J Cancer Res Clin Oncol (1992) 118:61-66 Cancer mesearch Clinical �9 �9 Springer-Verlag 1992

Phase I study of doxorubicin, ICRF-187 and granulocyte/macrophage-colony-stimulating factor Christina Walsh, Ronald H. Blum, Ruth Oratz, Alec Goldenberg, Andrea Downey, and James L. Speyer

Kaplan Cancer Center, New York University Medical Center, New York, USA

Received 5 July 1991/Accepted 14 August 1991

Summary. A group of 16 patients with advanced malig- nancy were entered on a phase I trial of escalating doses of doxorubicin with ICRF-187 for cardioprotection and granulocyte/macrophage-colony-stimulating factor (GMCSF) for bone marrow protection. Patients received intravenous ICRF-187 (dose ratio 20:1 ICRF-187:dox- orubicin) 30 min prior to doxorubicin. GMCSF at a dose of 15 ~tg kg -1 day-1 was self-administered subcuta- neously on days 3-14 of the cycle. Doxorubicin was ad- ministered every 21 days. Substantial hematological and non-hematological toxicity was seen. Fever, malaise, and pulmonary symptoms, thought to be due to GMCSF, were not eliminated by reduction in the GMCSF dose to 10 or 5 ~tg kg- 1 day- 1. Severe hematological toxicity was seen despite GMCSF administration and it was not pos- sible to escalate the doxorubicin dose above 72 mg/m 2 with this combination. Dose escalation of doxorubicin may be more feasible with the use of other growth factors or growth factor combinations.

Key words: Doxorubicin - GMCSF - Dose escalation

Introduction

Concepts of both dose response and dose intensity are major factors in determining the design of current clinical trials of chemotherapy. The theory of dose responsive- ness predicts that the greater the dose of chemotherapy administered the greater will be the response. As calcu- lated by Hryniuk (in mg m -2 week-l), dose intensity takes account of not only the total dose of chemotherapy

Offprint requests to: C. Walsh, NYU Medical Center, Division of Oncology, Old Bellevue C&D Bldg, Room 556, 462 First Avenue, New York, NY 10016, USA Abbreviations: GMCSF, granulocyte/macrophage-colony-stimulat- ing factor; ICRF-187, (+_)-l,2-bis(3,5-dioxopiperazinyl-l-yl)- propane; CMF, cyclophosphamide/methotrexate/5-fluorouracil; FDC, 5-fluorouracil/doxorubicin/cyclophosphamide; ANC, abso- lute neutruophil count

but also the time span over which it is delivered (Hryniuk et al. 1987). In breast cancer treatment data exist to sup- port the importance of both dose response and dose in- tensity, but most of the studies are retrospective. Hryniuk has retrospectively analyzed the relationship between dose intensity, response, and survival in the treatment of advanced breast cancer with cyclophosphamide/metho- trexate/5-fluorouracil (CMF). Within the bounds of his assumptions, this analysis showed that the received dose intensity correlated with survival (Hryniuk et al. 1987). He found a similar relationship between dose intensity and 3-year disease-free survival in the adjuvant treatment of stage II breast cancer (Hryniuk et al. 1987). Received dose intensity was also the major factor correlating with outcome in Bonadonna's retrospective analysis of his ad- juvant breast cancer treatment data (Bonnadona and Valagussa 1981). Jones et al. have reported higher re- sponse rates with increasing doses of doxorubicin in the treatment of advanced breast cancer (Jones et al. 1987). Tannock et al. (1988) performed a prospective trial of standard CMF vs the same drugs at 50% lower doses. This study showed improved response and survival for the standard dose regimen. However, prospective ran- domized trials showing the benefit of high-dose vs stan- dard-dose regimens are lacking.

Doxorubicin is the most active agent in the treatment of breast cancer. Drug-induced myelosuppression and cumulative dose-related cardiomyopathy create impedi- ments to increasing either the dose or dose intensity of doxorubicin delivery. In an attempt to develop a dose-in- tense regimen of doxorubicin that would have acceptable toxicity we designed a phase I study of escalating doses of doxorubicin given with granulocyte/macrophage- colony-stimulating factor (GMCSF) for bone marrow protection and (_+)-l,2-bis(3,5-dioxopiperazinyl-l-yl)- propane (ICRF-187) for cardioprotection.

