SEARCH

SEARCH BY CITATION

Keywords:

  • epoetin-α;
  • radiotherapy;
  • esophageal carcinoma;
  • tumor oxygenation

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

It has been suggested that hemoglobin levels of 12–14 g/dL are optimal for tumor oxygenation, radiosensitivity, and prognosis. In this prospective study, the authors evaluated the effectiveness of epoetin-α to maintain hemoglobin levels at 12–14 g/dL during radiotherapy (RT) for patients with UICC Stage III esophageal carcinoma, and they examined the impact of erythropoetin on overall survival (OS), metastatic-free survival (MFS), and local control (LC).

METHODS

Sixty patients who received RT between March, 2001 and September, 2004, were included in this prospective, nonrandomized study. Thirty patients received epoetin-α (150 IU/kg 3 times per week) during RT (Group A), and 30 patients did not receive epoetin-α (Group B). Epoetin-α was started at hemoglobin levels < 13 g/dL and was stopped at hemoglobin levels ≥ 14 g/dL. Hemoglobin was measured before RT and once weekly during RT.

RESULTS

Both patient groups were balanced for age, gender, performance status, tumor location/length, histology, grading, tumor classification, lymph node status, chemotherapy, treatment (45–50.4 grays [Gy] plus resection vs. 45.0–50.4 Gy vs. 59.4–66.0 Gy), and hemoglobin level before RT. In 20 of 30 patients (67%) from Group A and in 3 of 30 patients (10%) from Group B, ≥ 60% of hemoglobin levels during RT were 12–14 g/dL (P = 0.003). The median change in hemoglobin was + 0.4 g/dL per week in Group A and − 0.4 g/dL per week in Group B. LC was significantly better in Group A (66% vs. 38% at 1 year, respectively; P = 0.012), a trend was observed for OS (59% vs. 33%, respectively; P = 0.08), and MFS did not differ significantly (43% vs. 38%, respectively; P = 0.34). No epoetin-α-related toxicity was observed.

CONCLUSIONS

Epoetin-α was effective in maintaining the hemoglobin levels at 12–14 g/dL during RT. The application of epoetin-α significantly improved LC, and a trend was observed for OS. Cancer 2005. © 2005 American Cancer Society.

Radiobiologic data suggest that tumor hypoxia is associated with an increased resistance to radiation-induced tumor cell kill, which is the consequence of less radiation-induced production of cytotoxic free radicals and less fixation of DNA damage.1 Thus, tumor hypoxia results in a worse prognosis. Tumor oxygenation can be affected by several factors, such as adequacy of blood supply, microcirculation, and oxygen-carrying capacity of the blood, which is represented by the hemoglobin level. Anemia (hemoglobin < 12 g/dL) is not uncommon in cancer patients and occurs in up to 80% of the those undergoing radiotherapy (RT)2

Tumor oxygenation can be confounded not only by anemia but also by a relatively high hemoglobin level. The latter is associated with a drop in nutritive perfusion after an increase in viscous resistance to flow. According to the literature, hemoglobin levels of 12–14 g/dL are considered optimal for tumor oxygenation.3 Thus, the patient's prognosis may be improved by maintaining hemoglobin levels during RT within this optimal range.

A decrease in the hemoglobin level during RT can be improved by the administration of recombinant human erythropoietin.2 Whether the effect of erythropoietin on the hemoglobin level is associated with an improvement of tumor control and survival currently is a matter of strong debate. Some authors have suggested a beneficial effect for erythropoietin with respect to outcome4–6; whereas, in a recently published prospective study from Germany, other authors suggested a worse outcome for patients who receive erythropoietin during RT.7 However, that study has been criticized heavily because of major methodological problems that confounded the results. For instance, the hemoglobin levels in the placebo group were more favorable with respect to tumor oxygenation than the hemoglobin levels in the erythropoietin group.

