Abstract

Background

Anticancer chemotherapy is thought to be effective by means of direct cytotoxicity on tumor cells. Alternative mechanisms of efficacy have been ascribed to several common anticancer agents, including cyclophosphamide (CTX), methotrexate (MTX), anthracyclines and taxanes, postulating an antiangiogenic activity.

Patients and methods

We evaluated the clinical efficacy and impact on serum vascular endothelial growth factor (VEGF) levels of low-dose oral MTX and CTX in patients with metastatic breast cancer. MTX was administered 2.5 mg bd on days 1 and 2 each week and CTX 50 mg/day administered continuously.

Results

Sixty-four patients were enrolled, 63 were evaluable: Eastern Cooperative Oncology Group (ECOG) performance status 0–1, ≥2 sites of metastatic disease (n = 50 patients), progressive disease at study entry (n = 51), 1 regimen for metastatic disease (n = 32) and ≥2 regimens (n = 20). Among the 63 evaluable patients, there were two complete remissions (CR), 10 partial remissions (PR) for an overall response rate of 19.0% (95% CI 10.2% to 30.9%) and an overall clinical benefit (CR+ PR+ stable disease >24 weeks) of 31.7% (95% CI 20.6% to 44.7%). Grade ≥2 leucopenia was registered in only 13 patients. The median serum VEGF level for the subgroup of patients on treatment for at least 2 months decreased with treatment from 315 pg/ml (95% CI 245 to 435) at baseline to 248 pg/ml (95% CI 205 to 311) at 2 months (P <0.001). Both responders and non-responders showed similar reductions in serum VEGF (P = 0.78). After 6 months patients still on treatment had a median VEGF level of 195 pg/ml (95% CI 96 to 355), which was significantly lower than the median baseline values (P = 0.001).

Conclusions

Continuously low-dose CTX and MTX is minimally toxic and effective in heavily pretreated breast cancer patients. A drop in VEGF was associated with the treatment and so alternative hypotheses, other than that of direct toxicity on tumor cells, must be favored when trying to explain the anticancer effect.

Received 29 March 2001; revised 26 June 2001; accepted 5 July 2001.

Introduction

Neoangiogenesis plays a key role in tumor progression of solid neoplasm [1–4]. Tumor cells may induce angiogenesis via the release of numerous growth factors, prostaglandin, etc., and by their attraction of inflammatory cells which in turn release multiple angiogenic stimuli. A number of antiangiogenic agents have been recently discovered, and some are under early clinical evaluation [5–7]. In animal models, treatment with angiogenesis inhibitors has a proven antitumor effect in vivo, and can both reduce metastases and lead to regression of primary growth by necrosis following capillary retraction [8, 9].

There is evidence that several common anticancer agents including cyclophosphamide (CTX), doxorubicin (Adriamycin) and paclitaxel (Taxol) have antiangiogenic activity in the animal model [8, 10, 11]. Low-dose methotrexate (MTX) inhibits endothelial cell proliferation in vitro, and blocks endothelial cell growth factor-induced neovascularization in the rabbit cornea assay [10]. Cytotoxic agents including CTX and MTX [11] markedly inhibited the growth of chick embryo. It is therefore possible that conventional cytotoxic agents may exert a tumor suppressive effect through an antiangiogenic mechanism. Antiangiogenic therapy appears to be improved when administered over a long period of time without treatment interruptions. Therefore, oral therapy in doses, which would be tolerable on a long run, should be considered. Among orally available drugs for the treatment of breast cancer, significant bioavailability was demonstrated for both CTX and MTX [12, 13].

Vascular endothelial growth factor (VEGF) is an angiogenic polypeptide that is detected in a large variety of malignant human tumors including breast, brain, lung and gastrointestinal tract cancers [14, 15]. Overexpression of VEGF increased tumor growth, angiogenesis and metastases, whereas anti-VEGF antibodies inhibited the growth of tumor xenografts [16, 17]. Moreover, growth inhibition of glioma was observed in nude mice bearing this tumor derived from cells transfected with an antisense VEGF sequence [18]. Values of VEGF demonstrated a correlation with tumor bulk. Patients with disseminated breast cancer had higher serum VEGF concentrations than those with localized cancer. In particular, VEGF was found to be abnormally elevated in the serum of more than 70% of patients with advanced breast cancer [14]. VEGF content in tumor cytosols is a predictor of relapse free survival and overall survival in primary breast cancer and it might also predict outcome after adjuvant endocrine treatment [19].

We therefore evaluated the clinical efficacy and tolerance of low-dose, oral CTX and MTX in metastatic breast cancer patients and also the impact of this regimen upon serum VEGF values.