ICRF-187 was added for cardioprotection on the ba- sis of results of our prior randomized trial of 5- fluorouracil, doxorubicin, and cyclophosphamide (FDC) vs FDC with ICRF-187. That study showed that treat- ment with ICRF-187 at a 20:1 ICRF-187/doxorubicin ra-

62

tio reduced cardiac toxicity as measured both by clinical criteria and by laboratory measurement of changes in the left-ventricular ejection fraction (Speyer et al. 1988). ICRF-187 is a bisdioxopiperazine compound that is hy- drolyzed intracellularly to a bidentate chelator similar in structure to EDTA. Preclinical study in several animal models showed that ICRF-187 could confer significant protection against anthracycline toxicity (Herman and Ferrans 1981; Herman et al. 1974, 1981). The mechanism by which ICRF-187 reduces cardiac toxicity of anthracy- clines is postulated to be chelation of iron at intracellular sites preventing the formation of the Fe-Adriamycin complex. This complex is necessary for generation of free radicals that cause damage to ceils (Filipi et al. 1987; Doroshow et al. 1980). For the current study we elected to continue to use the 20:1 ICRF-187 : doxorubicin ratio that had been effective for cardioprotection in our prior trial.

The addition of GMCSF for bone marrow support was based on studies showing that use of granulocyte- colony-stimulating factor (GCSF) or GMCSF could re- duce the number of neutropenic days after aggressive chemotherapy and could also permit the delivery of more chemotherapy doses without reductions or delays. Ant- man et al. showed that use of GMCSF with the MAID regimen reduced the mean number of days of absolute neutropenia (absolute neutrophil count, ANC<500/ mm 3) in sarcoma patients (Antman et al. 1988). Gabri- love et al. found that GCSF in combination with MVAC chemotherapy significantly reduced the number of days of ANC<1000/mm 3 and significantly increased the number of patients able to receive scheduled day-14 che- motherapy (Gabrilove et al. 1988). The goal of the study was to develop a regimen with a maximal dose intensity of doxorubicin that could be tested in a randomized trial vs standard-dose doxorubicin.

Table 1. Treatment levels

Level Drug" Indended Actual dose dose

1 Doxorubicin (mg/m 2) day 1 50 50 ICRF-187 (mg/m 2) day 1 1000 1000 GMCSF (Ixg/kg) s.c. on days 3-14 15 15 Doxorubicin (mg/m 2) day 1 72 72 ICRF-187 (mg/m 2) day 1 1440 1440 GMCSF (p.g/kg) s.c. on days 3-14 15 15

Doxorubicin (mg/m 2) day i 86 60 ICRF-187 (mg/m 2) day 1 1720 1220 GMCSF (ttg/kg) s.c. on days 3-14 15 15 Doxorubicin (mg/m 2) day 1 105 50 ICRF-187 (mg/m 2) day 1 2100 1000 GMCSF (Ixg/kg) s.c. on days 3-14 15 5

b.i.d, b

50 1000

5

Doxorubicin (mg/m z) day 1 ICRF-187 (mg/m 2) day 1 GMCSF (~tg/kg) s.c. on days 3-14

a Doxorubicin and ICRF-187 were administered intravenously every 21 days. GMCSF, granulocyte/macropbage-colony-stimulat- ing factor

b.i.d., twice daily

As will be discussed in Results, the dose of GMCSF was reduced for the finale 6 patients in the study. Patients continued in the study un- til they developed progressive disease or unacceptable toxicity, or re- fused further therapy.

Four patients were initially assigned to each dosage level of dox- orubicin. If no dose-limiting toxicity was observed in 3 weeks, dose escalation occurred for the next group. If dose-limiting toxicity oc- curred in 1 patient, 2 additional patients were added to the dose level. Dose-limiting toxicity was defined as any of the following: ANC < 500/mm 3 for 5 days or more, platelet count < 20 000/mm 3, systemic infection with neutropenia, clinical congestive heart fail- ure, decline in left-ventricular ejection fraction of 20% from baseline or decline to below 45%, other ECOG toxicity of grade 3 or 4 ex- cepting nausea, vomiting, or alopecia. The maximum tolerated dose was defined as that level at which 2 patients experienced dose-limit- ing toxicity.