Thus, the potential effect of erythropoietin on prognosis in the treatment of patients with malignant disease remains to be clarified. In the current prospective study, we evaluated the impact of epoetin-α (ERYPO 10000; Janssen-Cilag, Neuss, Germany) both on hemoglobin levels during RT and on patient outcomes (overall survival [OS], metastatic-free survival [MFS], and local control [LC]) in patients with UICC Stage III esophageal carcinoma.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Sixty patients who received treatment either with RT or with RT plus chemotherapy for UICC Stage III esophageal carcinoma between March 2001 and September 2004, were included in this prospective, nonrandomized study. Thirty patients had received recombinant human erythropoietin (epoetin-α) after informed consent was obtained to avoid hemoglobin levels < 12 g/dL during RT (Group A). These patients received the entire therapy, including chemotherapy, in our department. Thirty patients did not receive epoetin-α (Group B), including either patients who received chemotherapy in other departments and institutions or patients who refused administration of epoetin-α.

Blood transfusions were given to patients with hemoglobin levels < 9 g/dL. Five patients in Group A and seven patients in Group B received from two to four transfusions during RT.

Epoetin-α

Epoetin-α in isotonic NaCl solution was administered subcutaneously (150 IU/kg) 3 times per week. Epoetin-α treatment was initiated after the hemoglobin level became < 13 g/dL, and epoetin-α administration was stopped after the hemoglobin level reached 14 g/dL again. Iron was substituted orally (200–300 mg per day) if the ferritin level was < 100 ng/mL or if ferritin saturation was < 20%.

Radiotherapy

RT was delivered with 6–16 megavolt (MV) photons with a daily dose of 1.8 grays (Gy) or 2.0 Gy 5 days per week. The total radiation dose was 45.0–50.4 Gy in 22 patients and 59.4–66.0 Gy in 28 patients. A dose of 45.0–50.4 Gy was applied either in combination with surgery or alone as palliative treatment. The initial RT fields (up to a dose of 50.0–50.4 Gy) had superior and inferior margins of 5 cm beyond the primary macroscopic tumor volume. The lateral, anterior, and posterior margins were a minimum of 2 cm beyond the primary macroscopic tumor volume. Regional lymph nodes were included. After irradiation of the initial fields, a boost dose (9–16 Gy) was delivered to the primary tumor with 2-cm margins and to enlarged lymph nodes with a minimum margin of 1 cm.

Chemotherapy

5-Fluorouracil (1000 mg/m2 per day) was administered as a continuous infusion for 120 hours (Days 1–5 of each course) in Weeks 1, 5, 9, and 13. Cisplatin at 75 mg/m2 was administered as an intravenous bolus over 60 minutes on Day 1 of each course. If from two to four courses of chemotherapy were given, then two courses were administered concurrently with RT.

Surgery

Patients with squamous cell carcinoma underwent radical en-bloc resection of the esophagus and two-field lymphadenectomy. Patients with adenocarcinoma underwent a transhiatal esophagectomy. Esophageal continuity was restored by gastric tube. Preferentially, the route taken was through the posterior mediastinum.

Potential Prognostic Factors

The following potential prognostic factors were investigated with respect to OS, MFS, and LC: patient age (≤ 60 years vs. > 60 years), gender, Eastern Cooperative Oncology Group (ECOG) performance status (1 vs. 2–3), location of tumor (upper one-third vs. middle one-third vs. lower one-third), tumor length (< 7 cm vs. ≥ 7 cm; determined on endoscopy), histology (squamous cell carcinoma vs. adenocarcinoma), histologic grade (Grades 1–2 vs. Grade 3), T classification (T3 vs. T4), lymph node status (N0 vs. N+), chemotherapy (0–1 courses vs. 2–4 courses), treatment approach (45.0–50.4 Gy plus surgery vs. 45.0–50.4 Gy alone vs. 59.5–66.0 Gy alone), hemoglobin level prior to RT (< 12 g/dL vs. ≥ 12 g/dL), and application of epoetin-α during RT (Group A vs. Group B).

