Molecular Determinants of the Inflammatory Breast Cancer Phenotype

December 1, 2008

Since its early descriptions by Lee and Tannenbaum in 1924, inflammatory breast cancer (IBC) has been recognized as the most aggressive and lethal form of primary breast cancer. With distinct clinical, pathologic, biologic, and molecular features, IBC presents unique challenges and opportunities to breast oncologists and breast cancer researchers.

Despite advances in multimodality treatment, inflammatory breast cancer (IBC) remains the most aggressive and lethal form of breast cancer. The use of primary human IBC cell lines and functional in vivo xenograft cancer models have revealed characteristics innate to IBC thought to confer a strong metastatic potential and aggressive phenotype. Classic descriptive markers in IBC (eg, estrogen and progesterone receptor status) often guide optimal therapy and aid in development of new diagnostic and prognostic technologies. Recent IBC research has examined two genes, RhoC GTPase and WISP3, which are concordantly altered in the majority of IBC tumors but not in non-IBC specimens. RhoC serves as a transforming oncogene by regulation of genes involved in the cell cycle, secretion of angiogenic factors, and activity of insulin-like growth factor (IGF). WISP3 functions as a tumor-suppressor gene by modulation of IGF activity and resultant inhibition of cell proliferation, growth, and angiogenesis. Continued research with molecular analysis technology is imperative in order to harness differential gene expression and fully discover a signature profile of IBC. The ultimate goal is to reveal the specific molecular determinants that underlie its aggressive phenotype so that we may accurately identify markers of disease, improve diagnostic tools and predictors of response to treatment, and even suggest targeted IBC-specific therapies that afford improved survival.

Since its early descriptions by Lee and Tannenbaum in 1924, inflammatory breast cancer (IBC) has been recognized as the most aggressive and lethal form of primary breast cancer. With distinct clinical, pathologic, biologic, and molecular features, IBC presents unique challenges and opportunities to breast oncologists and breast cancer researchers.

Definition and Epidemiology

IBC is rare in the United States and Western Europe, accounting for only 2.0% of all breast malignancies with an overall incidence of 2.5 per 100,000 woman-years.[1] IBC is characterized clinically by the rapid development of an enlarged breast that is edematous, reddened, warm, and often tender with thickened and brawny overlying skin that resembles an orange skin (“peau d’orange”). A mass is present in approximately one-third to one-half of cases on clinical or radiologic examination, and mammography is nearly always abnormal, with the most common findings being skin and trabecular thickening as well as axillary lymphadenopathy.

The pathologic hallmark of IBC on biopsy is invasion of the dermal lymphatic vessels by tumor emboli that impede the flow of lymph fluid, causing the clinical presentation to mimic classic inflammation. In fact, it is this embolic invasion, and not an inflammatory infiltrate, that leads to the erythema, edema, and induration seen in this syndrome and resulted in the misnomer “inflammatory” breast cancer. Other histologic features include extensive cellular pleomorphism, high histologic grade, a high intratumoral microvessel density, and highly atypical mitotic figures.[2-3]

Biologic Features
Three biologic features make IBC a unique form of locally advanced breast cancer (LABC). First, the tumor is rapidly progressive; in fact, the rate of progression is so rapid as to be easily misdiagnosed as acute infection. This symptom is often used to differentiate primary IBC from recurrent or untreated LABC with inflammatory signs. Second, the tumor is highly angiogenic and angioinvasive. And finally, the aggressive nature and angiogenicity are intrinsic to the tumor and are present from its inception rather than as late events during progression. This was elegantly demonstrated after immunocompromised mice were transplanted with a human IBC xenograft called MARY-X. Whereas implantation of noninflammatory human tumors into the mammary fat pads of nude mice led to the development of isolated subcutaneous nodules, implantation of MARY X resulted in exclusive growth of IBC within lymphatics and blood vessels with marked overlying skin erythema.[4]

These three traits seen in IBC are believed to confer an extremely high potential for metastasis. According to staging guidelines published by the American Joint Committee on Cancer, IBC is considered a T4d tumor. Thus, all patients with IBC are classified at diagnosis as being stage IIIb, IIIc, or IV, depending on nodal status and the presence of metastases.[5] The majority of patients have clinically overt axillary lymph node involvement, and up to 36% have distant metastases at the time of diagnosis.[6]

Despite recent advances in multimodality treatment (particularly the addition of anthracycline-based combined chemotherapy), the prognosis of IBC remains relatively poorer than for non-inflammatory breast cancer (non-IBC), with a median survival of 2.9 years compared to those with non-T4 breast cancer (> 10 years) and LABC (6.4 years).[1] The still discouraging survival associated with IBC is presumably related to its rapid progression and strong metastatic potential, as the prognosis was almost uniformly fatal prior to the advent and use of multimodality systemic chemotherapy.

