Immunologic and Biologic Properties of the Primary Tumor During Prolonged Neoadjuvant Chemoimmunotherapy

January 1, 2002

Patients with locally advanced cancers have a poor prognosis when treated with radiotherapy and/or surgery alone. The appearance of distant metastases shortly after removal of the primary tumor indicates that micrometastases are already present at the time of diagnosis. We observed a favorable outcome in patients with locally advanced breast cancer treated with a prolonged regimen of neoadjuvant chemotherapy plus granulocyte-macrophage colony-stimulating factor (GM-CSF [Leukine]) compared with patients receiving fewer chemotherapy cycles prior to surgery and radiotherapy. These results can partly be explained by the dose-intensive regimen used, but biologic and immunologic processes inherent to the prolonged presence of the primary tumor and its draining lymph nodes might also contribute to the beneficial outcome. The effects of the prolonged presence of the primary tumor during chemotherapy and GM-CSF administration on the antitumor immune response, and more specifically the functional properties of dendritic cells and T cells, are currently being investigated in a multicenter randomized clinical trial comparing prolonged neoadjuvant chemotherapy plus cytokines with a conventional treatment schedule. Aside from investigations concerning the immune system, other biologic processes, such as tumor angiogenesis, are being investigated at the same time. [ONCOLOGY 16(Suppl 1):32-39, 2002]

ABSTRACT: Patients with locally advanced cancers have a poor prognosis when treated with radiotherapy and/or surgery alone. The appearance of distant metastases shortly after removal of the primary tumor indicates that micrometastases are already present at the time of diagnosis. We observed a favorable outcome in patients with locally advanced breast cancer treated with a prolonged regimen of neoadjuvant chemotherapy plus granulocyte-macrophage colony-stimulating factor (GM-CSF [Leukine]) compared with patients receiving fewer chemotherapy cycles prior to surgery and radiotherapy. These results can partly be explained by the dose-intensive regimen used, but biologic and immunologic processes inherent to the prolonged presence of the primary tumor and its draining lymph nodes might also contribute to the beneficial outcome. The effects of the prolonged presence of the primary tumor during chemotherapy and GM-CSF administration on the antitumor immune response, and more specifically the functional properties of dendritic cells and T cells, are currently being investigated in a multicenter randomized clinical trial comparing prolonged neoadjuvant chemotherapy plus cytokines with a conventional treatment schedule. Aside from investigations concerning the immune system, other biologic processes, such as tumor angiogenesis, are being investigated at the same time. [ONCOLOGY 16(Suppl 1):32-39, 2002]

Treatment of patients with locally advanced cancerstill poses a major problem for clinicians. Patients with locally advancedtumors have shown a poor prognosis when treated by surgery and/or radiotherapyalone. Using these two treatment modalities local control can be reached in anumber of patients, but distant metastases usually appear shortly after removalof the primary tumor. This observation was made after local treatment of severaltumor types, including breast cancer, esophageal cancer, stomach cancer, bladdercancer, and head and neck cancer, thus implying that distant micrometastases arealready present at the time of diagnosis of a locally advanced tumor in mostpatients. The addition of systemic adjuvant and later also neoadjuvantchemotherapy showed an improvement of prognosis for these patients. Adjuvantchemotherapy improves survival in patients with colon[1] and breast cancer.[2]Adjuvant active specific immunotherapy with an autologous tumor cell-bacilleCalmette-Guérin (BCG) vaccine improved disease-free survival in patients withstage II colon cancer.[3]

Neoadjuvant or primary chemotherapy improves organ preservation in patientswith head and neck carcinomas without compromising survival.[4] This approach,often combined with radiotherapy, has also been tested in bladder,[5,6]esophageal,[7,8] and gastric cancer.[9] Although down-staging is often possiblewith this approach, no survival benefit has been demonstrated so far. We triedto make use of the advantages of neoadjuvant chemotherapy by combining it withimmunotherapy in the treatment of locally advanced cancer. Because the tumorantigens that are capable of inducing a specific immune response are unknown formany tumor types, we chose to test an autovaccination approach in the form ofprolonged neoadjuvant chemotherapy combined with immunomodulating cytokines. Inorder to study immunologic and biologic events during treatment of locallyadvanced tumors by a new multidisciplinary approach, we selected locallyadvanced breast cancer as a model. Patients with locally advanced breast cancerhave a primary tumor and regional lymph nodes, which are relatively easilyaccessible for sequential biopsies in order to study tumor biology.

Locally Advanced Breast Cancer

Locally advanced breast cancer includes patients with stage IIB (> 5 cm indiameter), IIIA, and IIIB tumors, according to the American Joint Committee onCancer (AJCC) guidelines.[10] This results in a heterogeneous group of breastneoplasms with different biologic and clinical characteristics.[11] Patientspresent with large primary tumors and no signs of distant metastases at the timeof diagnosis. Increasing tumor size may result in increased heterogeneity andresistance to treatment.[12] When treated with local therapy alone, patientswith locally advanced breast cancer have a poor prognosis; the 5-year overallsurvival rate is 20%.[13-15] For patients with inflammatory breast cancer the5-year overall survival rate is even worse—only 5%.[16-18] When treated withlocal therapy alone, systemic metastases tend to develop rapidly, indicating theexistence of micrometastases in most patients at diagnosis.[10] By means ofimmunohistochemistry or reverse polymerase chain reaction assay, cancer cellscan be detected in blood and bone marrow in up to 30% of patients withearly-stage breast cancer at time of surgery.[19,20]