Materials and methods

Eligibility. For entry to this study patients were required to have his- tological proof of advanced non-hematological malignancy, a per- formance status of 0, 1, or 2 by Eastern Cooperative Oncology Groupe (ECOG) criteria, a resting left-ventricular ejection fraction of at least 45% (lower limit of normal at our institution), a white blood count > 4 000/mm 3, platelet count > 100 000/ram 3, bilirubin < 1.5 mg/dl, creatinine < 1.5 mg/dl, and a 4-week interval from prior chemotherapy or radiotherapy. Patients were not eligible if they had received any prior anthracycline or if they had received more than 30 Gy to 50% of the pelvic bone structure and lower spine. The protocol and the consent signed by all patients were ap- proved by the Institutional Board or Research Associates, New York University Medical Center.

Study design. This was a phase I study in which the doses of doxoru- bicin and ICRF-187 were to be escalated for each group of patients. The GMCSF dose was to be the same for all patients. ICRF-187 and GMCSF (Schering-Sandoz) were provided by the National Cancer Institute. The dose ratio of ICRF-187 to doxorubicin was kept con- stant at 20:1. Table 1 shows the study scheme. Doxorubicin and ICRF-187 were given intravenously every 21 days. ICRF-187 was given over 15 rain starting 30 min prior to doxorubicin. GMCSF was self-administered subcutaneously on days 3-14 of each chemo- therapy cycle. Initially the dose of GMCSF was 15 ~tg kg- 1 day- 1.

Study evaluations. Pre-study evaluation included history and physi- cal, complete blood count, electrolytes, liver functions, renal func- tions, chest X-ray, electrocardiogram, urine analysis, tumor mea- surement, and resting gated radionuctide cardiac scan (MUGA). During the study, a complete blood count with differential was ob- tained every other day between days 3 and 14 of the cycle or until ANC reached 1000/mm 3. Physical examination and blood chemis- tries were repeated every 3 weeks. Scans and other tests required for tumor measurement were repeated every 2 cycles. The M U G A scan, electrocardiogram and chest X-ray were to be repeated at 300 rag/ m 2, 450 mg/m 2, and every 100 mg/m 2 thereafter of the cumulative doxorubicin dose.

Dose modifications. If on the day of scheduled treatment ANC was below 1500/mm 3 treatment was delayed, and the ANC checked weekly until a level of 1500/mm ~ was reached. A patient whose ANC failed to return to 1500/mm 3 within 28 days was removed from study. There were no dose reductions of chemotherapy. If the white blood count exceeded 20000/mm 3 during GMCSF adminis- tration the drug was discontinued and restarted if the count declined to 7000/mm 3 during days 3-14 of the cycle.

Response criteria. Complete response was defined as the disappear- ance of all clinical evidence of tumor and symptoms on two occa- sions measured at least 1 month apart. Partial response was defined as at least a 50% reduction in the size of all measurable tumor areas

63

as measured by the product of the largest diameter and its perpen- dicular, or by a decrease of 50% or more in evaluable disease lasting at least 1 month, with no development of new lesions. Bone-only le- sions, showing healing on X-ray, were considered a partial response. Progressive disease was defined as the appearance of new lesions or a 25% increase in the size of any lesion.

Results

Patient population

A total of 16 patients were entered in this study. Their characteristics are shown in Table 2. Eight patients had been previously treated with chemotherapy for meta- static disease; 3 patients had received adjuvant chemo- therapy. Of the 11 patients with prior chemotherapy, 10 had received only one regimen, whether adjuvant or metastatic. A single patient had recieved multiple re- gimens, and 5 patients had not received prior chemother- apy. All but I patient had an ECOG performance status of 0 or 1. All patients were evaluated for toxicity and re- sponse.

