9 μg/ml. These drug concentrations are clinically achievable. Survival data were fitted to the LQ model, and the area under the curve (AUC) value was obtained with numerical integration. The type of interaction was determined by comparing the AUC ratio of the 2 drugs with the survival fraction (SF) of paclitaxel alone. With all cell lines tested the growth‐inhibitory effect of simultaneous paclitaxel and cisplatin was at least additive. The effect of the tested combination on the UM‐SCV‐1A and UT‐SCV‐3 cell lines was clearly supra‐additive with all paclitaxel concentrations tested, and the UM‐SCV‐4 and UM‐SCV‐7 cell lines exhibited a supra‐additive effect with increasing paclitaxel concentrations. The degree of supra‐additivity was dose‐dependent in the UM‐SCV‐7 cell line with increasing synergy at higher paclitaxel doses. In the current study the combination of paclitaxel and cisplatin had a clear additive or supra‐additive cytotoxic effect on the vulvar SCC cell lines, and it has been successfully used in other gynecologic malignancies; therefore concurrent paclitaxel and cisplatin also deserves further testing in clinical settings in advanced‐stage vulvar carcinoma, which has a poor prognosis.