Systemic treatment with chemotherapy alone fails to achieve local control,due to either primary drug resistance[21] or induced drug resistance of thetumor cells.[22] Combined-modality treatment with neoadjuvant chemotherapy hassignificantly improved the survival rate of patients with locally advancedbreast cancer over the past 2 decades. The aim of this approach is to renderpatients with large tumors operable and to eradicate micrometastases at an earlystage, when they are still sensitive to chemotherapy.[10,23,24] Presenttreatment schedules usually consist of three to four conventional-doseneoadjuvant chemotherapy cycles prior to local therapy by means of surgeryand/or radiotherapy, often followed by a number of adjuvant postoperativechemotherapy cycles.[10,23] Clinical response rates range between 30% and 80%,with 10% to 30% clinical complete remissions. The 4- or 5-year disease-freesurvival rates vary from 33% to 40%, and the 5-year overall survival rate variesfrom 40% to 60% with long-term survival in 15% of patients.[25-28]

Doxorubicin and cyclophosphamide (Cytoxan, Neosar) are among the mosteffective chemotherapeutic agents in breast cancer, with reported response ratesof 25% to 50% when used as single agents and 45% to 60% when in standardcombinations. For both agents, investigators have observed immunomodulatingproperties enhancing antitumor immunity.[29] Because of increasing evidence of asteep dose-response effect in breast cancer for a number of chemotherapeuticagents, including doxorubicin and cyclophosphamide,[30-34] we developed adose-intensive regimen for doxorubicin plus cyclophosphamide.[32,35]

Earlier, we reported the results of a study of 42 patients with locallyadvanced breast cancer treated with four to six cycles of neoadjuvantchemotherapy with moderately high-dose cyclophosphamide, doxorubicin, andgranulocyte-macrophage colony-stimulating factor (GM-CSF [Leukine]) prior tosurgery and postoperative radiotherapy.[23] GM-CSF was chosen as a hematopoeticgrowth factor instead of the less toxic granulocyte colony-stimulating factor(G-CSF [Neupogen]) because of the former’s reported immunostimulatoryeffects.[23,35,36] This regimen showed a high clinical response rate of 98%,with 50% clinical complete remissions. Remarkably, analysis of the data showed atrend toward improvement in disease-free and overall survival in patients whoreceived more chemotherapy cycles. The 3-year disease-free survival rates are0%, 54%, and 67%, respectively, for patients who received four, five, and sixcycles. The 3-year overall survival rates are 20%, 62%, and 79%, respectively,for patients who received four, five, and six cycles (see Figure1).[23]

We hypothesized that these promising results are not only related to the 50%increase in the chemotherapy dose, but also related to biologic and immunologicconsequences of the prolonged presence of the primary tumor and draining lymphnodes in situ.[23] Release of antiangiogenic factors was shown in a preclinicalmodel of the Lewis lung carcinoma. Such factors were shown to inhibit the growthof distant metastases.[24] Circulating angiogenesis inhibitors produced by theprimary tumor in locally advanced breast cancer patients, for instance, mightinhibit the growth of micrometastases and add to the therapeutic effects ofprolonged neoadjuvant chemotherapy. The long-term administration of GM-CSF mighthave resulted in the stimulation of antigen-presenting cells and cytotoxic Tcells, thus enhancing the antitumor immune response.[23,36]

For many years investigators have been focusing mainly on the directantitumor and toxic effects of different chemotherapeutic agents, while biologicand immunologic factors inherent to the primary tumor might also add to thetherapeutic effects. In order to study these events we have initiated theSpinoza trial, an international randomized clinical trial of locally advancedbreast cancer patients comparing six neoadjuvant cycles to three neoadjuvantcycles combined with three adjuvant cycles, and GM-CSF to G-CSF administration(see Figure 2). This study investigates immunologic and biologic parameters inbiopsy and mastectomy specimens and in the peripheral blood of all patients. Thebiologic and immunologic consequences of the prolonged presence of the primarytumor and its draining lymph nodes, prolonged neoadjuvant chemotherapy, andprolonged GM-CSF administration will be described in further detail below.

Tumor Immunology in Locally Advanced Breast Cancer

In recent years more attention has been paid to the function of the immunesystem in cancer patients, and the possibility of enhancing the antitumoreffects of the immune system. A number of tumor-associated antigensoverexpressed in breast tumors have been identified. These include HER2/neu, CEA,and MUC1. These antigens have been shown to induce T-cell responses in healthydonors and breast cancer patients.[37-40] T-cell responses are induced bypresentation of tumor antigens by professional antigen-presenting cells. Theseinclude dendritic cells, macrophages/monocytes, and B cells.[41] Dendritic cellsare widely regarded as the most professional and potent antigen-presentingcells. They play a key role in the initiation of an antitumor immuneresponse.[41-44] Dendritic cells are derived from bone marrow progenitor cells.They produce circulating precursors that migrate to the tissues, where theyremain as immature cells with high phagocytic capacity. After phagocytosis andprocessing of antigens these cells migrate to the lymphoid tissues. They presentantigens on their cell surface in association with MHC-I and MHC-II molecules.Dendritic cells are able to activate CD8-positive cytotoxic T lymphocytes andCD4-positive T cells. Dendritic cell infiltration in the primary tumor has beenassociated with a reduced incidence of metastatic disease and prolongedsurvival, for several tumor types.[45-50]