Effect of Epoetin-α

Groups A and B also were compared in terms of the proportion of patients with at least 60% of hemoglobin levels during RT that were within the optimal range of 12–14 g/dL. This endpoint was chosen according to the data presented in Figure 1 and according to data from the literature.3 In addition, the weekly hemoglobin changes (increase or decrease) during RT were measured for both groups. The latter evaluation was performed for the total number of weeks of RT, for weeks during which concurrent chemotherapy was administered, and for weeks without chemotherapy.

thumbnail image

Figure 1. Overall recurrence rates were related to the proportion of patients who had hemoglobin (hgb) levels within the optimal range of 12–14 g/dL during radiotherapy: According to the literature, these levels are considered optimal for tumor oxygenation (see Vaupel et al., 20023).

Download figure to PowerPoint

Statistical Analysis

LC was defined as the absence of local disease progression demonstrated on endoscopy studies and/or computed tomography scans, which were performed at regular intervals. OS, MFS, and LC were calculated for each potential prognostic factor using the Kaplan–Meier method8 and were measured from the first day of RT or, in patients who received preoperative RT, from the day of surgery. Differences between the Kaplan–Meier curves were evaluated with the log-rank test. Results were considered significant at P values < 0.05.

Potential prognostic factors that were identified as significant in univariate analysis were evaluated in a multivariate analysis. The multivariate analysis was performed with a Cox proportional-hazards model.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Groups A and B were balanced well for the investigated potential prognostic factors (Table 1). The results of the univariate analysis with respect to OS, MFS, and LC at 1 year related to the potential prognostic factors are summarized in Table 2. The ECOG performance status achieved significance for LC and OS, the number of chemotherapy courses achieved significance for MFS and OS, and the hemoglobin level prior to RT achieved significance for LC, MFS, and OS.

Table 1. Comparison between Patients who Received Epoetin-α during Radiotherapy (Group A) and Patients who Did Not Receive Epoetin-α (Group B) with Respect to Potential Prognostic Factors for Outcome
Potential prognostic factorNo. of patientsP value
Group A (n = 30)Group B (n = 30)
  1. ECOG: Eastern Cooperative Oncology Group; SCC: squamous cell carcinoma; CT: chemotherapy; Gy: grays; RT: radiotherapy.

Age > 60 yrs16130.72
Female gender680.82
ECOG performance status = 112150.72
Tumor location   
 Upper one-third15150.96
 Lower one-third43 
Tumor length < 7 cm14130.98
Histology   
 SCC25280.57
 Adenocarcinoma52 
Histologic Grade 313160.72
Tumor classification   
 T314110.71
 T41619 
Positive lymph node status26261.0
Two to four CT courses21230.84
Treatment approach   
 45.0–50.4 Gy plus resection960.75
 45.0–50.4 Gy alone43 
 59.4–66.0 Gy alone1721 
Hemoglobin prior to RT   
 < 12 g/dL11130.85
 > 14 g/dL53 
Table 2. Univariate Analysis of Overall Survival, Metastasis-Free Survival, and Local Control at 1 Year Related to Potential Prognostic Factors and Administration of Epoetin-α
Potential prognostic factorLCMFSOS
%P value%P value%P value
  1. LC: local control; MFS; metastasis-free survival; OS: overall survival; ECOG: Eastern Cooperative Oncology Group; SCC: squamous cell carcinoma; Gy: grays; RT: radiotherapy.

Age      
 ≤ 60 yrs43 26 37 
 > 60 yrs590.55520.35520.44
Gender      
 Female41 35 31 
 Male530.77420.68490.38
ECOG performance status      
 165 38 59 
 2–3360.05320.40310.044
Tumor location      
 Upper one-third50 36 34 
 Middle one-third54 42 44 
 Lower one-third570.82630.64470.77
Tumor length      
 < 7 cm54 51 52 
 ≥ 7 cm480.82320.10380.21
Histology      
 SCC49 43 45 
 Adenocarcinoma750.19220.61440.95
Histologic grade      
 Grade 1–255 47 46 
 Grade 3510.72330.89440.73
Tumor classification      
 T358 58 54 
 T4470.59300.15380.39
Lymph node status      
 Negative (N0)55 43 25 
 Positive (N+)510.63330.54480.12
No. of chemotherapy courses      
 0–148 30 25 
 2–4540.42450.05520.044
Treatment approach      
 45.0–50.4 Gy plus resection52 57 52 
 45.0–50.4 Gy alone0 0 14 
 59.4–66.0 Gy alone550.54420.10480.11
Hemoglobin prior to RT      
 < 12 g/dL39 30 41 
 12–14 g/dL75 66 62 
 > 14 g/dL00.00900.002130.001

