The challenge of targeting metastasis.

Metastases that are resistant to conventional therapy are the major cause of death from cancer. In most patients, metastasis has already occurred by the time of diagnosis. Thus, the prevention of metastasis is unlikely to be of therapeutic benefit. The biological heterogeneity of metastases presents a major obstacle to treatment. However, the growth and survival of metastases depend on interactions between tumor cells and host homeostatic mechanisms. Targeting these interactions, in addition to the tumor cells, can produce synergistic therapeutic effects against existing metastases.

Cross-species hybridization of microarrays for studying tumor transcriptome of brain metastasis.

Although the importance of the cellular microenvironment (soil) during invasion and metastasis of cancer cells (seed) has been well-recognized, technical challenges have limited the ability to assess the influence of the microenvironment on cancer cells at the molecular level. Here, we show that an experimental strategy, competitive cross-species hybridization of microarray experiments, can characterize the influence of different microenvironments on cancer cells by independently extracting gene expression data of cancer and host cells when human cancer cells were xenografted into different organ sites of immunocompromised mice. Surprisingly, the analysis of gene expression data showed that the brain microenvironment induces complete reprogramming of metastasized cancer cells, resulting in a gain of neuronal cell characteristics and mimicking neurogenesis during development. We also show that epigenetic changes coincide with transcriptional reprogramming in cancer cells. These observations provide proof of principle for competitive cross-species hybridization of microarray experiments to characterize the effect of the microenvironment on tumor cell behavior.

The biology of brain metastasis.

BACKGROUND: It is estimated that at least 200 000 cases of brain metastases occur each year in the US, which is 10 times the number of patients diagnosed with primary brain tumors. Brain metastasis is associated with poor prognosis, neurological deterioration, diminished quality of life, and extremely short survival. Favorable interactions between tumor cells and cerebral microvascular endothelial cells encourage tumor growth in the central nervous system, while tumor cell interactions with astrocytes protect brain metastases from the cytotoxic effects of chemotherapy. CONTENT: We review the pathogenesis of brain metastasis and emphasize the contributions of microvascular endothelial cells and astrocytes to disease progression and therapeutic resistance. Animal models used to study brain metastasis are also discussed. SUMMARY: Brain metastasis has many unmet clinical needs. There are few clinically relevant tumor models and no targeted therapies specific for brain metastases, and the mean survival for untreated patients is 5 weeks. Improved clinical outcomes are dependent on an enhanced understanding of the metastasis-initiating population of cells and the identification of microenvironmental factors that encourage disease progression in the central nervous system.

The pathogenesis of cancer metastasis: the 'seed and soil' hypothesis revisited.

Researchers have been studying metastasis for more than 100 years, and only recently have we gained insight into the mechanisms by which metastatic cells arise from primary tumours and the reasons that certain tumour types tend to metastasize to specific organs. Stephen Paget's 1889 proposal that metastasis depends on cross-talk between selected cancer cells (the 'seeds') and specific organ microenvironments (the 'soil') still holds forth today. It is now known that the potential of a tumour cell to metastasize depends on its interactions with the homeostatic factors that promote tumour-cell growth, survival, angiogenesis, invasion and metastasis. How has this field developed over the past century, and what major breakthroughs are most likely to lead to effective therapeutic approaches?

The role of the organ microenvironment in brain metastasis.

More than 40% of patients with lung cancer and breast cancer develop brain metastasis. With improved local control and therapy of metastasis to visceral organs, the morbidity and mortality due to late diagnosed brain metastasis are projected to rise. The median survival for untreated patients is 1-2 months, which may be extended to 6 months with surgery, radiotherapy, and chemotherapy. The development of a relevant mouse model for the establishment and growth of brain metastasis has advanced our understanding of the biology and therapy of this most feared consequence of cancer. Injection of murine or human tumor cells into the internal carotid artery of mice produces experimental metastases in specific regions of the brain that are not due to patterns of initial cell arrest, motility, or invasiveness, but rather to the ability of metastatic tumor cells to exploit homeostatic mechanisms and proliferate. Immunohistochemical and morphometric analyses demonstrate that the density of blood vessels within experimental metastases in brains of mice or in clinical specimen of human lung cancer brain metastases is lower than that in the adjacent tumor-free brain parenchyma. However, brain metastasis-associated blood vessels are dilated and contain numerous dividing endothelial cells. Immunohistochemical analysis also reveals that tumor cells located less than 100 µm from a blood vessel are viable, whereas more distant tumor cells undergo apoptosis. Tumor cells within brain metastasis produce VEGF which induces permeability in adjacent vessels. The BBB in metastases that are larger than 0.25 mm in diameter is leaky. Metastases in the brain are resistant to chemotherapeutic drugs. The venerable "seed and soil" hypothesis suggests that the outcome of metastasis depends on the interaction between unique tumor cells and the specific organ microenvironment. The demonstration that activated astrocytes whose physiological role is to protect neurons from toxic substances can be exploited by tumor cells for protection from chemotherapeutic drugs suggests new approaches to the treatment of this fatal disease.