In patients with breast cancer a pattern of compartmentalization of immatureand mature dendritic cells characteristic of active immune response has beenobserved. Immature dendritic cells are found in the centers of tumors, whilemature dendritic cells are found in the peritumoral areas and lymphoid tissuessurrounded by clusters of T cells.[45] Tumor-specific T cells have been found incancer patients, but they fail to control tumor growth. This can be explained bythe existence of mechanisms for evading immune surveillance. A number of studieshave suggested the existence of immunosuppressive conditions in patients withvarious tumors.[51] Several cytokines produced by tumor cells, such asinterleukin (IL)-10, transforming growth factor (TGF)-beta, IL-6, and vascularendothelial growth factor (VEGF), have been shown to inhibit dendritic cells’maturation in vitro. IL-10 and TGF-beta have also shown inhibition of T-celleffector functions in vitro.[51-55] Tumor-induced Fas-mediated early apoptosisof dendritic cells and T cells has also been reported.[56]

In patients with locally advanced breast cancer, investigators have found adysfunction of dendritic cells and T lymphocytes. Dysfunction of T lymphocytesin breast cancer patients was found to be caused by a functional disturbance ofdendritic cells. It was shown that T cells from breast cancer patients can bestimulated by dendritic cells of healthy controls.[41,57] In more advancedstages of breast cancer, depression of T-cell reactivity becomes progressive andis significantly related to survival.[58] Dendritic cells are optimally equippedto activate T cells. Prolonged administration of GM-CSF in combination withneoadjuvant chemotherapy might help dendritic cells to overcome theseimmunosuppressive conditions present in breast cancer patients.

Consequences of Prolonged Neoadjuvant Chemotherapy

We hypothesized that the proposed schedule of six cycles of neoadjuvantchemotherapy may have an effect on the observed immunologic dysfunction.Destruction of tumor cells by chemotherapy and the resulting reduction in tumormass in patients treated with neoadjuvant chemotherapy may also reduce theproduction of several tumor-derived factors. These might include theimmunosuppressive cytokines mentioned earlier. These cytokines (eg, IL-10, TGF-beta,IL-6, and VEGF) are potential inhibitors of dendritic cell maturation and T-cellactivation.[51-54] Apoptosis and necrosis induced by chemotherapy will lead tothe release of tumor-associated antigens. It has been shown that dendritic cellscan efficiently endocytose apoptotic cell fragments and present epitopes derivedfrom these antigen-containing apoptotic bodies to cytotoxic T lymphocytes, aftertransportation to the draining lymph nodes.[59,60]

It has been shown that uptake of apoptotic bodies from normal cell turnoverby immature dendritic cells, without appropriate secondary signals, may maintainperipheral T-cell tolerance.[61] The simultaneous release of necrotic cells andan abundance of apoptotic bodies that occurs during chemotherapeutic treatmentmay provide appropriate danger signals for the maturation and activation ofdendritic cells.[62,63] Prolonged neoadjuvant chemotherapy regimens might thusprovide more time and enhance the release of appropriate signals for thedifferent components of the immune system to engage in an active antitumorimmune response.

Consequences of GM-CSF Administration

Because of the evidence for a steep dose-response relation of a number ofchemotherapeutic agents in breast cancer patients, dose-intensive regimens weredeveloped to improve survival.[30-34] Hematopoietic growth factors such as G-CSFand GM-CSF were added to these schedules to improve recovery frommyelosuppression after dose-intensive chemotherapy. As mentioned earlier, wechose GM-CSF as a hematopoietic growth factor instead of the less toxic G-CSFbecause of its reported immunomodulating effects. GM-CSF stimulates variouseffector functions of mature neutrophils, eosinophils, monocytes, andmacrophages, which might add to the antitumor effects ofchemotherapy.[32,35,36,64]

Earlier we reported on the results of a study comparing subcutaneous andintravenous administration of GM-CSF that showed that the subcutaneous routeincreased efficacy and reduced toxicity.[65] Moreover, GM-CSF also affects therecruitment and activation of macrophages and dendritic cells. It increasessurvival of these cells, and it has been shown that GM-CSF protects dendriticcells from apoptotic death.[36,64,66] By means of these mechanisms GM-CSF mightincrease the efficiency of tumor antigen presentation and the subsequentactivation of tumor-specific cytotoxic T lymphocytes.[36,64] Treatment with GM-CSFmight result in increased numbers of dendritic cells infiltrating the primarytumor site. Previous studies have shown a better prognosis for patients with anincreased number of tumor-infiltrating dendritic cells.[67] This may beexplained by protection of tumor-infiltrating lymphocytes from tumor-inducedcell death by dendritic cells. Fas-mediated apoptosis of T lymphocytes can beinduced by tumor cells expressing Fas ligand.[68-72] Dendritic cells have beenreported to protect these cells by B7/CD28 interactions and possibly by IL-12production.[73,74]

In patients with locally advanced breast cancer treated with neoadjuvantchemotherapy and GM-CSF, we observed recruitment of dendritic cells andmacrophages in the axillary lymph nodes (see Figure3). This may have thebeneficial effect of enhancement of antitumor immunity. Thus, prolongedadministration of GM-CSF might have a favorable effect on disease-free andoverall survival.

Prolonged Presence of Primary Tumor and Axillary Lymph Nodes

In the regional lymph nodes, dendritic cells present their antigens to thecirculating T cells. The lymph nodes function as a natural meeting point.Furthermore, they provide an optimal microenvironment for activation and primingof T cells.[11,59,60] As long as the primary tumor is in situ, release ofantitumor antigens, which can be processed by dendritic cells, can occur(especially when additional apoptosis and necrosis are being caused bychemotherapy).[11,61-63] The prolonged presence of the primary tumor andaxillary lymph nodes is regarded as beneficial for the activation of a T-cellimmune response. It provides optimal circumstances for dendritic cells toexpress tumor antigens and encounter, bind, and prime tumor-specific cytotoxic Tlymphocytes.[11] The generation of T-cell responses against tumor-associatedantigens may provide a highly specific and long-lasting defense against theoutgrowth of breast cancer metastases.