The univariate analysis also demonstrated that the application of epoetin-α was associated significantly with LC (Fig. 2) but not with MFS (Fig. 3). For OS, a trend was observed in favor of epoetin α (Fig. 4).

thumbnail image

Figure 2. Local control is illustrated for patients who received epoetin-α (n = 30 patients) versus patients who did not receive epoetin-α (n = 30 patients).

Download figure to PowerPoint

thumbnail image

Figure 3. Metastatic-free survival is illustrated for patients who received epoetin-α (n = 30 patients) versus patients who did not receive epoetin-α (n = 30 patients).

Download figure to PowerPoint

thumbnail image

Figure 4. Overall survival is illustrated for patients who received epoetin-α (n = 30 patients) versus patients who did not receive epoetin-α (n = 30 patients).

Download figure to PowerPoint

The potential prognostic factors that achieved significance in the univariate analysis were included in the multivariate analysis. The administration of epoetin α (P = 0.007) and ECOG performance status (P = 0.027) maintained significance for LC, whereas the hemoglobin level prior to RT lost significance for LC (P = 0.55). No potential prognostic factor maintained significance for MFS or for OS. At the least, a strong trend was demonstrated for the number of chemotherapy courses with respect to MFS (P = 0.07).

In 20 of 30 patients (67%) from Group A and in 3of 30 patients (10%) from Group B, ≥ 60% of hemoglobin levels during RT were within the optimal range of 12–14 g/dL (P = 0.003). Comparisons of both groups with respect to median changes in hemoglobin per week, for the total number of weeks of RT, for the weeks with concurrent chemotherapy, and for the weeks without chemotherapy are demonstrated in Table 3. No epoetin-α-related adverse effects were observed.

Table 3. Median Changes (Decrease or Increase) in Hemoglobin Levels for the Total Number of Weeks of Radiotherapy, for the Weeks with Concurrent Chemotherapy and for the Weeks without Chemotherapy
MeasureChange in hemoglobin level per week (g/dL)
Group A (with epoetin-α)Group B (without epoetin-α)
MedianRangeMedianRange
  1. RT: radiotherapy; CT: chemotherapy.

Total weeks of RT+0.4−0.9 to +1.2−0.4−1.5 to +0.7
Weeks with concurrent CT−0.30.9 to +0.6−0.7−1.5 to +0.2
Weeks without concurrent CT+0.5−0.6 to +1.2−0.3−1.2 to +0.7

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Erythropoietin is a cytokine that has shown the ability to increase hemoglobin levels by increasing erythropoiesis.9 In the current series, the course of the weekly hemoglobin levels was much more favorable in patients who received epoetin-α than in the other patients, either during administration of RT with concurrent chemotherapy or during RT alone (Table 3). The median weekly increase in hemoglobin levels during RT in our series was consistent with data from the literature10–12 that describe an average increase of 0.4–0.7 g/dL, depending on the erythropoietin dose.

The hemoglobin level represents the oxygen-carrying capacity of the blood, which is associated with tumor oxygenation. It has been suggested that tumor oxygenation is important for the effect of RT, because tumor hypoxia leads to an increased resistance to radiation-induced tumor cell kill.1

A hemoglobin level of 12–14 g/dL is considered optimal for tumor oxygenation.3 At hemoglobin levels < 12 g/dL, a worsening of the tumor oxygenation can be explained by the reduced oxygen-carrying capacity of the blood, which has been demonstrated in head and neck carcinoma.13 At hemoglobin levels > 14 g/dL, the worsening of the tumor oxygenation status occurs most likely due to a drop in nutritive perfusion after a drastic increase in viscous resistance to flow.