Patients and methods

Patient selection

Patients included in the study were required to have histologically confirmed metastatic breast carcinoma that had either progressed, or not, after a first line of chemotherapy for metastatic disease. Starting in June 1998, patients that asked for a regimen characterized by few side effects were included in the study, although they had not previously received chemotherapy for metastatic disease. Other inclusion criteria were measurable or evaluable disease, age ≤80 years, performance status Eastern Cooperative Oncology Group (ECOG) ≤3, adequate bone marrow reserve defined as white blood cells >4000 mm3 and platelets >100000mm3, adequate renal function (serum creatinine <120 µmol/l) and hepatic function (serum bilirubin <20 µmol/l, aspartate aminotransferase <60 IU/l). Any prior hormonal therapy or chemotherapy must have been discontinued at least 4 weeks before study entry. All patients must have recovered from any prior chemotherapy, radiotherapy or surgery before study entry. Each patient included in the study gave a written informed consent.

Study evaluation and treatment

Baseline evaluation included clinical examination, chest X-ray, liver ultrasound or CT scan, bone nuclear scan (plus segmental bone radiographs when bone scans were positive), ECG, complete biochemical and hematological tests. Complete blood count was then repeated every 14 days and biochemical tests every 28 days.

Treatment was administered as follows: MTX orally at a dose of 2.5 mg bd (10 am, 5 pm) on day 1 and 2 every week and CTX orally at a dose of 50 mg/day (9 am). No antiemetic treatment was prescribed. Serum VEGF was determined at baseline and every month thereafter. Fifteen milliliters of peripheral blood were collected partly into a citrate tube, for plasmatic VEGF determination, and partly into a clot-activator tube for serum VEGF determination. As the value of serum VEGF increased during clot formation and reached a plateau after 60 min [20] the samples were allowed to clot at room temperature for 2 h and then centrifuged for 10 min at 1900g. The aliquoted samples were then stored at –30°C until analysis was performed. VEGF was assessed by an ELISA kit (Quantikine®, R&D Systems, Minneapolis, USA). All samples and standards were assayed in duplicate. The serum was stocked in aliquots at –30°C until analysis was performed. The samples were analyzed in duplicate by a sandwich enzyme-linked immunosorbent assay (ELISA), using a monoclonal antibody specific for VEGF165 (Quantikine®). In order to evaluate possible differences in VEGF values between serum and plasma, 29 patients had a control value of VEGF in both serum and plasma for each sampling.

Side effects and response

Toxicity was evaluated according to NCI–CTG criteria by clinical and laboratory investigations. Treatment was withheld and delayed for 1 week in case of a neutrophil count <1000 mm3 and/or platelet count <75000 mm3. A 50% dose reduction in the total amount of drug administered in each cycle was prescribed after hematological recovery. In the case of a neutrophil count <1500 mm3 but >1000 mm3 and/or platelet count <100000 mm3 but >75000 mm3, therapy was administered with a 50% dose reduction in the total amount of drug administered in each cycle. Re-escalation of drug doses was attempted if close monitoring was possible.

In the event of grade ≥2 anorexia, nausea, vomiting, diarrhea, stomatitis, dryness of the mouth, epigastric pain or increase in transaminases, all therapy was postponed until symptoms subsided. A 50% reduction of combined CTX and MTX therapy was performed for the next cycle, with subsequent re-escalation to full dosage if tolerated. Any other non-hematological grade 3 toxicity was managed by a 50% reduction of dosage in the next cycle, which was not commenced until full recovery had occurred.

Assessment of response was performed according to WHO criteria after every 8 weeks of therapy. Complete remission (CR) was defined as the disappearance of all known lesions on two separate measurements at least 4 weeks apart. Partial remission (PR) was defined as a reduction of each lesion by at least 50%. Stable disease (SD) was defined as a decrease of <50% or an increase of <25% with no new lesions, and progressive disease as an increase of >25% or appearance of new lesions. Overall success rate was defined as the proportion of patients who achieved CRs, PRs or SDs for at least 24 weeks. Case report forms were reviewed by an independent panel composed of two medical oncologists.

Statistical analysis

The aim of the study was to obtain at least a 25% overall success rate. With 63 patients the power to detect a difference in response rate from 10% to 30% is 98% and from 10% to 25% is 91%. Estimated curves of survival and time to progression were plotted from the first day of treatment using the Kaplan–Meier method; response duration was measured from the date of achievement of response. Confidence intervals (CI) for the response rates were calculated using exact binomial methods. Given the non-normality of VEGF distribution in our sample (Shapiro–Wilk test), the non-parametric Wilcoxon matched pairs test was applied to compare VEGF values at different assay times, while the Mann–Whitney U test was performed to compare responders and non-responders with respect to their baseline VEGF levels. Relative changes in VEGF levels were calculated as ratio of the values measured at 2 months and at baseline and compared between responders and non-responders by the Mann–Whitney U test. Confidence intervals for the medians were obtained using 2000 bootstrap samples. Log transformed VEGF and platelet values were used in the correlation analysis as well as to estimate the effect of VEGF in the prediction of a response using logistic regression.