Antiangiogenic Effects in Locally Advanced Breast Cancer

Although at the time of diagnosis there are no signs of distant metastases,rapid growth of micrometastases after initial treatment with surgery and/orradiotherapy alone has been observed.[10] Recently, it was demonstrated thatlocal recurrences following mastectomy do not grow at a constant rate, as hadbeen assumed for a long time. They start to grow after a period of dormancy.[75]In prevascular (dormant) metastases the apoptotic rate of tumor cells is highlyelevated compared to growing tumors, while the proliferation rate is equal inboth. In several preclinical models it was observed that further development ofdistant metastases was inhibited in the presence of the primary tumor. Distantmetastases were kept dormant for a long time by tumor-derived factors that blockneovascularization.[76]

In the Lewis lung carcinoma model the circulating angiogenesis inhibitorangiostatin (an internal fragment of plasminogen) was identified as the agentresponsible for the dormancy of micrometastases. Growth of distant metastaseswas being inhibited as long as the primary tumor was in situ.[24] The sameinvestigators later identified another antiangiogenic factor (endostatin) andshowed that it was able to regress primary tumors in mice.[77] Tumor growth isdependent on the formation of new vessels (angiogenesis). A delicate balancebetween pro- and antiangiogenic factors regulates the dormancy ofmicrometastases and the moment that new vessel formation will be triggered,allowing tumor growth. Angiogenesis inhibitors keep tumors dormant until thebalance becomes in favor of proangiogenic factors. Then new vessel formation isturned on, allowing tumor growth (angiogenic switch).[78]

A number of different cell types and the factors they produce have beenidentified as involved in the production and release of angiogenic andantiangiogenic factors, and thus influence this balance. These cells includefibroblasts, macrophages, endothelial cells, and platelets.[79-81] Some of theidentified antiangiogenic factors are fragments of larger extracellular matrixand plasma proteins: fibronectin, prolactin, thrombospondin, collagen XVIII,prothrombin, and plasminogen, for instance. Many of these factors are generatedin the process of tissue remodeling.[77,82-84] Several circulating angiogenicand antiangiogenic factors have been identified in breast cancerpatients.[85,86] It was observed that the serum of some breast cancer patientswas able to stimulate endothelial cell proliferation in vitro, whereas the serumof other patients inhibited endothelial cell proliferation. This inhibitioncould be blocked by adding a specific antibody against thrombospondin,indicating its antiangiogenic effects.[85]

Platelet activation plays an important role in tumor-induced angiogenesis.Platelets are carriers of angiogenic factors, which are released upon plateletactivation.[86] Furthermore, in soft-tissue sarcomas, activated platelets thatare releasing VEGF and dense vascularization with intense VEGF expression havebeen observed.[87] In addition, VEGF-stimulated endothelial cells have beenshown to cause platelet adhesion and activation.[88] Macrophages are responsiblefor the production of both angiogenic and antiangiogenic factors. They produceangiogenic tumor growth factors (eg, TNF-alpha) and proteases. They also produceantiangiogenic protease inhibitors.[89] Macrophages are also responsible for theproduction of an elastase-like enzyme that induces angiostatin (antiangiogenicfactor) production out of plasminogen.[90] The mechanisms regulating the balancebetween angiogenic and antiangiogenic factors produced by macrophages are stillunknown.

When Lewis lung carcinoma cells were transfected with the GM-CSF gene anincreased number of tumor-infiltrating macrophages was observed and elevatedlevels of angiostatin were found. Tumor growth was inhibited at the same time.These results indicate that GM-CSF might play a role in generating angiogenesisinhibitors.[91] Earlier we reported results of a comparison between mastectomyspecimens and pretreatment biopsies of patients treated with neoadjuvantchemotherapy and GM-CSF. We observed an increase in macrophages in themastectomy specimens. These macrophages may play a role in the production ofantiangiogenic factors, as in the Lewis lung carcinoma model.[92] We hypothesizethat release of antiangiogenic factors by the primary tumor and its stroma, theintratumoral macrophages in particular, are capable of producing peptides thatmay keep micrometastases dormant.

Interaction of Biologic and Immunologic Processes

It is impossible to regard the antiangiogenic and immunomodulating effects ofour treatment schedule as two completely separate concepts, since there is someoverlap. VEGF, a proangiogenic factor, has been shown to inhibit dendritic cellmaturation in vitro and in vivo.[51,54] GM-CSF has a beneficial effect ondendritic cell maturation and might thus overcome this inhibition. The GM-CSF-inducedrecruitment and maturation of dendritic cells might lead to increased levels ofcytokines such as IL-12. IL-12 has considerable antiangiogenic effects and atthe same time plays an important role in the initiation of effectivecell-mediated antitumor immune responses.[93,94] On the other hand, someinvestigators have observed a stimulation of microvascular growth by inducingproliferation and migration of endothelial cells as an effect of GM-CSF.[95,96]Pro- and antiangiogenic factors might also affect the function of the immunesystem. Adhesion molecules important for leukocyte-endothelial interactions aredown-regulated by proangiogenic factors and up-regulated by inhibitors ofangiogenesis.[97] Thus many different processes seem to influence each other.Further investigation into the effects of these interactions is necessary.