Racial/Geographic Disparities
An intriguing aspect of IBC epidemiology is the disparity between black and white women with regard to age at diagnosis, incidence, and survival. Hance and colleagues examined these phenomena by evaluating Surveillance, Epidemiology, and End Results (SEER) program data between 1988 and 2000 with respect to IBC.[1] Black women were noted to have a higher age-adjusted incidence rate of IBC compared to white women (3.1 vs 2.2 cases per 100,000 woman-years), and if tumors were estrogen receptor (ER)-negative, a statistically significant younger age at diagnosis (Kolmogorov-Smirnov value [KS] = 0.15, P = .01).

Perhaps the most concerning epidemiologic issue is the stark contrast in survival between the races. Median survival in black women compared to white women was worse for both IBC (2.0 vs 3.0 years) and LABC (3.1 vs 7.5 years). African-American race has been shown to be an independent predictor of elevated risk for breast cancer mortality, advanced stage at diagnosis, and tumor aggressiveness.[7-11] This last quality is evidenced by a higher histologic grade in tumor specimens and higher rates of locoregional recurrence despite equivalent response rates to local and systemic therapy. These data suggest a biologic difference between IBC tumors in blacks and whites, and perhaps with further study, these differences can be elucidated and utilized for improved prognostic and therapeutic outcomes.

More recently, Lo and colleagues uncovered important differences in the molecular and histologic characteristics of IBC between Egyptian and US patients, indicating that the North African patients tended to have many more emboli and higher expression of the RhoC oncogene.[12] These features suggest a more aggressive form of the disease, which presents in a qualitatively identical manner throughout the populations studied so far. It would be of great interest and importance to uncover the risk factors and/or environmental variables that may foster this more aggressive subtype, even within a very aggressive cancer such as IBC. It is important to note that these tumors in North Africa are not neglected stage III indolent tumors. The latter may show tumor invasion into dermal lymphatics but they seldom show the profusion of dermal lymphatic emboli that characterizes the main tumor burden in IBC (see Figures 1 and 2).

In summary, IBC presents with a unique clinical syndrome brought about by well defined histologic features that are rare in other subtypes of breast cancer. As we will summarize below, the puzzle that is IBC has begun to be assembled from the perspective of its molecular determinants.

Molecular Genetics of IBC and Animal Models

In recent years, the primary endeavor of IBC research has been the elucidation of the molecular and genetic alterations (so-called genetic determinants) that underlie its aggressive phenotype and differentiate it from non-IBC. This has been accomplished through a body of work that was spearheaded by the development of two IBC cell lines, SUM149 and SUM190, derived from primary human IBC tumors.[13] With these cell lines, functional in vivo xenograft cancer models have been created. These are implanted into the mammary fat pads of immunocompromised mice.[4,14] The SUM149 xenografts form primary tumors that often metastasize to the lungs through lymph and blood vessels.

Molecular analysis technology-including comparative genomic hybridization, gene-expression profiling, cDNA and tissue microarrays, and differential display technology-has been harnessed to compare the genetic distinctions among different forms of breast cancer in order to define a signature profile of IBC. The ultimate goal of this research is to utilize an integrated profiling approach to gene-expression differences and thereby accurately identify markers of disease, improve diagnostic tools and predictors of response to treatment, and even suggest targeted IBC-specific therapies that afford improved survival.

Classic Descriptive Markers in IBC

Initial research on the molecular biology of IBC was done by evaluating biologic markers known or thought to have prognostic significance in non-IBC such as the status of hormone receptors, epidermal growth factor receptor (EGFR), and the p53 tumor-suppressor gene. While these molecular markers may serve as general prognostic tools, they are not specific to IBC and therefore cannot be employed for diagnostic purposes. It is useful, however, to briefly review their status in IBC.