Hematological toxicity and infection

Table 3 shows the nadir granulocyte and platelet counts seen in the first 2 cycles for all patients. The first 10 pa- tients received GMCSF at a dose of 15 gg kg -~ day -1. At a doxorubicin dose of 50 mg/m 2, five of the eight

Table 2. Patient characteristics

Characteristic a No. patients

Male 2 Female 14

Primary tumor type Breast 7 NSC lung 3 Ovary 2 Sarcoma 2 Endometrial 1 Unknown 1 ~ 1

Prior treatment Adjuvant chemotherapy

CMF 3 Metastatic chemotherapy

MVP 3 CEP 1 CDDP Ctx 1 5FU Leuc 1 Multiple 1 CMF 1

No chemotherapy 5 Prior radiotherapy 5

Performance status 0 8 1 7 2 1

" NSC, non-small-cell; CEP, cyclophosphamide, etoposide, pla- tinum; CDDP Ctx, platinum, cyclophosphamide; 5FU Leuc, 5-fluorouracil, leucovorin; MVP, mitomycin, vinblastine, platinum; CMF, cyclophosphamide, methotrexate, 5-fluorouracil

cycles resulted ifi granulocyte nadirs below 500/mm 3. The duration of these nadirs was 4 days or less in all cases. Of the 8 cycles, 2 resulted in neutropenic fever (defined as fever with ANC<1000/mm 3 requiring hospitalization for antibiotics) without a documented source of infec- tion.

The maximally tolerated dose level was doxorubicin 72 mg/m 2 with ICRF-187 1440 mg/m 2, and 3 patients were entered at this dose level. Two patients had dose- limiting hematological toxicity. One of these patients died with documented polymicrobial sepsis during neu- tropenia and the second had documented bacteremia with a granulocyte count below 500/ram 3 for 6 days. One patient at this dose level had a platelet nadir of 22 000/ mm 3 but no clinical bleeding.

For the next group of 3 patients the dose of doxorubi- cin was reduced to 60 mg/m 2 with 1200 mg ICRF. As can be seen in Table 3, 3 of 6 cycles showed granulocyte na- dirs of <500/mm 3 but no nadirs lasted longer than 4 days and no episodes of neutropenic fever occurred.

Though dose-limiting hematological toxicity did not occur in the group of patients receiving 60 mg/m 2 dox- orubicin, non-hematological toxicity was substantial in all patients receiving 15 gg/kg GMCSF daily (see Table 4 and below). In order to define better the role of GMCSF causing non-hematological toxicity in this combination the doses of both doxorubicin and GMCSF were reduced for the next patients. Table 3 shows the doses received and the nadirs encountered in the last 6 patients. Among this group of patients receiving 50 mg/m 2 doxorubicin and 5 or 10 ggkg -1 day -1 GMCSF, 3 of 8 cycles had na- dirs below 500/mm 3 but one lasted longer than 3 days. One patient had neutropenic fever and a documented bacterial bursitis.

Hematological toxicity did not appear to correlate with prior treatment. Three of five patients with no prior chemotherapy had granulocyte nadirs below 500/mm 3 in the first cycle.

Non-hematological toxicities

Table 4 shows the occurrence of non-hematological tox- icities of grade 2 or 3. These toxicities occurred with equal frequency at all dose levels of doxorubicin/ICRF- 187. Though it is not possible definitively to assign toxic- ity to one drug in the combination, many of these toxic- ities may have been due to GMCSF. Almost all patients had grade 2 or 3 fever and local skin reaction of mild to moderate severity at sites of GMCSF injection. Two pa- tients also developed a generalized maculopapular rash but it was possible to continue treatment in these cases. Also frequent was a syndrome of dyspnea and pleuritic chest pain, which occurred in patients without known pulmonary disease as well as in patients with pulmonary tumor. These symptoms were often transient but they were recurrent and were not limited to the first day of GMCSF dosing. They were not temporally related to the day's GMCSF dosing but they generally disappeared to- tally within 24~8 h after the last GMCSF dose of the cycle.