A potential impact of the hemoglobin levels during RT on outcome has been described for gynecologic tumors in two retrospective analyses.14, 15 Thus, it appears that it is important to maintain hemoglobin levels during RT within the optimal range of 12–14 g/dL to improve treatment outcome, which may be achieved by the administration of epoetin-α. According to our current results, holding the administration of epoetin-α should be considered if the hemoglobin level reaches 14 g/dL.

However, the potential impact of epoetin-α on outcome is debated heavily. Several studies have administered erythropoietin during RT to improve the prognosis of patients with malignancies. One retrospective analysis4 and two randomized studies, which recently were published as abstracts,5, 6 suggested a benefit for erythropoietin with respect to LC.

Conversely, results from a randomized study in Germany suggested a worse outcome for patients who received erythropoietin during RT.7 However, that study was criticized heavily because of major methodological problems, which confounded the results. The investigated groups were not balanced for relevant prognostic factors, approximately one-third of the patients were not treated per protocol, and differences were no longer significant when the analysis was limited to patients who were treated according to the protocol. A major point of criticism was that the hemoglobin levels were more favorable in the placebo group than in the treatment group (mean hemoglobin values of 12.4 g/dL vs. 14.8 g/dL at 4 weeks and of 12.9 g/dL vs. 15.4 g/dL at 9 weeks), which was explained by over-treatment with erythropoietin.

In contrast to the German study published by Henke et al.,7 we stopped the administration of epoetin-α if a hemoglobin level of 14 g/dL was reached. In the majority of our patients, the treatment with epoetin-α resulted in optimal hemoglobin levels of 12–14 g/dL. Keeping the hemoglobin level within this range was associated with significantly better LC and with a trend toward better OS; however, MFS was not improved significantly. These findings support the hypothesis that anemia results in tumor hypoxia and a decreasing tumor radiosensitivity. However, it may be questioned whether our results, which were found in patients with UICC Stage III esophageal carcinoma, are consistent with epoetin α administration during RT of tumors other than Stage III esophageal carcinoma. Thus, more prospective studies are required.

In the current series, the two patient groups were distributed evenly with respect to all potential prognostic factors, so that a relevant selection bias could be excluded. This even distribution strengthens the results of the current study, although it was not randomized. However, prospective observational studies can provide the same level of internal validity as randomized, controlled trials.16, 17 Furthermore, randomized trials may produce inconsistent results and can have limited external validity if there are design problems.

Erythropoietin receptors have been demonstrated in tumor cell lines in vitro.18 However, in most of these cell lines, the receptors do not have a functional role, because there is no effective downstream signaling, although increased proliferation of tumor cells has been described.19, 20 Erythropoietin receptors also have been described in in vivo models and in human tissue samples.21, 22 In animal models, the application of erythropoietin has not led to increased tumor growth.23, 24 Thus, the role and the clinical relevance of erythropoietin receptors remain to be clarified.