Results

Sixty-four patients were enrolled between July 1997 and May 2000. From June 1998, 44 patients were enrolled, 11 of them untreated with chemotherapy for metastatic disease. One patient was considered ineligible because of failure to meet one of the eligibility criteria. All but 11 patients had previously received at least one line of chemotherapy for metastatic disease and 20 patients received at least two lines. Fifty-one patients had progressive disease at study entry. All the patients had good performance status and 10 patients were ≥70 years old (Table 1).

Oral CTX plus MTX produced two CRs and 12 PRs, providing an objective response rate of 19.0% (95% CI 10.2% to 30.9%) in the intention-to-treat analysis of all eligible patients by peer review assessment. An additional eight patients had long-term disease stabilization [no change (NC) after 24 weeks], providing an overall success (CR + PR + NC after 24 weeks) rate of 31.7%. (95% CI 20.6% to 44.7%). Five patients among those who achieved prolonged disease stabilization had progressive disease at study entry. The characteristics of responding patients are shown in Table 2. All but three patients were pre-treated for metastatic disease and 10 of them had progressive disease at study entry. Three responding patients previously received CTX, MTX and fluorouracil (CMF)-like chemotherapy as adjuvant therapy and one for metastatic disease. Three patients received the combination of Adriamycin and cyclophosphamide (AC) for metastatic disease with progressive disease documented during the AC treatment in one patient. The median time to response was 2.7 (95% CI 2.0 to 3.3) months and, among responding patients, the median duration of response was 6.8 (95% CI 3.7 to 9.7) months. For all 63 patients the median time to progression was 2.8 (95% CI 2.1 to 5.9) months. Twenty-six per cent of patients were still responding after 12 months.

A total of 361 months of therapy were administered, with a median administration time per patient of 2.5 months. Only 37 (10%) of the cycles were delayed and 25 (7%) of the courses administered at reduced dosages. Reason for delay and dose reduction was mainly due to leukopenia and an increase in transaminases. Table 3 summarizes the side effects observed. Treatment was well tolerated. The most frequently encountered toxicity was grade I leukopenia, which was observed in 35% of cases. An increase in transaminase values was registered in 32 cases. Six out of nine patients that presented grade III hepatic toxicity had concomitant liver metastases. A complete recovery of function was achieved with either reduction or transient interruption of MTX.

Forty-eight patients had serum VEGF levels measured at baseline and at 2 months (Table 4). The median VEGF level for the group of 48 patients decreased with treatment from 314.5 pg/ml (95% CI 245 to 435) at baseline to 248.5 pg/ml (95% CI 205 to 310.9) at 2 months (P <0.001, Wilcoxon matched pairs test), with a median relative change of 0.75 (95% CI 0.62 to 0.83). Among responding patients the median reduction in serum VEGF was 71.5 pg/ml (95% CI 42.5 to 130), while among non-responding patients the median reduction was 73 pg/ml (95% CI –44 to 208). There is no significant difference in the median reductions (P = 0.77). Although the reduction in VEGF was of a similar magnitude in the two subgroups of patients, the reduction was statistically significant only in the subgroup of responding patients (P <0.001) (Figure 1). The median relative changes in responding and non-responding patients were 0.75 (95% CI 0.63 to 0.88) and 0.83 (05% CI 0.50 to 1.11), respectively. The median baseline VEGF level did not differ between the two groups (P = 0.45, Mann–Whitney U test). In 16 patients presenting long-term response the value of VEGF at 6 months remained at lower levels than baseline, median 195 pg/ml (95% CI 96 to 355) while response was maintained.

Baseline values after 2 months, and the ratio of after 2 months to baseline medians, (and 95% CI) are presented for platelets and VEGF in Table 4. Data are shown for responders, non-responders and both groups combined. The P values shown in Table 4 are for a comparison of responders with non-responders. Platelet values were higher at baseline among patients who subsequently did not respond (P = 0.01). As shown in Table 4 a decrease in the median number of platelets was observed only for non-responding patients. The median number of platelets did not change significantly for the whole sample of patients. No significant correlation was detected between serum VEGF values and platelets at baseline [correlation coefficient (r) = 0.25, P = 0.10] or at 2 months (r= 0.14, P = 0.36). Using logistic regression analysis there is no evidence that baseline serum VEGF is associated with predicting a response (P = 0.23). The relative change in serum VEGF from baseline to 2 months is also not predictive of a response (P = 0.69).