64

Table 3. Hematological toxicity and in- fection a Patient Dox dose GMCSF Cycles Gran Plt

no. (mg/m 2) dose received nadir 1 nadir 1 (lag kg- 1 day- 1)

Gran nadir 2

Plt nadir 2

1 50 15 2 350 110 2 50 15 2 528 105 3 50 15 2 252 b 117 4 50 15 5 135 156

5 72 15 2 14 b 42 6 72 15 1 57 b'~ 22 7 72 15 4 60 90

8 60 15 2 56 116 9 60 15 3 2800 107

10 60 15 2 35 128

11 50 5 b.i.d. 2 200 70 12 50 5 b.i.d. 1 1188 172 13 50 5 b.i.d. 1 440 132 14 50 5 b.i.d. 2 525 166

15 50 5 1 407 b 128 16 50 5 1 1008 72

1326 288 170 u

1050

280 b

140 b

736 3016 490

532

1004

98 54 60

182

41

87

42 125 74

49

105

Table 4. Number of patients with non- hematological toxicity of grade 2 or 3 severity a

a Dox, doxorubicin: Gran, granulocytes; Pit, platelets b Cycles with neutropenic fever requiring intravenous antibiotics ~ Duration of absolute neutrophil count < 500/mm 3 longer than 5 days

Treatment Fever Chills Ma- Ar- Skin N/V Muco Dia laise thral-

gias

Dox 50 mg/m 2 3 3 2 1 4 1 0 0 4 GMCSF 15 gg/kg n=4

Dox 60 mg/m 2 3 2 3 2 3 3 0 1 3 GMCSF 15 lag/kg n=3

Dox 72 mg/m 2 3 2 2 2 2 2 1 1 2 GMCSF 15 lag/kg n=3

Dox 50 mg/m 2 1 1 4 3 4 0 1 GMCSF 10 lag/kg (5 lag/kg b.i.d.) n=4

Dox 50 mg/m 2 1 0 1 0 0 1 2 1 GMCSF 5 lag/kg n=2

Pulm/ pleur

a N/V, nausea and vomiting; Muco, mucositis; Dia, diarrhea; Pulm/pleur, pulmonary pleuritic; Dox, doxorubicin

One pa t i en t wi th p u l m o n a r y metas tases exper ienced grade 4 p u l m o n a r y toxic i ty du r ing the second cycle o f t r ea tmen t a t a d o x o r u b i c i n dose level o f 50 m g / m 2. The pa t i en t exper ienced severe dyspnea on days 6 and 7 o f this cycle when the whi te b l o o d coun t was in the 7000- 8000/mm 3 range. She was t r ea ted wi th oxygen and diuret ics . S y m p t o m s resolved over 24 h and an eject ion f rac t ion o b t a i n e d the next day was unchanged . The fol- lowing day , when her s y m p t o m s h a d subsided, her whi te coun t was 19000 /mm 3. G M C S F was withheld . Af te r 2 days, when the white b l o o d coun t was be low 7000/mm 3, G M C S F was res ta r t ed wi thou t recurrence o f s y m p t o m s and the dos ing cycle was comple ted . Because no pa t i en t r eached 300 m g / m 2 comula t ive doxorub ic in , and no cl inical ca rd i ac toxic i ty was seen, this was on ly pa t i en t to have a second M U G A scan.

Because o f the f requency and severi ty o f side-effects tha t were felt to be due to G M C S F , the dose o f this agen t was r educed to 5 g g / k g / d a y for two final g roups o f pa - t ients. D o x o r u b i c i n was reduced to 50 m g / m 2 for b o t h groups . Tab le 4 shows the toxic i ty tha t was seen. Pa t i en t numbe r s are small , bu t i t is c lear t ha t cons t i tu t iona l s y m p t o m s a n d the s y n d r o m e o f p u l m o n a r y s y m p t o m s descr ibed above con t inued to occur.

Response data

N o pa r t i a l or comple t e responses occurred . The m e a n n u m b e r o f cycles received was 2. Twelve pa t ien ts wi th s table disease d i scon t inued t r ea tmen t because o f dose- l imi t ing toxic i ty or refusal to cont inue , while 4 pa t ien ts left the s tudy because o f progress ive disease.

65

Discussion

Toxicity continues to limit the success of attempts to in- crease dose intensity of doxorubicin administration, both alone and in combination. Hortobagyi et el. (1987) tested dose intensification of 5-fluorouracil, doxorubicin, and cyclophosphamide in patients with breast cancer, main- taining the patients in a protected hospital environment during the period of neutropenia. Hematological toxicity was very severe and doxorubicin could not be escalated to the intended dose because of gastrointestinal toxicity.