Selection of brain metastasis-initiating breast cancer cells determined by growth on hard agar.

An approach that facilitates rapid isolation and characterization of tumor cells with enhanced metastatic potential is highly desirable. Here, we demonstrate that plating GI-101A human breast cancer cells on hard (0.9%) agar selects for the subpopulation of metastasis-initiating cells. The agar-selected cells, designated GI-AGR, were homogeneous for CD44(+) and CD133(+) and five times more invasive than the parental GI-101A cells. Moreover, mice injected with GI-AGR cells had significantly more experimental brain metastases and shorter overall survival than did mice injected with GI-101A cells. Comparative gene expression analysis revealed that GI-AGR cells were markedly distinct from the parental cells but shared an overlapping pattern of gene expression with the GI-101A subline GI-BRN, which was generated by repeated in vivo recycling of GI-101A cells in an experimental brain metastasis model. Data mining on 216 genes shared between GI-AGR and GI-BRN breast cancer cells suggested that the molecular phenotype of these cells is consistent with that of cancer stem cells and the aggressive basal subtype of breast cancer. Collectively, these results demonstrate that analysis of cell growth in a hard agar assay is a powerful tool for selecting metastasis-initiating cells in a heterogeneous population of breast cancer cells, and that such selected cells have properties similar to those of tumor cells that are selected based on their potential to form metastases in mice.

The seed and soil hypothesis revisited--the role of tumor-stroma interactions in metastasis to different organs.

The fact that certain tumors exhibit a predilection for metastasis to specific organs has been recognized for well over a century now. An extensive body of clinical data and experimental research has confirmed Stephen Paget's original "seed and soil" hypothesis that proposed the organ-preference patterns of tumor metastasis are the product of favorable interactions between metastatic tumor cells (the "seed") and their organ microenvironment (the "soil"). Indeed, many of the first-line therapeutic regimens, currently in use for the treatment of human cancer are designed to target cancer cells (such as chemotherapy) and also to modulate the tumor microenvironment (such as antiangiogenic therapy). While some types of tumors are capable of forming metastases in virtually every organ in the body, the most frequent target organs of metastasis are bone, brain, liver and the lung. In this review, we discuss how tumor-stromal interactions influence metastasis in each of these organs.

Tumor cell-organ microenvironment interactions in the pathogenesis of cancer metastasis.

The process of cancer metastasis is sequential and selective and contains stochastic elements. The growth of metastases represents the endpoint of many lethal events that few tumor cells can survive. Primary tumors consist of multiple subpopulations of cells with heterogeneous metastatic properties, and the outcome of metastasis depends on the interplay of tumor cells with various host factors. The findings that different metastases can originate from different progenitor cells account for the biological diversity that exists among various metastases. Even within a solitary metastasis of proven clonal origin, however, heterogeneity of biological characteristics can develop rapidly. The pathogenesis of metastasis depends on multiple interactions of metastatic cells with favorable host homeostatic mechanisms. Interruption of one or more of these interactions can lead to the inhibition or eradication of cancer metastasis. For many years, all of our efforts to treat cancer have concentrated on the inhibition or destruction of tumor cells. Strategies both to treat tumor cells (such as chemotherapy and immunotherapy) and to modulate the host microenvironment (including the tumor vasculature) should offer additional approaches for cancer treatment. The recent advances in our understanding of the biological basis of cancer metastasis present unprecedented possibilities for translating basic research to the clinical reality of cancer treatment.

Formation of solid tumors by a single multinucleated cancer cell.