In 29 patients for whom measurements of VEGF values in both serum and plasma for each sample were taken, a good correlation between plasma and serum levels was observed at baseline as indicated by the correlation coefficient of 0.51 (P <0.01). At baseline the median plasma level of VEGF was 29.2 pg/ml (95% CI 23.8 to 40.9) while at 2 months it was 34.8 pg/ml (95% CI 24.3 to 39.5); there was no evidence of a change in plasma VEGF from baseline to 2 months (P = 0.94) nor that there was a difference between responders and non-responders (P = 0.17).

Discussion

A dose-related cytotoxic effect on tumor cells has been postulated as the main cause of anti-neoplastic efficacy using chemotherapy for the treatment of solid tumors. This was based mainly upon the historical development of anticancer agents tested primarily in leukemia or cell cultures. Several studies in advanced breast cancer were, therefore, aimed at increasing the dose of chemotherapeutic agents to take advantage of the steep dose-intensity effect [21]. In fact, the principle of dose intensity is supported by experimental model systems where a small increase in drug dose may result in a large addition in tumor cell kill [21–24]. Bonadonna et al. first reported the favorable outcome of patients treated with adjuvant cyclophosphamide, fluorouracil and methotrexate at 5 and 10 years when given the full-dose treatment compared with those given a reduced dose [25]. The issue of dose intensity and efficient tumor kill in advanced breast cancer is even more controversial [26–28] since interactions between drugs, tumor cells, stroma, hormones (and endocrine organs), growth factors and vessels is highly complex [29].

In view of the goal of palliative treatment in this setting, it is reasonable to consider disease stabilization (i.e. NC for 24 weeks or longer) as a clinical outcome [30]. The overall success rate (CR + PR + NC for 24 weeks or longer) of 32% suggests that a meaningful percentage of patients in our study benefited from treatment with low-dose chemotherapy. The observation of a 19% response rate, using a regimen with CTX and MTX at a very low-dose level, indicates that cytotoxicity cannot possibly explain entirely the efficacy of this regimen. In fact, direct cytotoxicity of anticancer agents is expected to cause bone marrow suppression, hair loss and altered mucosal repair. Conversely, this regimen demonstrated no significant hematological side effects or hair loss. Only 21% of the patients presented a grade >1 leucopenia; 12%, grade >1 neutropenia; and 8% had some hair loss. Moreover, no significant correlation was detected between the tumor shrinkage and the most commonly used surrogate indicator of cytotoxicity, i.e. leucopenia. The absence of significant bone marrow suppression, mucositis or hair loss, usually observed with standard dose CTX (± MTX), leads to a hypothetical alternative target. One plausible explanation for cell-growth inhibition might relate to the antiangiogenic effects of CTX (and also doxorubicin but not 5-fluorouracil) as postulated by Folkman et al., based upon a neovascularization model in the mouse cornea [9]. Also, low-dose MTX inhibited endothelial cell proliferation in vitro, and inhibited neovascularization by endothelial cell growth factor in the rabbit cornea assay [10].

We assessed a possible correlation between angiogenic pathways and efficacy of the regimen by repeated determination of VEGF concentrations in the serum of patients during treatment. Table 4 illustrates the association between VEGF levels and response. Values were elevated in all patients before treatment and no significant difference in baseline values was detected between patients who subsequently responded to the treatment and patients that subsequently failed to respond.

Recently published results indicate that VEGF in the bloodstream is transported by blood cells, including leukocytes and platelets [31, 33], and a correlation between VEGF value and platelet number has been observed [33]. The blood cells of cancer patients contain greatly elevated amounts of this major angiogenic growth factor [32, 33]. Therefore changes in VEGF during chemotherapy might be related with chemotherapy-induced thrombocytopenia and a subsequent rebound increase in platelet numbers rather than the persistent decrease in tumor-derived VEGF [34]. Moreover the finding of lower VEGF in patients undergoing chemotherapy compared with those untreated may reflect effects that not only have an influence on platelet numbers but also on platelet volume [35]. Although a possible influence of platelet volume on VEGF values cannot be excluded in the present study, no significant change in the platelet numbers was observed and serum VEGF significantly decreased after 2 months of treatment. Conversely, in a subgroup of patients where we tested both serum and plasma concentrations of VEGF, we showed that although there is a good correlation between serum and plasmatic values, there was no evidence of a change in plasma VEGF from baseline to 2 months. The role of platelets and particularly of platelet-derived VEGF in tumor biology is obscure, but not negligible. Cancer patients frequently have elevated platelet counts and an increased platelet consumption compared with healthy individuals. Moreover, thrombocytosis is a negative prognostic factor in some cancers, and antitumor effects and improvements in terms of survival have been observed with anticoagulation therapy [36]. Other factors may influence serum VEGF concentrations other than tumor and platelets and therefore might have influenced the results observed. Studies have found variable correlations with hormones and have variably reported changes in the levels of VEGF concentration according to the phase of the menstrual cycle [37–39]. Finally, VEGF concentrations in peripheral blood mononuclear cells have been reported in cancer patients [40].