Conclusion

Prolonged neoadjuvant chemotherapy has shown favorable results in patientswith locally advanced breast cancer. Patients treated with six cycles showedmore favorable results than patients receiving four or five cycles. Theprolonged presence of the draining lymph nodes in combination with the repeatedtumor antigen release from the primary tumor and dendritic cell recruitment andactivation, which might be further stimulated by GM-CSF administration, mayaccount for the observed increase in survival in locally advanced breast cancer.Furthermore, factors inhibiting new vessel formation produced by the primarytumor might inhibit growth of micrometastases, thus enhancing the effects ofchemotherapy. We hypothesized that these immunologic and biologic processesinherent to the primary tumor and its draining lymph nodes, and their prolongedpresence during chemotherapy, might be responsible for the observed improvement.In order to study these biologic and immunologic events in tissue samples andperipheral blood of patients with locally advanced breast cancer, we initiatedan international randomized clinical trial using locally advanced breast canceras a model for other locally advanced tumors. We also started a phase II studyin esophageal cancer with neoadjuvant gemcitabine (Gemzar), cisplatin (Platinol),and GM-CSF.

References:

1. Macdonald JS, Astrow AB: Adjuvant therapy of colon cancer. Semin Oncol28(1):30-40, 2001.

2. Cole BF, Gelber RD, Gelber S, et al: Polychemotherapy for early breastcancer: An overview of the randomised clinical trials with quality-adjustedsurvival analysis. Lancet 358:277-286, 2001.

3. Vermorken JB, Claessen AM, van Tinteren H, et al: Active specificimmunotherapy for stage II and stage III human colon cancer: A randomised trial.Lancet 353:345-350, 1999.

4. Pignon JP, Bourhis J, Domenge C, et al: Chemotherapy added to locoregionaltreatment for head and neck squamous-cell carcinoma: Three meta-analyses ofupdated individual data. MACH-NC Collaborative Group. Meta-Analysis ofChemotherapy on Head and Neck Cancer. Lancet 355:949-955, 2000.

5. Advanced Bladder Cancer Overview Collaboration: Neoadjuvant cisplatin foradvanced bladder cancer. Cochrane Database Syst Rev 2:CD001426, 2000.

6. Neoadjuvant cisplatin, methotrexate, and vinblastine chemotherapy formuscle-invasive bladder cancer: A randomised controlled trial. InternationalCollaboration of Trialists. Lancet 354:533-540, 1999.

7. Walsh TN, Noonan N, Hollywood D, et al: A comparison of multimodal therapyand surgery for esophageal adenocarcinoma. N Engl J Med 335(7):462-467, 1996.

8. Kelsen DP, Ginsberg R, Pajak TF, et al: Chemotherapy followed by surgerycompared with surgery alone for localized esophageal cancer. N Engl J Med339(27):1979-1984, 1998.

9. Kelsen D, Karpeh M, Schwartz G, et al: Neoadjuvant therapy of high-riskgastric cancer: A phase II trial of preoperative FAMTX and postoperativeintraperitoneal fluorouracil-cisplatin plus intravenous fluorouracil. J ClinOncol 14(6):1818-1828, 1996.

10. Beahrs OH, Henson DE, Hutter RVP, et al (eds): Handbook for Staging ofCancer, pp 161-167 [from the Manual for Staging of Cancer, 4th ed].Philadelphia, J.B. Lippincott, 1993.

11. Luykx-de Bakker SA, Verheul HM, de Gruijl TD, et al: Prolongedneoadjuvant treatment in locally advanced tumours: A novel concept based onbiological considerations. Ann Oncol 10:155-160, 1999.

12. Schnipper L: Clinical implications of tumor-cell heterogeneity. N Engl JMed 314(22):1423-1431, 1986.

13. Booser DJ, Hortobagyi GN: Treatment of locally advanced breast cancer.Semin Oncol 19:278-285, 1992.

14. Hortobagyi GN: Multidisciplinary management of advanced primary andmetastatic breast cancer. Cancer 74(1 suppl):416-423, 1994.

15. Valagussa P, Zambetti M, Bonadonna G, et al: Prognostic factors inlocally advanced noninflammatory breast cancer. Long-term results followingprimary chemotherapy. Breast Cancer Res Treat 15:137-147, 1990.

16. Graham MV, Perez CA, Kuske RR, et al: Locally advanced (noninflammatory)carcinoma of the breast: Results and comparison of various treatment modalities.Int J Radiat Oncol Biol Phys 21:311-318, 1991.

17. Jaiyesimi IA, Buzdar AU, Hortobagyi G: Inflammatory breast cancer: Areview. J Clin Oncol 10(6):1014-1024, 1992.

18. Valero V, Buzdar AU, Hortobagyi GN: Locally advanced breast cancer.Oncologist 1(1/2):8-17, 1996.

19. Braun S, Pantel K, Muller P, et al: Cytokeratin-positive cells in thebone marrow and survival of patients with stage I, II, or III breast cancer. NEngl J Med 342(8):525-533, 2000.

20. Smith BM, Slade MJ, English J, et al: Response of circulating tumor cellsto systemic therapy in patients with metastatic breast cancer: Comparison ofquantitative polymerase chain reaction and immunocytochemical techniques. J ClinOncol 18(7):1432-1439, 2000.

21. Lankelma J, Dekker H, Luque FR, et al: Doxorubicin gradients in humanbreast cancer. Clin Cancer Res 5(7):1703-1707, 1999.