BACKGROUND: Large multinucleated cells (MNCs) commonly exist in tumorigenic cancer cell lines that are used widely in research. However, the contributions of MNCs to tumorigenesis are unknown. METHODS: In this study, MNCs were characterized in the murine fibrosarcoma cell line UV-2237 in vitro and in vivo at the single-cell level. RESULTS: The authors observed that MNCs originated from a rare subpopulation of mononuclear cells and were positive for a senescent marker, ß-galactosidase. In addition, MNCs were responsible for the majority of clonogenic activity when cultured in hard agar; they were more resistant to chemotherapeutic agents than mononuclear cells; they could undergo asymmetric division (producing mononuclear cells) and self-renewal in vitro and in vivo; and, most important; a single MNC produced orthotopic, subcutaneous tumors (composed mainly of mononuclear cells) that gave rise to spontaneous lung metastases in nude mice. CONCLUSIONS: The current results indicated that the growth of MNCs may be arrested under stress and that MNCs are highly resistant to chemotherapy and can generate clonal, orthotopic, metastatic tumors.

Circulating monocytes expressing CD31: implications for acute and chronic angiogenesis.

To identify the roles of various circulating cells (eg, endothelial and/or stem and progenitor cells) in angiogenesis, we parabiosed a wild-type syngeneic mouse with a transgenic syngeneic green fluorescent protein mouse. Following the establishment of a common circulation between these parabionts, we investigated acute (7 to 10 days), subacute (2 to 3 weeks), and chronic (4 to 6 weeks) phases of angiogenesis in wild-type mice using wound healing, implanted gel foam fragments, and subcutaneous tumor assays, respectively. We found that under in vitro conditions, circulating murine monocytes expressed F4/80, CD31, and vascular endothelial growth factor receptor 2, but neither CD133 nor von Willebrand factor, whereas murine endothelial cells expressed CD31, vascular endothelial growth factor receptor 2, and von Willebrand factor, but neither CD133 nor F4/80. Immunofluorescence analysis revealed that green fluorescent protein-positive cells in the walls of new vessels in wounds, gel foam blocks, and tumors expressed both F4/80 and CD31, that is, macrophages. Pericytes, cells that express both CD31 and desmin, were found both in the walls of tumor-associated vessels and within tumors. Collectively, these data demonstrate that monocytes (ie, cells that express both CD31 and F4/80) may be recruited to the site of tissue injury and directly contribute to angiogenesis, reaffirming the close relationships between various cell types within the reticuloendothelial system and suggesting possible targets for anticancer treatments.

Modification of the primary tumor microenvironment by transforming growth factor alpha-epidermal growth factor receptor signaling promotes metastasis in an orthotopic colon cancer model.

The transforming growth factor alpha (TGFalpha)/epidermal growth factor receptor (EGFR) signaling pathway appears to play a critical role in colon cancer progression, but the cellular and molecular mechanisms that contribute to metastasis remain unknown. KM12C colon cancer cell clones expressing high (C9) or negligible (C10) levels of TGFalpha were implanted into the cecal walls of nude mice. C9 tumors formed autocrine and paracrine EGFR networks, whereas C10 tumors were unable to signal through EGFR. The tumor microenvironment of C9, but not C10, contained cells enriched in vascular endothelial growth factor (VEGF) A, interleukin-8, and matrix metalloproteinases-2 and -9 and had a high vascular surface area. C9 tumors recruited a macrophage population that co-expressed F4/80 and lymphatic vessel endothelial hyaluronic acid receptor and produced VEGFC. The mean lymphatic density of C9 tumors was threefold higher than that of C10 tumors. C9, but not C10, tumor cells metastasized to regional lymph nodes in all mice and to the liver in 5 of 10 mice. Forced expression of TGFalpha in C10 tumor cells led to the generation of autocrine and paracrine EGFR signaling, macrophage recruitment, enhanced blood and lymphatic vascular surface areas, and increased lymphatic metastasis. Collectively, these data show that activation of TGFalpha-EGFR signaling in colon cancer cells creates a microenvironment that is conducive for metastasis, providing a rationale for efforts to inhibit EGFR signaling in TGFalpha-positive colon cancers.

Neoadjuvant platelet derived growth factor receptor inhibitor therapy combined with docetaxel and androgen ablation for high risk localized prostate cancer.