Hormone Receptor Status
It is well known that the expression of estrogen and progesterone receptors (ER and PR, respectively) correlates with clinical outcome and survival in breast cancer patients. In general, tumors that lack expression of ER and PR are more aggressive and exhibit a poorer response to hormone treatment as well as an overall decreased disease-free survival rate.[15] A study conducted by Paradiso and colleagues examined hormone receptor status in both IBC and non-IBC tumors.[16] IBC, as compared to non-IBC, was more often associated with ER negativity (56% vs 36%) and PR negativity (70% vs 49%), a finding confirmed in later studies using immunohistochemistry on tissue microarrays. Hormone receptor status is used in IBC to guide use of hormonal therapy following systemic chemotherapy and definitive locoregional therapy.

Epidermal Growth Factor Receptors

EGFR and ErbB2 (HER2) are members of the epidermal growth factor receptor gene family, and their expression is amplified in human breast cancers, resulting in uncontrolled growth and angiogenesis. Because they are generally associated with increased aggressiveness (EGFR has been shown to be an independent adverse prognostic factor in breast cancer, especially when it is coexpressed with ErbB2) and a worse clinical outcome, and because of the development of specific targeted therapies (eg, trastuzumab [Herceptin], a monoclonal antibody that inhibits HER2; lapatinib [Tykerb], a dual tyrosine kinase inhibitor that inhibits both EGFR and HER2), these tyrosine kinase receptors have been scrutinized as a possible molecular marker of IBC.

In a study by Le and colleagues, IBC tumors were found to be associated with a positive expression of EGFR (58%) and ErbB2 (60%).[17] Moreover, EGFR transcripts were found more frequently in IBC samples than in non-IBC.[18,19] However, in spite of a trend toward EGFR and ERBB2 overexpression in IBC vs non-IBC, these markers do not robustly distinguish between the aforementioned categories.

p53 Tumor Suppressor Gene
The most common mutations found in human breast cancers and human cancers in general are mutations of the p53 tumor-suppressor gene. This gene produces a protein that modulates active cell replication by induction of apoptosis or cell-cycle arrest, and this serves to preserve the genome and avoid potentiation of harmful errors in the DNA replication process. Up to 60% of IBC cases possess a mutated p53 gene, which allows accumulation of subsequent gene mutations, instability of the genome, and progression of disease.[20] These mutations are also associated with a decreased response to systemic chemotherapy.

Aziz and colleagues showed that IBC tumors were more likely to overexpress p53, compared to other tumors matched by patient age and tumor grade (69% vs 48%).[21] Moll and colleagues conducted a study using immunohistochemistry that evaluated p53 in IBC. In this study of 27 patients, p53-positive samples stained for p53 in either cytoplasm or the nucleus, and three distinct groups were identified: (1) tumors with high nuclear p53 expression, (2) tumors with cytoplasmic p53, and (3) tumors with no detectable p53. The gene sequence from the nuclear specimens contained several missense mutations in exons 5 through 8 (the “mutational hot spot” of p53 in most cancers), whereas the cytoplasmic p53 groups (positive and negative) contained wild-type p53. Thus, direct missense mutations and nuclear exclusion/cytoplasmic sequestration of wild-type p53 were thought to contribute to the altered function of p53 in human IBC.[22]

The p53 gene is also a relatively useful prognostic marker in IBC. Patients with a p53 mutation and nuclear overexpression of the protein product possessed an 8.6-fold higher risk of death, compared with patients who had neither mutation nor overexpression. When combined with ER negativity, p53 overexpression as a prognostic indicator became even more powerful. In ER-negative patients with p53 overexpression, the risk of death was increased 17.9-fold, compared to patients with p53 overexpression alone (2.8-fold).[23] In general, patients with IBC and p53 mutations have larger tumors and more disseminated disease at the time of diagnosis.[24] It is clearly important to take into account p53 status when developing new technologies for diagnosis of IBC. That said, p53 studies have been extremely small and, therefore, should be considered hypothesis-generating rather than definitive.