22. Abolhoda A, Wilson AE, Ross H, et al: Rapid activation of MDR1 geneexpression in human metastatic sarcoma after in vivo exposure to doxorubicin.Clin Cancer Res 5(11):3352-3356, 1999.

23. Honkoop AH, Luykx-de Bakker SA, Hoekman K, et al: Prolonged neoadjuvantchemotherapy with GM-CSF in locally advanced breast cancer. Oncologist4:106-111, 1999.

24. O’Reilly MS, Holmgren L, Shing Y, et al: Angiostatin: A novelangiogenesis inhibitor that mediates the suppression of metastases by a Lewislung carcinoma. Cell 79:315-328, 1994.

25. Attia-Sobol J, Ferriere JP, Cure H, et al: Treatment results, survivaland prognostic factors in 109 inflammatory breast cancers: Univariate andmultivariate analysis. Eur J Cancer 29A:1081-1088, 1993.

26. Chevallier B, Bastit P, Graic Y, et al: The Centre H. Becquerel studiesin inflammatory nonmetastatic breast cancer. Combined modality approach in 178patients. Br J Cancer 67:594-601, 1993.

27. Karlsson YA, Malmstrom PO, Hatschek T, et al: Multimodality treatment of128 patients with locally advanced breast carcinoma in the era of mammographyscreening using standard polychemotherapy with 5-fluorouracil, epirubicin, andcyclophosphamide: Prognostic and therapeutic implications. Cancer 83:936-947,1998.

28. Rouesse J, Friedman S, Sarrazin D, et al: Primary chemotherapy in thetreatment of inflammatory breast carcinoma: A study of 230 cases from theInstitut Gustave-Roussy. J Clin Oncol 4:1765-1771, 1986.

29. Machiels JP, Reilly RT, Emens LA, et al: Cyclophosphamide, doxorubicin,and paclitaxel enhance the antitumor immune response of granulocyte/macrophage-colonystimulating factor-secreting whole-cell vaccines in HER-2/neu tolerized mice.Cancer Res 61(9):3689-3697, 2001.

30. Carmo-Pereira J, Costa FO, Henriques E, et al: A comparison of two dosesof adriamycin in the primary chemotherapy of disseminated breast carcinoma. Br JCancer 56(4):471-473, 1987.

31. Frei E III, Teicher BA, Holden SA, et al: Preclinical studies andclinical correlation of the effect of alkylating dose. Cancer Res48(22):6417-6423, 1988.

32. Honkoop AH, Hoekman K, Wagstaff J, et al: Dose-intensive chemotherapywith doxorubicin, cyclophosphamide and GM-CSF fails to improve survival ofmetastatic breast cancer patients. Ann Oncol 7:35-39, 1996.

33. Hryniuk W, Bush H: The importance of dose intensity in chemotherapy ofmetastatic breast cancer. J Clin Oncol 2(11):1281-1288, 1984.

34. Jones RB, Holland JF, Bhardwaj S, et al: A phase I-II study ofintensive-dose adriamycin for advanced breast cancer. J Clin Oncol 5(2):172-177,1987.

35. Hoekman K, Wagstaff J, van Groeningen CJ, et al: Effects of recombinanthuman granulocyte-macrophage colony-stimulating factor on myelosuppressioninduced by multiple cycles of high-dose chemotherapy in patients with advancedbreast cancer. J Natl Cancer Inst 83:1546-1553, 1991.

36. Weisbart RH, Gasson JC, Golde DW: Colony-stimulating factors and hostdefense. Ann Intern Med 110(4):297-303, 1989.

37. Disis ML, Calenoff E, McLaughlin G, et al: Existent T-cell and antibodyimmunity to HER-2/neu protein in patients with breast cancer. Cancer Res54:16-20, 1994.

38. Schlom J, Kantor J, Abrams S, et al: Strategies for the development ofrecombinant vaccines for the immunotherapy of breast cancer. Breast Cancer ResTreat 38(1):27-39, 1996.

39. Southall PJ, Boxer GM, Bagshawe KD, et al: Immunohistologicaldistribution of 5T4 antigen in normal and malignant tissues. Br J Cancer61(1):89-95, 1990.

40. von Mensdorff-Pouilly S, Verstraeten AA, Kenemans P, et al: Survival inearly breast cancer patients is favorably influenced by a natural humoral immuneresponse to polymorphic epithelial mucin. J Clin Oncol 18(3):574-583, 2000.

41. Gabrilovich DI, Corak J, Ciernik IF, et al: Decreased antigenpresentation by dendritic cells in patients with breast cancer. Clin Cancer Res3:483-490, 1997.

42. Banchereau J, Steinman RM: Dendritic cells and the control of immunity.Nature 392:245-252, 1998.

43. Luykx-de Bakker SA, de Gruijl TD, Scheper RJ, et al: Dendritic cells: Anovel therapeutic modality. Ann Oncol 10:21-27, 1999.

44. Mayordomo JI, Zorina T, Storkus WJ, et al: Bone marrow-derived dendriticcells serve as potent adjuvants for peptide-based antitumor vaccines. Stem Cells15:94-103, 1997.

45. Bell D, Chomarat P, Broyles D, et al: In breast carcinoma tissue,immature dendritic cells reside within the tumor, whereas mature dendritic cellsare located in peritumoral areas. J Exp Med 190:1417-1426, 1999.

46. Coppola D, Fu L, Nicosia SV, et al: Prognostic significance of p53,bcl-2, vimentin, and S100 protein-positive Langerhans cells in endometrialcarcinoma. Hum Pathol 29:455-462, 1998.