PURPOSE: Platelet derived growth factor receptor inhibitor therapy improves the efficacy of taxane chemotherapy in preclinical models of prostate cancer. Men with high risk localized prostate cancer were treated with platelet derived growth factor receptor inhibitor therapy, docetaxel and hormone ablation in the preoperative setting, and clinicopathological outcomes were evaluated. MATERIALS AND METHODS: A total of 36 men with cT2 or greater disease, Gleason grade 8-10, serum prostate specific antigen more than 20 ng/ml or cT2b and prostate specific antigen more than 10 ng/ml and Gleason 7 disease, without radiological evidence of metastases, were scheduled to receive intramuscular leuprolide, 600 mg daily oral imatinib and 30 mg/m(2) weekly docetaxel x 4 every 42 days for 3 cycles before radical prostatectomy (beta [0.02, 1.98] prior on the possibility of pathological complete remission). Unresectable disease, postoperative prostate specific antigen 0.2 ng/ml or greater, or administration of postoperative radiation or hormones were defined as treatment failure. RESULTS: A total of 39 men were registered over 15 months. Median patient age was 57 years (range 44 to 71). Risk factors included T3 disease (22 of 39), Gleason 8-10 disease (31 of 39) and prostate specific antigen more than 20 ng/ml (12 of 39). Three men were ineligible or declined therapy, 29 of 36 (81%) received 3 cycles of therapy and 7 of 36 (19%) discontinued therapy related to toxicity. Grades 3-4 toxicity included rash (4), diarrhea (4), fatigue (6) and neutropenia (1). The surgical approach was feasible, without excessive or unusual complications such as wound dehiscence. No pathological complete remissions were defined. At a median followup of 39 months 53% were free from progression. CONCLUSIONS: Evidence for a favorable impact of platelet derived growth factor receptor inhibitor therapy on the efficacy of neoadjuvant docetaxel and hormonal ablation in high risk localized prostate cancer was not obtained.

Gene expression profile of metastatic human pancreatic cancer cells depends on the organ microenvironment.

To determine the influence of the microenvironment on changes in gene expression, we did microarray analysis on three variant lines of a human pancreatic cancer (FG, L3.3, and L3.6pl) with different metastatic potentials. The variant lines were grown in tissue culture in the subcutis (ectopic) or pancreas (orthotopic) of nude mice. Compared with tissue culture, the number of genes of which the expression was affected by the microenvironment was up-regulated in tumors growing in the subcutis and pancreas. In addition, highly metastatic L3.6pl cells growing in the pancreas expressed significantly higher levels of 226 genes than did the L3.3 or FG variant cells. Growth of the variant lines in the subcutis did not yield similar results, indicating that the orthotopic microenvironment significantly influences gene expression in pancreatic cancer cells. These data suggest that investigations of the functional consequence of gene expression require accounting for experimental growth conditions.

Zonal heterogeneity for gene expression in human pancreatic carcinoma.

Using Affymetrix HG-U133 Plus 2.0 array and laser capture microdissection techniques, we determined whether different zones of the same pancreatic tumor exhibited differential expression of genes. Human L3.6pl pancreatic cancer cells were implanted into the pancreas of nude mice. Three weeks later when tumors were 7 to 9 mm in diameter, gene expression patterns in tumor cells within the central and peripheral zones were compared, and 1,222 genes showed statistically significant differences. Bioinformatic functional analysis revealed that 346 up-regulated genes in the peripheral zone were related to cytoskeleton organization and biogenesis, cell cycle, cell adhesion, cell motility, DNA replication, localization, integrin-mediated signaling pathway, development, morphogenesis, and IkappaB kinase/nuclear factor-kappaB cascade; 876 up-regulated genes in the central zone were related to regulation of cell proliferation, regulation of transcription, transmembrane receptor protein tyrosine kinase signaling pathways, response to stress, small GTPase-mediated signal transduction, hexose metabolism, cell death, response to external stimulus, carbohydrate metabolism, and response to wounding. The reliability of the microarray results were confirmed by in situ hybridization analysis of the expression of two genes. Collectively, the data showed zonal heterogeneity for gene expression profiles in tumors and suggest that characterization of zonal gene expression profiles is essential if microarray analyses of genetic profiles are to produce reproducible data, predict disease prognosis, and allow design of specific therapeutics.

Impact of sentinel lymphadenectomy on survival in a murine model of melanoma.