Jones et al. (1987) gave escalating doses of doxorubi- cin as a single agent on a daily x 3 schedule given monthly. They were not able to increase the median dose intensity of doxorubicin administered over standard dos- age schedules. Both hematological and cardiac toxicities were dose-limiting, with 16 of 26 patients removed from study for declining left-ventricular ejection fraction (Jones et al. 1987). They did report a high rate of re- sponse.

More recently Bronchud et al. (1989) gave escalating doses of doxorubicin every 14 days in combination with continous infusion of GCSF on days 2-11 of the cycle. Six patients received 125mg/m 2 every 14 days for 3 cycles and 2 patients received 150 mg/m 2 every 14 days for 3 cycles. Though this study demonstrated success with increasing the dose intensity of doxorubicin deliv- ery, problems remain. Out of 8 patients receiving the higher doses, 7 required hospitalization for neutropenic fever, and mucositis and desquamating dermatological toxicity were severe. In addition, only a limited number of doses were given, keeping the total amount below 450 mg/m z. If this type of regimen were continued over time the potential for cardiotoxicity would become a ma- jor issue.

The goal of our study was to establish a dose-intense regimen of doxorubicin that could be given in repeated cycles with limited hematological and cardiac toxicity. If a pattern of acceptable toxicity with bone marrow recov- ery by 14 days was seen with this regimen, our intention was to change the frequency of doxorubicin dosing to ev- ery 14 days. However, in our study we were not able to escalate the dose of doxorubicin above 72 mg/m 2 every 21 days, a standard dose; 2 of 3 patients at that level ex- perienced dose-limiting toxicity and 1 patient died.

Since this was not a randomized study, we do not know whether the hematological toxicity would have been more severe in the absence of GMCSF. However, though GMCSF may have shortened the duration of na- dirs, both the hematological and non-hematological tox- icity of this combination were unacceptable.

We elected to use a 20: 1 ICRF-187 : doxorubicin ratio because this ratio demonstrated a cardioprotective effect in our prior randomized study of FDC vs FDC plus ICRF-187 in patients with advanced breast cancer. In that study of 150 patients the mean white blood cell and platelet nadirs were not different in the two arms for the first 2 cycles (Speyer et al. 1988). At higher doses ICRF- 187 does cause myelosuppression, the maximum toler- ated dose being 3.8 g m-2 week-1 in pretreated patients (Vogel et al. 1987). Though this is much higher than the

ICRF-187 dose achieved in this study, it may be that ICRF-187 made a greater contribution to myelosup- pression than expected.

The toxicity experienced by patients in this study was substantial. Grade 2 or 3 fever was nearly universal and complicated the monitoring of these patients during neu- tropenia, sometimes necessitating difficult clinical judg- ments as to the etiology of the fever. In 33 cycles admin- istered, there were 7 hospitalizations for antibiotic ad- ministration for neutropenic fever. In 4 of these episodes bacterial infection was documented. Also very disturbing to patients was a syndrome of intermittent shortness of breath and/or pleuritic chest pain. These symptoms had sometimes subsided by the time patients were seen for evaluation, but no physical findings or chest X-ray or electrocardiographic evidence of pleuritis or pericarditis were ever documented. These symptoms were not corre- lated with the height of the white blood cell count. Though a "first-dose" syndrome including such symp- toms has been reported with GMCSF administration (Lieschke et al. 1989), in our patients they were not limited to the first day of treatment but recurred throughout the treatment cycle. Supporting their associ- ation with GMCSF, these symptoms always subsided within 24 h of the end of GMCSF administration.

When this study was designed it was not clear to what degree a dose/response relationship exists for hemato- poietic stimulation with GMCSF. We selected a rela- tively high dose of GMCSF in an attempt to maximize the chance of escalating doxorubicin successfully. Marked non hematological toxicity caused study discon- tinuation in 5 of the 10 patients receiving 15 ~tg kg -1 day-1 GMCSF. Since it was felt that many of these tox- icities were related to GMCSF, dose reduction of this agent was implemented for two final groups of patients. Though the numbers are small, no clear-cut decrease in toxicity was demonstrated. Among the 10 patients receiv- ing 15 gg kg -1 day -1 GMCSF, 7 of than discontinued therapy because of grade 3 or 4 toxicity or refusal to con- tinue therapy, and 7 received only 1 or 2 cycles. In the fi- nal group of 6 patients receiving either 5 gg/kg twice daily or 5 gg/kg each day, 5 left the study for either tox- icity or refusal and all 6 received either 1 or 2 cycles. In view of this toxicity we elected not to proceed with this regimen. We plan to attempt doxorubicin/ICRF-187 dose escalation using combinations of growth factors for bone marrow support.

References

1. Antman K, Griffin J, Elias Aet al. (1988) Effect of rhGMCSF on chemotherapy induced myelosuppression. N Engl J Med 319:593-598

2. Bonadonna G, Valagussa P (1981) Dose response effect of ad- juvant chemotherapy in breast cancer. N Engl J Med 304:10- 15

3. Bronchud MH, Howell A, Crowther D, Hopwood P, Soza L, Dexter TM (1989) The use of granulocyte colony-stimulating factor to increase the intensity of treatment with doxorubicin in patients with advanced breast and ovarian cancer. Br J Cancer 60:121-125

66

4. Doroshow J, Locker G, Myers C (1980) Enzymic defenses of the mouse heart against reactive oxygen metabolites: alterations produced by doxorubin. JCI 65:128-135

5. Filippi J, Imandi A, Wolgemuth R (1987) Characterization of the cardioprotective effect of ICRF 187 on anthracycline car- diotoxicity. In: Hacker M, Lazo J, Tritton T (eds) Organ di- rected toxicities of anticancer drugs. Nijhoff, Boston, p 225

6. Gabrilove J, Jakubowski A, Scher H et al. (1988) Effect of GCSF on neutropenia and associated morbidity due to che- motherapy for TTC of the urothelium. N Engl J Med 318:1414- 1422

7. Herman E, Mhatre R, Chadwick D (1974) Modification of some of the toxic effects of daunomycin (NSC-62161) by pre- treatment with the antineoplastic agent ICRF-159 (NSC- 129943). Toxicol Appl Pharmacol 27:517-526

8. Herman EH, Ferrans VJ, Jordan W, Adralan B (1981) Reduc- tion of chronic daunorubicin cardiotoxicity by ICRF-187 in rabbits. Res Commun Chem Pathol Pharmacol 31:85-97

9. Herman EH, Ferrans VJ (1981) Reducation of chronic doxoru- bicin cardiotoxicity in dogs by pretreatment with (+)-l ,2- bis(3,5-dioxopiperazinyl-l-yl) propane (ICRF-187). Cancer Res 41:3436-3440

10. Hortobagyi G, Bodey G, Buzdar A et al. (1987) Evaluation of high-dose versus standard dose FAC chemotherapy for ad-

vanced breast cancer in protected environment units: a prospec- tive randomized study. J Clin Oncol 5:354-364

11. Hryniuk W, Figueredo A, Goodyear M (1987) Applications of dose intensity to problems in chemotherapy of breast and col- orectal cancer. Semin Oncol 14[Suppl 4]:3-11

12. Jones R, Holland F, Bhardwaj S et al. (1987) A Phase I - I I study of intensive dose adriamycin for advanced breast cancer. J Clin Oncol 5:172-177

13. Lieschke G, Maher D, Cebon J e t al. (1989) Effects of bac- terially synthesized rh GMCSF in patients with advanced ma- lignancy. Ann Int Med 110:357-364

14. Speyer JL, Green MD, Kramer EL, Rey M, Sanger J, Ward C, Dubin N, Ferrans V, Stecy P, Zeleniuch-Jacquotte A, Wernz J, Felt F, Meyers M, Slater W, Taubes S, Blum R, Muggia F (1988) Protective effect of ICRF-187 against chronic doxorubi- cin induced cardiac toxicity in women with advances breast cancer. N Engl J Med 319:745-752

15. Tannock I, Boyd N, DeBouer Ge t al. (1988) A randomized trial of two dose levels of cyclophosphamide, methotrexate and fluorouracil chemotherapy for patients with metastatic breast cancer. J Clin Oncol 6:1377-1388

16. Vogel C, Gorowski E, Devila E et al. (1987) Phase I clinical trial and pharmacokinetics of weekly ICRF-187 infusion in patients with solid tumor. Drugs 5:187-198