47. Gallo O, Bianchi S, Giannini A, et al: Correlations betweenhistopathological and biological findings in nasopharyngeal carcinoma and itsprognostic significance. Laryngoscope 101:487-493, 1991.

48. Ikeguchi M, Ikeda M, Tatebe S, et al: Clinical significance of dendriticcell infiltration in esophageal squamous cell carcinoma. Oncol Rep 5:1185-1189,1998.

49. Ishigami S, Aikou T, Natsugoe S, et al: Prognostic value of HLA-DRexpression and dendritic cell infiltration in gastric cancer. Oncology 55:65-69,1998.

50. Saito H, Tsujitani S, Ikeguchi M, et al: Relationship between theexpression of vascular endothelial growth factor and the density of dendriticcells in gastric adenocarcinoma tissue. Br J Cancer 78:1573-1577, 1998.

51. Ochsenbein AF, Klenerman P, Karrer U, et al: Immune surveillance againsta solid tumor fails because of immunological ignorance. Proc Natl Acad Sci U S A96(5):2233-2238, 1999.

52. Buelens C, Verhasselt V, De Groote D: Interleukin-10 prevents thegeneration of dendritic cells from human peripheral blood mononuclear cellscultured with interleukin-4 and granulocyte/macrophage-colony-stimulatingfactor. Eur J Immunol 27:756-762, 1997.

53. Gabrilovich D, Ishida T, Oyama T, et al: Vascular endothelial growthfactor inhibits the development of dendritic cells and dramatically affects thedifferentiation of multiple hematopoietic lineages in vivo. Blood 92:4150-4166,1998.

54. Gabrilovich DI, Chen HL, Girgis KR, et al: Production of vascularendothelial growth factor by human tumors inhibits the functional maturation ofdendritic cells [published erratum appears in Nat Med 2(11):1267, 1996]. Nat Med2:1096-1103, 1996.

55. Morel AS, Quaratino S, Douek DC, et al: Split activity of interleukin-10on antigen capture and antigen presentation by human dendritic cells: definitionof a maturative step. Eur J Immunol 27:26-34, 1997.

56. Gastman BR, Johnson DE, Whiteside TL, et al: Tumor-induced apoptosis of Tlymphocytes: Elucidation of intracellular apoptotic events. Blood95(6):2015-2023, 2000.

57. Lespagnard L, Gancberg D, Rouas G, et al: Tumor-infiltrating dendriticcells in adenocarcinomas of the breast: A study of 143 neoplasms with acorrelation to usual prognostic factors and to clinical outcome. Int J Cancer84:309-314, 1999.

58. Marrogi AJ, Munshi A, Merogi AJ, et al: Study of tumor infiltratinglymphocytes and transforming growth factor-beta as prognostic factors in breastcarcinoma. Int J Cancer 74(5):492-501, 1997.

59. Albert ML, Sauter B, Bhardwaj N: Dendritic cells acquire antigen fromapoptotic cells and induce class I-restricted CTLs. Nature 392:86-89, 1998.

60. Inaba K, Turley S, Yamaide F, et al: Efficient presentation ofphagocytosed cellular fragments on the major histocompatibility complex class IIproducts of dendritic cells. J Exp Med 188(11):2163-2173, 1998.

61. Steinman RM, Turley S, Mellman I, et al: The induction of tolerance bydendritic cells that have captured apoptotic cells. J Exp Med 191(3):411-416,2000.

62. Gallucci S, Lolkema M, Matzinger P: Natural adjuvants: Endogenousactivators of dendritic cells. Nat Med 5(11):1249-1255, 1999.

63. Sauter B, Albert ML, Francisco L, et al: Consequences of cell death:exposure to necrotic tumor cells, but not primary tissue cells or apoptoticcells, induces the maturation of immunostimulatory dendritic cells. J Exp Med191(3):423-434, 2000.

64. Armitage JO: Emerging applications of recombinant human granulocyte-macrophagecolony-stimulating factor. Blood 92:4491-4508, 1998.

65. Honkoop AH, Hoekman K, Wagstaff J, et al: Continuous infusion orsubcutaneous injection of granulocyte-macrophage colony-stimulating factor:Increased efficacy and reduced toxicity when given subcutaneously. Br J Cancer74:1132-1136, 1996.

66. Rabinovich GA, Riera CM, Iribarren P: Granulocyte-macrophagecolony-stimulating factor protects dendritic cells from liposome-encapsulateddichloromethylene diphosphonate-induced apoptosis through a Bcl-2-mediatedpathway. Eur J Immunol 29(2):563-570, 1999.

67. Lotze MT, Shurin M, Davis I, et al: Dendritic cell based therapy ofcancer. Adv Exp Med Biol 417:551-569, 1997.

68. Lu L, Qian S, Hershberger PA, et al: Fas ligand (CD95L) and B7 expressionon dendritic cells provide counter-regulatory signals for T cell survival andproliferation. J Immunol 158:5676-5684, 1997.

69. O’Connell J, O’Sullivan GC, Collins JK, et al: The Fas counterattack:Fas-mediated T cell killing by colon cancer cells expressing Fas ligand. J ExpMed 184:1075-1082, 1996.

70. O’Connell J, Bennett MW, O’Sullivan GC, et al: The Fas counterattack:A molecular mechanism of tumor immune privilege. Mol Med 3(5):294-300, 1997.

71. Walker PR, Saas P, Dietrich PY: Role of Fas ligand (CD95L) in immuneescape: The tumor cell strikes back. J Immunol 158:4521-4524, 1997.