Lymphatic mapping and sentinel lymph node biopsy-also termed sentinel lymphadenectomy (SL)-has become a standard of care for patients with primary invasive cutaneous melanoma. This technique has been shown to provide accurate information about the disease status of the regional lymph node basins at risk for metastasis, provide prognostic information, and provide durable regional lymph node control. The potential survival benefit afforded to patients undergoing SL is controversial. Central to this controversy is whether metastasis to regional lymph nodes occurs independent of or prior to widespread hematogenous dissemination. A related area of uncertainty is whether tumor cells residing within regional lymph nodes have increased metastatic potential. We have used a murine model of primary invasive cutaneous melanoma based on injection of B16-BL6 melanoma cells into the pinna to address two questions: (1) does SL plus wide excision of the primary tumor result in a survival advantage over wide excision alone; and (2) do melanoma cells growing within lymph nodes produce a higher incidence of hematogenous metastases than do cells growing at the primary tumor site? We found that SL significantly improved the survival of mice with small primary tumors. We found no difference in the incidence of lung metastases produced by B16-BL6 melanoma cells growing exclusively within regional lymph nodes and cells growing within the pinna.

Targeting the EGFR, VEGFR, and PDGFR on colon cancer cells and stromal cells is required for therapy.

Immunohistochemical analysis of human colon cancers growing in the cecal walls of nude mice revealed that epidermal growth factor receptor (EGFR) and vascular endothelial growth factor receptor 2 (VEGFR2) were expressed by different tumor cells and tumor-associated endothelial cells, whereas platelet-derived growth factor receptor (PDGFR)beta was expressed by tumor-associated endothelial cells and pericytes. We hypothesized that treatment of nude mice with AEE788 (an inhibitor of EGFR and VEGFR phosphorylation) and STI571 (an inhibitor of PDGFRbeta phosphorylation) combined with irinotecan would overcome the intratumoral heterogeneity of these growth factors and efficiently inhibit colon cancer growth and metastasis. We implanted HT29 and KM12SM cells into the cecal walls of nude mice. Two weeks later, the mice were treated with oral vehicle solution; oral AEE788, oral STI571, or intraperitoneal injection of irinotecan as single agents; or the various combinations of these agents. We then assessed the mice for tumor growth and metastasis. Immunohistochemical analyses revealed that oral AEE788 suppressed proliferation and increased apoptosis of tumor cells and tumor-associated endothelial cells. Oral STI571 increased apoptosis of tumor-associated endothelial cells and pericytes. The combination of AEE788, STI571, and irinotecan produced the greatest inhibition of primary tumor growth and metastasis. Collectively, these data demonstrate that only targeting multiple tyrosine kinase receptors on colon cancer cells and tumor-associated stromal cells can overcome the effects of biologic heterogeneity for resistance to treatment and has the potential to improve therapeutic outcome for patients with this disease.

Molecular mechanism of hTid-1, the human homolog of Drosophila tumor suppressor l(2)Tid, in the regulation of NF-kappaB activity and suppression of tumor growth.

hTid-1, a human homolog of the Drosophila tumor suppressor l(2)Tid and a novel DnaJ protein, regulates the activity of nuclear factor kappaB (NF-kappaB), but its mechanism is not established. We report here that hTid-1 strongly associated with the cytoplasmic protein complex of NF-kappaB-IkappaB through direct interaction with IkappaBalpha/beta and the IKKalpha/beta subunits of the IkappaB kinase complex. These interactions resulted in suppression of the IKK activity in a J-domain-dependent fashion and led to the cytoplasmic retention and enhanced stability of IkappaB. Overexpression of hTid-1 by using recombinant baculovirus or adenovirus led to inhibition of cell proliferation and induction of apoptosis of human osteosarcoma cells regardless of the p53 expression status. Adherent cultured cells transduced with Ad.hTid-1 detached from the dish surface. Morphological changes consistent with apoptosis and cell death were evident 48 h after Ad.EGFP-hTid-1 transduction. In contrast, cells transduced with Ad.EGFP or Ad.EGFP-hTd-1DeltaN100, a mutant that has the N-terminal J domain deletion and that lost suppressive activity on IKK, continued to proliferate. Similar data were obtained with A375 human melanoma cells. Ad.EGFP or Ad.EGFP-hTd-1DeltaN100 ex vivo-transduced A375 cells injected subcutaneously into nude mice produced growing tumors, whereas Ad.EGFP-hTid-1-transduced cells did not. Collectively, the data suggest that hTid-1 represses the activity of NF-kappaB through physical and functional interactions with the IKK complex and IkappaB and, in doing so, it modulates cell growth and death.