Important aspects such as the chronic use of CTX (and MTX) should be raised for the sake of discussion, especially because the response duration might last for several months. Cyclophosphamide, given together with MTX and fluorouracil has been reported to slightly increase the incidence of leukemia [41]. Dealing with leukemic risk while discussing treatment for advanced breast cancer might be futile. However, the possible use of this regimen in the adjuvant setting might require a careful estimation of such risk. Even a year of treatment, using a cumulative CTX dose of 18 g will produce only a small increase in the risk of leukemia [41]. This figure is lower than that observed with anthracyclines, which are commonly used in the adjuvant setting [42]. Also, changes of transitional epithelium [43] and immunosuppression [44, 45] are associated with prolonged use of CTX. While instructions for an abundant fluid intake evade the former, the latter side effect might require a pause in the admin-istration of the drug(s). We encountered neither of these.

A side issue, which is worth mentioning, relates to the personal and economical costs of this regimen. Increased attention to patients’ quality of life favors the use of an active oral treatment [46]. Advanced breast cancer responsive to chemotherapy might require a treatment duration of several months in responders [47]. Subjective toxicity and frequent visits to care providers represents a significant burden in terms of personal costs to the patient. Furthermore, the utility of a treatment regimen which has an economic cost of about US$10 per month is obvious.

In conclusion, low-dose, oral CTX and MTX demonstrated significant efficacy in pre-treated metastatic breast cancer. Theoretically, treatments aimed at inhibiting angiogenesis should be chronically administered for a prolonged period. These treatments might have particular usefulness for subsets of patients with limited tumor burden, such as early breast cancer [48]. Moreover, combinations of therapy that inhibit angiogenesis plus cytotoxic therapy may be more effective than either type of therapy alone. Use of oral low-dose CTX and MTX should be investigated further as a strategy against tumor progression after standard chemotherapy in the adjuvant setting.

+

Correspondence to: Via Ripamonti 435, 20141 Milan, Italy. Tel: +39-02-57489439; Fax: +39-02-574829212; E-mail: marco.colleoni@ieo.it

Figure 1. Serum VEGF concentrations (pg/ml) at baseline and at 2 months for evaluable RP+RC+SD patients (A) and for progressive disease (PD) patients (B). The horizontal axes are plotted on a logarithmic scale.

Figure 1. Serum VEGF concentrations (pg/ml) at baseline and at 2 months for evaluable RP+RC+SD patients (A) and for progressive disease (PD) patients (B). The horizontal axes are plotted on a logarithmic scale.

Table 1.

Characteristics of eligible patients

Entered/eligible 64/63 
Median age (Range) 57 (36–80) 
Menopausal status pre/post 12/51 
Progressive disease at study entry 51 
ER and PgR negative 25 
ER and/or PgR positive 31 
Number of involved sites 13/23/27 
1/2/>2 16 
Tumor sites 24 
Lung 38 
Liver 34 
Soft tissues 14 
Bone 41 
Other 52 
Chemotherapy 32 
Adjuvant–neoadjuvant 11 
for metastatic disease  9 
 1 line  
 2 lines  
 ≥3 lines  
Entered/eligible 64/63 
Median age (Range) 57 (36–80) 
Menopausal status pre/post 12/51 
Progressive disease at study entry 51 
ER and PgR negative 25 
ER and/or PgR positive 31 
Number of involved sites 13/23/27 
1/2/>2 16 
Tumor sites 24 
Lung 38 
Liver 34 
Soft tissues 14 
Bone 41 
Other 52 
Chemotherapy 32 
Adjuvant–neoadjuvant 11 
for metastatic disease  9 
 1 line  
 2 lines  
 ≥3 lines  
Table 2.

Characteristics of responding patients

ER/PgRa Age Previous chemotherapyb PD at study entryc Sites of response 
49 3 lines Yes Soft tissue 
42 1 line No Bone, pleural infusion 
57 4 line Yes Liver 
63 0 lines Yes Soft tissue 
56 4 lines Yes Liver 
69 1 line Yes Lung 
67 1 line Yes lung 
64 1 line Yes Skin 
44 2 lines No Liver, breast 
53 0 lines Yes Skin 
39 1 line Yes Liver 
65 0 lines Yes Lung 
ER/PgRa Age Previous chemotherapyb PD at study entryc Sites of response 
49 3 lines Yes Soft tissue 
42 1 line No Bone, pleural infusion 
57 4 line Yes Liver 
63 0 lines Yes Soft tissue 
56 4 lines Yes Liver 
69 1 line Yes Lung 
67 1 line Yes lung 
64 1 line Yes Skin 
44 2 lines No Liver, breast 
53 0 lines Yes Skin 
39 1 line Yes Liver 
65 0 lines Yes Lung 

a1 = Er < 10%/PgR < 10%; 2 = other.

bFor advanced disease.

cPD = progressive disease.