72. Zeytun A, Hassuneh M, Nagarkatti M, et al: Fas-Fas ligand-basedinteractions between tumor cells and tumor-specific cytotoxic T lymphocytes: Alethal two-way street. Blood 90:1952-1959, 1997.

73. Cella M, Scheidegger D, Palmer-Lehmann K, et al: Ligation of CD40 ondendritic cells triggers production of high levels of interleukin-12 andenhances T cell stimulatory capacity: T-T help via APC activation. J Exp Med184:747-752, 1996.

74. Daniel PT, Kroidl A, Cayeux S, et al: Costimulatory signals throughB7.1/CD28 prevent T cell apoptosis during target cell lysis. J Immunol159:3808-3815, 1997.

75. Demicheli R, Terenziani M, Valagussa P, et al: Local recurrencesfollowing mastectomy: Support for the concept of tumor dormancy. J Natl CancerInst 86:45-48, 1994.

76. Holmgren L, O’Reilly MS, Folkman J: Dormancy of micrometastases:balanced proliferation and apoptosis in the presence of angiogenesissuppression. Nat Med 1:149-153, 1995.

77. O’Reilly MS, Boehm T, Shing Y, et al: Endostatin: An endogenousinhibitor of angiogenesis and tumor growth. Cell 88(2):277-285, 1997.

78. Hanahan D, Folkman J: Patterns and emerging mechanisms of the angiogenicswitch during tumorigenesis. Cell 86:353-364, 1996.

79. Fukumura D, Xavier R, Sugiura T, et al: Tumor induction of VEGF promoteractivity in stromal cells. Cell 94:715-725, 1998.

80. Pinedo HM, Verheul HM, D’Amato RJ, et al: Involvement of platelets intumour angiogenesis? Lancet 352:1775-1777, 1998.

81. Polverini PJ, Leibovich SJ: Induction of neovascularization in vivo andendothelial proliferation in vitro by tumor-associated macrophages. Lab Invest51:635-642, 1984.

82. Good DJ, Polverini PJ, Rastinejad F, et al: A tumor suppressor-dependentinhibitor of angiogenesis is immunologically and functionally indistinguishablefrom a fragment of thrombospondin. Proc Natl Acad Sci U S A 87:6624-6628, 1990.

83. Homandberg GA, Williams JE, Grant D, et al: Heparin-binding fragments offibronectin are potent inhibitors of endothelial cell growth. Am J Pathol120:327-332, 1985.

84. Lee TH, Rhim T, Kim SS: Prothrombin kringle-2 domain has a growthinhibitory activity against basic fibroblast growth factor-stimulated capillaryendothelial cells. J Biol Chem 273:28805-28812, 1998.

85. Morelli D, Lazzerini D, Cazzaniga S, et al: Evaluation of the balancebetween angiogenic and antiangiogenic circulating factors in patients withbreast and gastrointestinal cancers. Clin Cancer Res 4:1221-1225, 1998.

86. Verheul HM, Hoekman K, Luykx-de Bakker S, et al: Platelet: transporter ofvascular endothelial growth factor. Clin Cancer Res 3:2187-2190, 1997.

87. Verheul HM, Hoekman K, Lupu F, et al: Platelet and coagulation activationwith vascular endothelial growth factor generation in soft tissue sarcomas. ClinCancer Res 6(1):166-171, 2000.

88. Verheul HM, Jorna AS, Hoekman K, et al: Vascular endothelial growthfactor-stimulated endothelial cells promote adhesion and activation ofplatelets. Blood 96(13):4216-4221, 2000.

89. Joseph IB, Isaacs JT: Macrophage role in the anti-prostate cancerresponse to one class of antiangiogenic agents. J Natl Cancer Inst90(21):1648-1653, 1998.

90. Dong Z, Kumar R, Yang X, et al: Macrophage-derived metalloelastase isresponsible for the generation of angiostatin in Lewis lung carcinoma. Cell88:801-810, 1997.

91. Dong Z, Yoneda J, Kumar R, et al: Angiostatin-mediated suppression ofcancer metastases by primary neoplasms engineered to produce granulocyte/macrophagecolony-stimulating factor. J Exp Med 188:755-763, 1998.

92. Honkoop AH, Pinedo HM, de Jong JS, et al: Effects of chemotherapy onpathologic and biologic characteristics of locally advanced breast cancer. Am JClin Pathol 107:211-218, 1997.

93. Cavallo F, Di Carlo E, Butera M, et al: Immune events associated with thecure of established tumors and spontaneous metastases by local and systemicinterleukin 12. Cancer Res 59(2):414-421, 1999.

94. Duda DG, Sunamura M, Lozonschi L, et al: Direct in vitro evidence and invivo analysis of the antiangiogenesis effects of interleukin 12. Cancer Res60(4):1111-1116, 2000.

95. Bussolino F, Colotta F, Bocchietto E, et al: Recent developments in thecell biology of granulocyte-macrophage colony-stimulating factor and granulocytecolony-stimulating factor: activities on endothelial cells. Int J Clin Lab Res23(1):8-12, 1993.

96. Feder LS, Laskin DL: Regulation of hepatic endothelial cell andmacrophage proliferation and nitric oxide production by GM-CSF, M-CSF, and IL-1beta following acute endotoxemia. J Leukoc Biol 55(4):507-513, 1994.

97. Griffioen AW, Damen CA, Mayo KH, et al: Angiogenesis inhibitors overcometumor induced endothelial cell anergy. Int J Cancer 80:315-319, 1999.