Platelet-derived growth factor receptor inhibition and chemotherapy for castration-resistant prostate cancer with bone metastases.

PURPOSE: To further assess preclinical and early clinical evidence that imatinib mesylate, a platelet-derived growth factor receptor (PDGFR) inhibitor, modulates taxane activity in prostate cancer and bone metastases, a randomized study was conducted. EXPERIMENTAL DESIGN: Men with progressive castration-resistant prostate cancer with bone metastases (n = 144) were planned for equal randomization to i.v. 30 mg/m(2) docetaxel on days 1, 8, 15, and 22 every 42 days with 600 mg imatinib daily or placebo, for an improvement in median progression-free survival from 4.5 to 7.5 months (two-sided alpha = 0.05 and beta = 0.20). Secondary end points included differential toxicity and bone turnover markers, tumor phosphorylated PDGFR (p-PDGFR) expression, and modulation of p-PDGFR in peripheral blood leukocytes. RESULTS: Accrual was halted early because of adverse gastrointestinal events. Among 116 evaluable men (57 docetaxel + imatinib; 59 docetaxel + placebo), respective median times to progression were 4.2 months (95% confidence interval, 3.1-7.5) and 4.2 months (95% confidence interval, 3.0-6.8; P = 0.58, log-rank test). Excess grade 3 toxicities (n = 23) in the docetaxel + imatinib group were principally fatigue and gastrointestinal. Tumor p-PDGFR expression was observed in 12 of 14 (86%) evaluable bone specimens. In peripheral blood leukocytes, p-PDGFR reduction was more likely in docetaxel + imatinib-treated patients compared with docetaxel + placebo (P < 0.0001), as were reductions in urine N-telopeptides (P = 0.004) but not serum bone-specific alkaline phosphatase (P = 0.099). CONCLUSIONS: These clinical and translational results question the value of PDGFR inhibition with taxane chemotherapy in prostate cancer bone metastases and are at variance with the preclinical studies. This discordance requires explanation.

Expression of epidermal growth factor (EGF)/transforming growth factor-alpha by human lung cancer cells determines their response to EGF receptor tyrosine kinase inhibition in the lungs of mice.

Epidermal growth factor receptor (EGFR) has been extensively targeted in the treatment of non-small cell lung cancer, producing responses in a small number of patients. To study the role of ligand expression in mediating response to EGFR antagonism, we injected NCI-H441 [EGFR and EGF/transforming growth factor-alpha (TGF-alpha) positive] or PC14-PE6 (EGFR positive and EGF/TGF-alpha negative) human lung adenocarcinoma cells into the lungs of nude mice. We randomized the mice to receive treatment with the EGFR tyrosine kinase inhibitors gefitinib or AEE788 or vehicle. Treatment of mice bearing NCI-H441 but not PC14-PE6 lung tumors resulted in a significant reduction in primary tumor growth, pleural effusion, and lymph node metastasis. Immunohistochemical analyses revealed that NCI-H441 and PC14-PE6 cells expressed EGFR but that the expression of EGF/TGF-alpha was high in NCI-H441 cells and very low in PC14-PE6 cells. Consequently, EGFR was activated in both tumor and tumor-associated endothelial cells in the NCI-H441 tumors but not in the PC14-PE6 tumors. Antagonism of EGFR signaling by treatment of mice with AEE788 decreased proliferation and increased apoptosis of both tumor cells and tumor-associated endothelial cells in NCI-H441 tumors but not in PC14-PE6 tumors. However, after transfection of PC14-PE6 cells with TGF-alpha, lung tumors derived from the transfected cells expressed and activated EGFR in both tumor and tumor-associated endothelial cells and tumors responded to treatment with AEE788. Collectively, these results strongly suggest that the response of human lung cancers growing orthotopically in mice to the inhibition of EGFR signaling is determined by ligand (EGF/TGF-alpha) expression by tumor cells. Our findings provide an additional explanation for the susceptibility of lung cancers to treatment with EGFR tyrosine kinase inhibitors.