Table 3.

Side effects on 63 eligible patients

Side effect NCIC–CTG grade 
          
 No.  No.  No.  No.  No. 
Leukopenia 28 44  22 35  12 19  –  –  
Neutropenia 42 66  14 22   6 10  –  –  
Thrombocytopenia 60 95   3  5   0  –  –  –  – – 
Anaemia 54 86   8 12   0  –   2  – – 
Alopecia 58 92   4  6   1  2  –  –  – – 
Nausea/vomiting 47 75  13 20   3  5  –  –  – – 
Gastric pain 61 97   2  3   –  –  –  –  – – 
Mucositis 60 95   3  5   –  –  –  –  – – 
Transaminases 31 49  11 17  12 19  15  – – 
Side effect NCIC–CTG grade 
          
 No.  No.  No.  No.  No. 
Leukopenia 28 44  22 35  12 19  –  –  
Neutropenia 42 66  14 22   6 10  –  –  
Thrombocytopenia 60 95   3  5   0  –  –  –  – – 
Anaemia 54 86   8 12   0  –   2  – – 
Alopecia 58 92   4  6   1  2  –  –  – – 
Nausea/vomiting 47 75  13 20   3  5  –  –  – – 
Gastric pain 61 97   2  3   –  –  –  –  – – 
Mucositis 60 95   3  5   –  –  –  –  – – 
Transaminases 31 49  11 17  12 19  15  – – 
Table 4.

Serum modifications of VEGF (pg/ml) according to treatment response on 48 evaluable patients

 RP + RC + SD (n = 29) PD (n = 19)a All patients (n = 48) P valueb 
Pre-treatment VEGF 315.2 293 314.5 0.45 
Median (95% CI) (238–435) (218–688) (245–435)  
Post-treatment VEGF 222 304.3 248.5 0.23 
Median (95% CI) (148–262) (212–329.5) (205–310.9)  
VEGF ratio (post/pre) 0.75 0.83 0.75 0.53 
Median (95% CI) (0.62–0.77) (0.5–1.18) (0.62–0.85)  
Pre-treatment platelets ×109/l 209 269 230.5 0.01 
Median (95% CI) (170–241) (214–289) (206–252.7)  
Post-treatment platelets ×109/l 212.5 223 216.5 0.69 
Median (95% CI) (191–253) (195–246) (195–237)  
Platelet ratio (post/pre) 1.04 0.90 0.96 0.04 
Median (95% CI) (0.91–1.18) (0.78–0.99) (0.87–1.03)  
 RP + RC + SD (n = 29) PD (n = 19)a All patients (n = 48) P valueb 
Pre-treatment VEGF 315.2 293 314.5 0.45 
Median (95% CI) (238–435) (218–688) (245–435)  
Post-treatment VEGF 222 304.3 248.5 0.23 
Median (95% CI) (148–262) (212–329.5) (205–310.9)  
VEGF ratio (post/pre) 0.75 0.83 0.75 0.53 
Median (95% CI) (0.62–0.77) (0.5–1.18) (0.62–0.85)  
Pre-treatment platelets ×109/l 209 269 230.5 0.01 
Median (95% CI) (170–241) (214–289) (206–252.7)  
Post-treatment platelets ×109/l 212.5 223 216.5 0.69 
Median (95% CI) (191–253) (195–246) (195–237)  
Platelet ratio (post/pre) 1.04 0.90 0.96 0.04 
Median (95% CI) (0.91–1.18) (0.78–0.99) (0.87–1.03)  

aPD = progressive disease.

bP value from the Wilcoxon–Mann–Whitney U test comparing the medians of the responders and non-responders.

References

1.
Folkman J. The influence of angiogenesis research on management of patients with breast cancer.
Breast Cancer Res Treat
 
1995
;
36
:
109–
118.
2.
Kern FG, Lippman ME. The role of angiogenic growth factors in breast cancer progression.
Cancer Metastasis Rev
 
1996
;
15
:
213–
219.
3.
Leek RD, Lewis CE, Whitehouse R et al. Association of macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma.
Cancer Res
 
1996
;
15
:
4625–
4629.
4.
Weidner N. Angiogenesis in breast cancer.
Cancer Treat Res
 
1996
;
83
:
265–
301.
5.
Sledge GW Jr . Implications of the new biology for therapy in breast cancer.
Semin Oncol
 