Specific Genetic Determinants of Inflammatory Breast Cancer

In order to better define the genes that confer to IBC the features of increased aggressiveness and metastatic potential, Merajver and colleagues evaluated the differential expression of genes in an IBC cell line (SUM149) compared to an immortalized normal human mammary epithelial (HME) cell line. From this study, 17 transcripts were found to be expressed exclusively by one line or the other.[25] In situ hybridization was utilized in a blind analysis of the isolated 17 transcripts in 20 archival IBC and 30 stage-matched non-IBC specimens to further characterize their differential expression. Two of these genes, ARHC (encoding RhoC GTPase) and WISP3 or LIBC (“lost in inflammatory breast cancer”), were concordantly altered in approximately 80% of the IBC specimens, compared to none of the stage-matched non-IBC controls. IBC tumor specimens were found to overexpress RhoC GTPase in 90% of cases, compared to 38% in non-IBC. However, WISP3 was lost in 80% of IBC specimens, compared to 21% in non-IBC.[25]

This early landmark study suggested critical underlying genetic determinants of the aggressive IBC phenotype and gave rise to further investigations into the role of each gene in breast and other cancers. Other potential determinants of the IBC phenotype that have been elucidated include NF-κB, E-cadherin, and multiple angiogenic factors. This review will focus on RhoC GTPase and WISP3.

RhoC GTPase
RhoC GTPase is a member of the Ras superfamily of small GTP-binding proteins that have been highly conserved throughout evolution. The Rho GTPases are fundamental in transduction of extracellular signaling from growth factor receptors and integrins to the cytoplasm. This signaling pathway is highly complex, with multiple sites of regulation from catalysts and inhibitors. RhoC GTPase cycles between active and inactive states and is specifically involved in reorganization of the cytoskeleton primarily through regulation of actin and focal adhesion contacts. It effectively acts as a molecular “on/off switch” that controls signal transduction and subsequent motility, invasion, and metastatic potential.[26]

Several experiments have unequivocally shown that RhoC expression correlates with aggressive subtypes of human cancers, tumor progression, and overall poor clinical outcomes. The role RhoC GTPase plays in tumor metastasis was elucidated in a study by Hakem and colleagues, who genetically engineered mice to be deficient in RhoC. While the initiation of tumors remained unaffected in these RhoC-/- knockout mice, the ability of the tumors to invade and survive metastasis was severely inhibited, as evidenced by greatly decreased frequency and size of pulmonary metastases.[27]

Conversely, key discoveries were made when van Golen and colleagues first introduced RhoC GTPase gene overexpression into HME cell lines via transfection. These cells were found to upregulate genes involved with the cell cycle (eg, cyclin D1) and cell matrix adhesion (eg, fibronectin). They secreted increased amounts of angiogenic factors such as vascular endothelial growth factor (VEGF), interleukin (IL)-6, and IL-8 than controls-a finding that was negated with pan-Rho inhibitor treatment.[28] RhoC was also found to modulate the activity of insulin-like growth factor binding protein 2 (IGFBP-2), a molecule that serves as a potent inhibitor of the mitogen IGF. With introduction of RhoC GTPase overexpression into immunocompromised mice, the formation of new, large, anchorage-independent tumor colonies was observed.[29]

These findings led to the concept that RhoC GTPase serves as a transforming oncogene in IBC that contributes to the phenotype of increased motility, invasiveness, and metastatic potential.[30]

Wnt-inducible signaling pathway protein 3 (WISP3) is a gene that produces an insulin-like growth factor binding protein-related protein called IGFBPrP-9. This protein, identified as a member of the CCN family of proteins, is rich in cysteine moieties and has conserved DNA sequences that include a domain homologous to IGFBP, among others. WISP3 plays important roles in both normal cell physiology and tumorigenesis.

The role of WISP3 in defining the aggressive phenotype of IBC was examined by introducing WISP3-containing expression vector into the SUM149 immortal HME cell line, which does not contain wild-type WISP3. The resulting transfectants formed fewer and smaller colonies, lost motility and invasion ability, and demonstrated less angiogenicity. In fact, when WISP3 was introduced into immunocompromised mice, tumor growth was suppressed and the animal survived longer.[31] Thus, it appears that WISP3 serves as a tumor-suppressor gene in IBC, and its level of expression correlates with the phenotype’s level of aggression. Wild-type WISP3 is thought to inhibit cell proliferation, growth, and angiogenesis by modulating the activity of IGF.[32]

Because of their concordant expression, the relationship between RhoC GTPase and WISP3 in the pathogenesis of IBC was investigated using an antisense approach. This study, conducted by Kleer and colleagues, generated WISP3 knockout organisms, which were found to have increased levels of RhoC expression and increased invasion and metastatic potential. However, RhoC levels decreased when WISP3 was re-introduced. These two genes appear to act in concert to contribute to the invasiveness of IBC.[33] Further research is being conducted to better elucidate the tumor suppressor role of WISP3 and the transforming oncogene role of RhoC in breast epithelium.