The administration of epoetin-α increases hemoglobin during RT and is effective in maintaining the hemoglobin levels within the range considered optimal for tumor oxygenation and radiosensitivity (12–14 g/dL). Treatment with epoetin-α significantly improves LC in patients with UICC Stage III esophageal carcinoma. Furthermore, there is a trend toward better survival. Over-treatment with epoetin-α has to be avoided, and the application should be stopped if the hemoglobin level reaches 14 g/dL.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Vaupel P, Thews O, Hockel M. Treatment resistance of solid tumors: role of hypoxia and anemia. Med Oncol. 2001; 18: 243259.
  • 2
    Harrison LB, Shasha D, Homel P. Prevalence of anemia in cancer patients undergoing radiotherapy: prognostic significance and treatment. Oncology. 2002; 63(Suppl 2): 1118.
  • 3
    Vaupel P, Thews O, Mayer A, et al. Oxygenation status of gynecologic tumors: what is the optimal hemoglobin level? Strahlenther Onkol. 2002; 178: 727731.
  • 4
    Glaser CM, Millesi W, Kornek GV, et al. Impact of hemoglobin level and use of recombinant erythropoietin on efficacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx. Int J Radiat Oncol Biol Phys. 2001; 50: 705715.
  • 5
    Blohmer JU, Wurschmidt F, Petry U, et al. Sixth interim analysis of a prospective, randomized, open and controlled AGO- and NOGGO-intergroup study: sequential adjuvant chemo-radiotherapy with versus without Epoetin alfa for patients with high-risk cervical cancer [abstract]. Proc Am Soc Clin Oncol. 2003; 22: 447.
  • 6
    Antonadou D, Cardamakis E, Sarris G, Tzigounis V, Throuvalas N. Effect of the administration of recombinant human erythropoietin in patients with pelvic malignancies during radiotherapy [abstract]. Radiother Oncol. 1998; 48(Suppl 1): S122.
  • 7
    Henke M, Laszig R, Rube C, et al. Erythropoietin to treat head and neck cancer patients with anemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet. 2003; 362: 12551260.
  • 8
    Kaplan EL, Meier P. Non parametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53: 457481.
  • 9
    Goodnough LT, Skikne B, Brugnara C. Erythropoietin, iron, and erythropoiesis. Blood. 2000; 96: 823833.
  • 10
    Lavey RS, Dempsey WH. Erythropoietin increases hemoglobin in cancer patients during radiotherapy. Int J Radiat Oncol Biol Phys. 1993; 27: 11471152.
  • 11
    Dusenbery KE, McGuie WA, Holt PJ, et al. Erythropoietin increases hemoglobin during radiotherapy for cervical cancer. Int J Radiat Oncol Biol Phys. 1994; 29: 10791080.
  • 12
    Sweeney PJ, Nicolae D, Ignacio L, et al. Effect of subcutaneous recombinant erythropoietin in cancer patients receiving radiotherapy: final report of a randomized, open-labeled Phase II trial. Br J Cancer. 1998; 77: 19962002.
  • 13
    Becker A, Stadler P, Lavay RS, et al. Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys. 2000; 46: 459466.
  • 14
    Grogan M, Thomas GM, Melamed I, et al. The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer. 1999; 86: 15281536.
  • 15
    Munstedt K, Volzing M, Von Georgi R. Hemoglobin levels during radiotherapy and their influence on local control and survival of patients with endometrial carcinoma. Oncol Rep. 2004; 11: 711717.
  • 16
    Rabeneck L, Viscoli CM, Horwitz RI. Problems in the conduct and analysis of randomized clinical trials: are we getting the right answers to the wrong questions? Arch Intern Med. 1992; 152: 507512.
  • 17
    Concato J, Shah N, Horwitz RI. Randomized controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000; 342: 18871892.
  • 18
    Westphal G, Niederberger L, Blum C, et al. Erythropoietin and G-CSF receptors in human tumor cells: expression and aspects regarding functionality. Tumori. 2002; 88: 150159.
  • 19
    Westenfelder C, Baranowski RL. Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int. 2000; 58: 647657.
  • 20
    Selzer E, Wachek V, Kodym R, et al. Erythropoietin receptor expression in human melanoma cells. Melanoma Res. 2000; 179: 527534.
  • 21
    Acs G, Zhang PJ, McGrath CM, et al. Hypoxia-inducible erythropoietin signaling in squamous dysplasia and squamous cell carcinoma of the uterine cervix and its potential role in cervical carcinogenesis and tumor progression. Am J Pathol. 2003; 162: 17891806.
  • 22
    Batra S, Perelman N, Luck LR, Shimada H, Malik P. Pediatric tumor cells express erythropoietin and a functional erythropoietin receptor that promotes angiogenesis and tumor cell survival. Lab Invest. 2003; 83: 14771487.
  • 23
    Stuben G, Thews O, Pottgen C, et al. Recombinant human erythropoietin increases the radiosensitivity of xenografted human tumours in anaemic nude mice. J Cancer Res Clin Oncol. 2001; 127: 346350.
  • 24
    Thews O, Kelleher DK, Vaupel P. Erythropoietin restores the anemia-induced reduction in cyclophosphamide cytotoxicity in rat tumors. Cancer Res. 2001; 61: 13581361.