1996
;
23
:
76–
81.
6.
Harris AL, Zhang H, Moghaddam A et al. Breast cancer angiogenesis: new approaches to therapy via antiangiogenesis, hypoxic activated drugs, and vascular targeting.
Breast Cancer Res Treat
 
1996
;
38
:
97–
108.
7.
Dickson RB, Johnson MD, Maemura M, Low J. Anti-invasion drugs.
Breast Cancer Res Treat
 
1996
;
38
:
121–
132.
8.
Klauber N, Parangi S, Flynn E et al. Inhibition of angiogenesis and breast cancer in mice by the microtubule inhibitors 2-methoxyestradiol and taxol.
Cancer Res
 
1997
;
57
:
81–
86.
9.
Folkman J. New perspectives in clinical oncology from angiogenesis research.
Eur J Cancer
 
1996
;
32
:
2534–
2539.
10.
Hirata S, Matsubara T, Saura R et al. Inhibition of in vitro vascular endothelial cell proliferation and in vivo neovascularisation by low-dose methotrexate.
Arthritis Rheum
 
1989
;
32
:
1065–
1073.
11.
Steiner R. Angiostatic activity of anticancer agents in the chick allantoic membrane (CHE–CAM) assay. In Steiner R, Weisz PB, Langer R (eds): Angiogenesis: Key Principles–Science–Technology–Medicine. Basel: Birkhaeuser Verlag
1992
;
449–
454.
12.
Wagner T, Fenneberg K. Pharmacokinetics and bioavailability of cyclophosphamide from oral formulations.
Arzneimittelforschung
 
1984
;
34
:
313–
316.
13.
Christophidis N, Vajda FJ, Lucas I et al. Comparison of intravenous and oral high-dose methotrexate in treatment of solid tumours.
Br Med J
 
1979
;
1
:
298–
300.
14.
Salven P, Maenpaa H, Orpana A. Serum vascular endothelial growth factor is often elevated in disseminated cancer.
Clin Cancer Res
 
1997
;
3
:
647–
651.
15.
Toi M, Inada K, Suzuki H, Tominaga T. Tumor angiogenesis in breast cancer: its importance as a prognostic indicator and the association with vascular endothelial growth factor expression.
Breast Cancer Res Treat
 
1995
;
36
:
193–
204.
16.
Senger DR, van de Water L, Brown L et al. Vascular permeability factor (VPF, VEGF) in tumor biology.
Cancer Metastasis Rev
 
1993
;
12
:
303–
324.
17.
Ferrara N. The role of vascular endothelial growth factor in pathological angiogenesis.
Breast Cancer Res Treat
 
1995
;
36
:
127–
137.
18.
Saleh M, Stacker SA, Wilks AF. Inhibition of growth of C6 glioma cells in vivo by expression of antisense vascular endothelial growth factor sequence.
Cancer Res
 
1996
;
56
:
393–
401.
19.
Linderholm B, Grankvist K, Wilking N et al. Correlation of vascular endothelial growth factor content with recurrences, survival, and first relapse site in primary node-positive breast carcinoma after adjuvant treatment.
J Clin Oncol
 
2000
;
18
:
1423–
1431.
20.
Hyodo I, Doi T, Endo H et al. Clinical significance of plasma vascular endothelial growth factor in gastrointestinal cancer.
Eur J Cancer
 
1998
;
34
:
2041–
2045.
21.
Henderson IC, Hayes DF, Gelman R. Dose–response in the treatment of breast cancer: a critical review.
J Clin Oncol
 
1988
;
6
:
1501–
1515.
22.
Von Hoff DD, Clark GM, Weiss GR et al. Use of in vitro dose–response effects in select antineoplastics for high-dose or regional administration regimens.
J Clin Oncol
 
1986
;
4
:
1827–
1834.
23.
Schabel FM, Griswold DP, Corbett TH et al. Increasing the therapeutic response rates to anticancer drugs by applying the basic principles of pharmacology.
Cancer
 
1984
;
54
:
1160–
1167.
24.
Hill RP, Stanley JA. Pulmonary metastases of the Lewis lung tumor-cell kinetics and response to cyclophosphamide at different sizes.
Cancer Treat Rep
 
1977
;
61
:
29–
36.
25.
Bonadonna G, Valagussa P. Dose–response effect of adjuvant chemotherapy in breast cancer.
N Engl J Med
 
1981
;
304
:
10–
15.
26.
Canellos GP. The dose dilemma.
J Clin Oncol
 
1988
;
6
:
1363–
1364.
27.
Redmond C, Fisher B, Wjeand HS. The methodological dilemma in retrospectively correlating the amount of chemotherapy received in adjuvant therapy protocols with disease-free survival.
Cancer Treat Rep
 