The above work exemplifies our collective progress toward understanding inflammatory breast cancer. There is an inherent sense of urgency for further discovery and research devoted to the underlying pathophysiology of IBC, largely due to its extremely aggressive nature and resultant poor survival rates. The goal of studying the genetic determinants of IBC is identification of markers that can lead to earlier diagnoses as well as molecular targets for novel therapies in an effort to give hope to those with this puzzling and serious form of breast cancer.

In particular, investigations into RhoC GTPase as a key IBC oncogene have revealed not only its power to confer exponentially increased tumor motility but also its value as a prognostic factor and its ability to serve as a target for new therapies. For example, a study by van Golen showed reversion of the aggressive IBC phenotype through treatment with farnesyl transferase inhibitors (FTIs).[34] It is also important to note that the advances made by exploring the role or RhoC GTPase in IBC may also be harnessed to improve outcomes in other aggressive cancer types whose phenotypes depend on RhoC such as metastatic pancreatic cancer and malignant melanoma.

Besides the discovery or development of individual targets, there is a pressing need in cancer to conduct comprehensive signaling research aimed at integrating pathways and understanding their dynamics. It is possible we are applying many drugs suboptimally due to our relatively insular approaches to individual pathways or targets. As we move closer to molecular definition of IBC, we believe that in the interim, tumor samples should be shared in the global community to validate the emerging molecular markers and genetic determinants in a setting where robust conclusions can be drawn.


1. Hance KW, Anderson WF, Devesa SS, et al: Trends in inflammatory breast carcinoma incidence and survival: The Surveillance, Epidemiology, and End Results Program at the National Cancer Institute. J Natl Cancer Inst 97:966-975, 2005.

2. Rosen PP: Rosen’s Breast Pathology, 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2001.

3. McCarthy NJ, Yang X, Linnoila IR, et al: Microvessel density, expression of estrogen receptor α, MIB-1, p53, and c-erbB-2 in inflammatory breast cancer. Clin Cancer Res 8:3857-3862, 2002.

4. Alpaugh ML, Tomlinson JS, Shao Z, et al: A novel human xenograft model of inflammatory breast cancer. Cancer Res 59:5079-5084, 1999.

5. Singletary SE, Allred C, Ashley P, et al: Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 20:3628-3636, 2002.

6. Jaiyesimi IA, Buzdar AU, Hortobagyi G: Inflammatory breast cancer: A review. J Clin Oncol 10:1014-1024, 1992.

7. Wojcik BE, Spinks MK, Optenberg SA: Breast carcinoma survival analysis for African American and white women in an equal-access health care system. Cancer 82:1310-1318, 1998.

8. Newman LA, Mason J, Cote D, et al: African-American ethnicity, socioeconomic status, and breast cancer survival: A meta-analysis of 14 studies involving over 10,000 African-American and 40,000 White American patients with carcinoma of the breast. Cancer 94:2844-2854, 2002.

9. Yood MU, Johnson CC, Blount A, et al: Race and differences in breast cancer survival in a managed care population. J Natl Cancer Inst 91:1487-1491, 1999.

10. Schwartz KL, Crossley-May H, Vigneau FD, et al: Race, socioeconomic status and stage at diagnosis for five common malignancies. Cancer Causes Control 14:761-766, 2003.

11. Henson DE, Chu KC, Levine PH: Histologic grade, stage, and survival in breast carcinoma: Comparison of African American and Caucasian women. Cancer 98:908-917, 2003.

12. Lo AC, Kleer CG, Banerjee M, et al: Molecular epidemiologic features of inflammatory breast cancer: A comparison between Egyptian and US patients. Breast Cancer Res Treat 112:141-147, 2008.

13. Ethier SP: Human breast cancer cell lines as models of growth regulation and disease progression. J Mammary Gland Biol Neoplasia 1:111-121, 1996.

14. Shirakawa K, Tsuda H, Heike Y, et al: Absence of endothelial cells, central necrosis, and fibrosis are associated with aggressive inflammatory breast cancer. Cancer Res 61:445-451, 2001.

15. Kaufman M: Review of known prognostic variables. Results Probl Cell Differ 140:77-87, 2000.

16. Paradiso A, Tommasi S, Brandi M, et al: Cell kinetics and hormonal receptor status in inflammatory breast carcinoma. Comparison with locally advanced disease. Cancer 64:1922-1927, 1989.

17. Le MG, Arriagada R, Bahi J, et al: Are risk factors for breast cancer similar in women with inflammatory breast cancer and in those with non-inflammatory breast cancer? Breast 15:355-362, 2006.

18. Charafe-Jauffret E, Tarpin C, Bardou VJ, et al: Immunophenotypic analysis of inflammatory breast cancers: Identification of an ‘inflammatory signature’. J Pathol 202:265-273, 2004.

19. Guerin M, Gabillot M, Mathieu MC, et al: Structure and expression of c-erbB-2 and EGF receptor genes in inflammatory and non-inflammatory breast cancer: Prognostic significance. Int J Cancer 43:201-208, 1989.

20. Turpin E, Bieche I, Bertheau P, et al: Increased incidence of ERBB2 overexpression and TP53 mutation in inflammatory breast cancer. Oncogene 21:7593-7597, 2002.

21. Aziz SA, Pervez S, Khan S, et al: Case control study of prognostic markers and disease outcome in inflammatory carcinoma breast: A unique clinical experience. Breast J 7:398-404, 2001.

22. Moll UM, Riou G, Levine AJ: Two distinct mechanisms alter p53 in breast cancer: Mutation and nuclear exclusion. Proc Natl Acad Sci USA 89:7262-7266, 1992.

23. Riou G, Le MG, Travagli JP, et al: Poor prognosis of p53 gene mutation and nuclear overexpression of p53 protein in inflammatory breast carcinoma. J Natl Cancer Inst 85:1765-1767, 1993.

24. Faille A, De Cremoux P, Extra JM, et al: p53 mutations and overexpression in locally advanced breast cancers. Br J Cancer 69:1145-1150, 1994.

25. van Golen KL, Davies S, Wu ZF, et al: A novel putative low affinity insulin-like growth factor-binding protein LIBC (lost in inflammatory breast cancer), and RhoC GTPase correlate with the inflammatory breast cancer phenotype. Clin Cancer Res 5:2511-2519, 1999.

26. Hall A: Rho GTPases and the actin cytoskeleton. Science 279:509-514, 1998.

27. Hakem A, Sanchez-Sweatman O, You-Ten A, et al: RhoC is dispensable for embryogenesis and tumor initiation but essential for metastasis. Genes Dev 19:1974-1979, 2005.

28. van Golen KL, Wu ZF, Oiao XT, et al: RhoC GTPase overexpression modulates induction of angiogenic factors in breast cells. Neoplasia 2:418-425, 2000.

29. Wu M, Wu ZF, Kumar-Sinha C, et al: RhoC induces differential expression of genes involved in invasion and metastasis in MCF10A breast cells. Breast Cancer Res Treat 84:3-12, 2004.

30. van Golen KL, Wu ZF, Oiao XT, et al: RhoC GTPase, a novel transforming oncogene for human mammary epithelial cells that partially recapitulates the inflammatory breast cancer phenotype. Cancer Res 60:5832-5838, 2000.

31. Kleer CG, Zhang Y, Pan Q, et al: WISP3 is a novel tumor suppressor gene of inflammatory breast cancer. Oncogene 21:3172-3180, 2002.

32. Kleer CG, Zhang Y, Pan Q, et al: WISP3 (CCN6) is a secreted tumor-suppressor protein that modulates IGF signaling in inflammatory breast cancer. Neoplasia 6:179-185, 2004.

33. Kleer CG, Zhang Y, Pan Q, et al: WISP3 and RhoC guanosine triphosphatase cooperate in the development of inflammatory breast cancer. Breast Cancer Res 6:110-115, 2004.

34. van golen KL, Bao L, DiVito MM, et al: Reversion of RhoC GTPase-induced inflammatory breast cancer phenotype by treatment with a farnesyl transferase inhibitor. Mol Cancer Ther 1:575-583, 2002.