1983
;
67
:
519–
526.
28.
Hrynjuk W, Bush H. The importance of dose intensity in chemotherapy of metastatic breast cancer.
J Clin Oncol
 
1984
;
2
:
1281–
1288.
29.
Schnipper LE. Clinical implications of tumor-cell heterogencity.
N Engl J Med
 
1986
;
314
:
1423–
1431.
30.
Hortobagyi GN. Drug therapy: treatment of metastatic breast cancer.
N Engl J Med
 
1998
;
339
:
974–
984.
31.
Salgado R, Vermeulen PB, Benoy I et al. Platelet number and interleukin-6 correlate with VEGF but not with bFGF serum levels of advanced cancer patients.
Br J Cancer
 
1999
;
80
:
892–
897.
32.
Salven P, Orpana A, Joensuu H. Leukocytes and platelets of patients with cancer contain high levels of vascular endothelial growth factor.
Clin Cancer Res
 
1999
;
5
:
487–
489.
33.
Adams J, Carder PJ, Downey S et al. Vascular endothelial growth factor (VEGF) in breast cancer: comparison of plasma, serum, and tissue VEGF and microvessel density and effects of tamoxifen.
Cancer Res
 
2000
;
60
:
2898–
2905.
34.
Verheul HMW, Hoekman K, Luykx-de Bakker S et al. Platelet: transporter of vascular endothelial growth factor.
Clin Cancer Res
 
1997
;
3
:
2187–
2190.
35.
Gunsilius E, Gastl G. Platelets and VEGF blood levels in cancer patients.
Br J Cancer
 ,
1999
;
81
:
184–
186.
36.
Pinedo HM, Verheul HMW, D’Amato RJ, Folkman J. Involvement of platelets in tumour angiogenesis?
Lancet
 
1998
;
352
:
1775–
1777.
37.
Heer K, Kumar H, Speirs V et al. Vascular endothelial growth factor in premenopausal women—indicator of the best time for breast cancer surgery?
Br J Cancer
 
1998
;
78
:
1203–
1207.
38.
Agrawal R, Conway GS, Sladkevicius P et al. Serum vascular endothelial growth factor (VEGF) in the normal menstrual cycle: association with changes in ovarian and uterine Doppler blood flow.
Clin Endocrinol
 
1999
;
50
:
101–
106.
39.
Benoy I, Vermeulen P, Wuyts H, Dirix L. Vascular endothelial cell growth factor (VEGF) serum concentrations change according to the phase of the menstrual cycle.
Eur J Cancer
 
1998
;
34
:
1298–
1299.
40.
Salven P, Orpana A, Joensuu H. Leukocytes and platelets of patients with cancer contain high levels of vascular endothelial growth factor.
Clin Cancer Res
 
1999
;
5
:
487–
491.
41.
Curtis RE, Boice JD Jr, Stovall M et al. Risk of leukemia after chemotherapy and radiation treatment for breast cancer.
N Engl J Med
 
1992
;
326
:
1745–
1751.
42.
Levine MN, Bramwell VH, Pritchard KI et al. Randomized trial of intensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy compared with cyclophosphamide, methotexate, and fluorouracil in premenopausal women with node positive breast cancer.
J Clin Oncol
 
1998
;
162
:
2651–
2658.
43.
Kimura M, Tomita Y, Morishita H, Takahashi K. Presence of mucosal change in the urinary bladder in nonhematuric patients with long-term exposure and/or accumulating high-dose cyclophosphamide. Possible significance of follow-up cystoscopy on preventing development of cyclophosphamide-induced hemorrhagic cystitis.
Urol Int
 
1998
;
61
:
8–
11.
44.
Winkelstein A, Ruben RL, Tolchin SF, Pollock BH. Mechanisms of immunosuppression: effect of cyclophosphamide on responses to influenza immunization.
J Lab Clin Med
 
1974
;
83
:
504–
510.
45.
Winkelstein A, Mikulla JM, Nankin HR et al. Mechanisms of immunosuppression: effects of cyclophosphamide on lymphocytes.
J Lab Clin Med
 
1972
;
80
:
506–
513.
46.
DeMario MD, Ratain MJ. Oral chemotherapy: rationale and future directions.
J Clin Oncol
 
1998
;
16
:
2557–
2567.
47.
Henderson IC. Chemotherapy for metastatic disease. In Harris JR, Hellman S, Henderson JC, (eds): Breast Diseases, 2nd edition. Philadelphia, PA: Lippincott;
1991
;
604–
665.
48.
Bergers G, Javaherian K, Lo KM et al. Effects of angiogenesis inhibitors on multistage carcinogenesis in mice.
Science
 
1999
;
30
:
808–
812.

Author notes

1Division of Medical Oncology, 2Division of Laboratory Medicine and Pathology, 3Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy