Short Term Treatment with
Escheria coli Recombinant
Human Granulocyte-Macrophage–Colony Stimulating
Factor Prior to Chemotherapy for Hodgkin Disease
M. Aglietta,M.D.1 F. Montemurro,M.D.1 F. Fagioli,M.D.2 C. Volta,M.D.3 B. Botto,M.D.4 M. Cantonetti,M.D.5 V. Racanelli,M.D.6 L. Teofili,M.D.7 R. Ferrara,Ph.D.8 S. Amadori,M.D.5 G. L. Castoldi,M.D.2 F. Dammacco,M.D.6 A. Levis,M.D.4
1Divisione Universitaria di Oncologia ed Ematologia,
Ospedale Mauriziano Umberto I-Istituto per la Ricerca e la Cura del Cancro (I.R.C.C.), Torino, Italy.
2Cattedra di Ematologia, Universita´ di Ferrara,
Arcispedale S. Anna, Ferrara, Italy.
3Clinica Medica Generale, Universita´ di Novara,
Ospedale Maggiore della Carita´, Novara, Italy.
4Divisione di Ematologia, Ospedale S. Giovanni
Battista, Torino, Italy.
5Divisione di Ematologia, Ospedale S. Eugenio,
Roma, Italy.
6Dipartimento di Scienze Biomediche ed
Oncolo-gia Umana, Universita´ degli Studi di Bari-Poli-clinico, Bari, Italy.
7Divisione di Ematologia, Policlinico
Gemelli-Uni-versita´ Cattolica, Roma, Italy.
8Novartis Farma S.p.A., Origgio, Italy.
Supported by a Consiglio Nazionale delle Ricerche (CNR) special project grant from the Applicazioni Cliniche della Ricerca Oncologica (ACRO), by the Italian Association for Cancer Research, and by No-vartis Farma S.p.A., Origgio (Varese), Italy. The authors thank Dr. Silvano Andorno for help with collecting data and Ms. Barbara Bertin for secretarial work.
Address for reprints: M. Aglietta, M.D., Divisione Universitaria di Oncologia ed Ematologia, Ospedale Mauriziano Umberto I, Largo Turati 62, 10128, Torino, Italy.
Received July 22, 1999; accepted October 1, 1999.
BACKGROUND.Granulocyte-macrophage– colony stimulating factor (GM-CSF) ad-ministration stimulates the proliferation of hemopoietic progenitors. Shortly (48 –96 hours) after its discontinuation, feedback phenomena occur and the pro-genitor proliferation rate drops below baseline levels. As the quiescence of hyper-plastic bone marrow suggests that hemopoietic cells may be refractory to the toxic effects of cytostatic drugs, the decision was made to test the hypothesis that GM-CSF given before chemotherapy may be myeloprotective.
METHODS.Fifty-six patients with newly diagnosed Stage II–IV Hodgkin disease, ages 18 –77 years, were randomized to receive GM-CSF (5g/kg subcutaneously) or placebo from Day 7 to Day 4 before each chemotherapy administration (6 cycles of a hybrid of mechlorethamine, vincristine, procarbazine, and prednisone with doxorubicin, bleomycin, vinblastine, and dacarbazine). The treatment was consid-ered a success if the delivery rate of chemotherapy was⬎90% after 3 cycles and ⬎80% after 6 cycles.
RESULTS. Thirty patients received GM-CSF and 26 placebo. The dose intensity (85.2% vs. 79.6%) and the overall success in terms of delivery rate (56.7% vs. 50%) were higher in the GM-CSF group, although these differences were not statistically significant. The neutrophil nadirs were higher in the GM-CSF group during the first three cycles and subsequently similar in both groups.
CONCLUSIONS.No significant differences in terms of myelotoxicity or drug delivery were observed between the two treatment arms. Although the myeloprotective effect of the prechemotherapy administration of GM-CSF seems to be minimal, the data indicate a safe timing between GM-CSF discontinuation and further chemo-therapy. Because cumulative myelotoxicity has been observed with other growth factors, given in the interval between the chemotherapy cycles, this may be relevant to the planning of rapid cycling. Cancer 2000;88:454 – 60.
© 2000 American Cancer Society.
KEYWORDS: Hodgkin disease; hybrid mechlorethamine; vincristine, procarbazine, and prednisone; doxorubicin, bleomycin, vinblastine, and dacarbazine; granulocyte-macrophage– colony stimulating factor; myeloprotection.
M
ore than 80% of patients with advanced Hodgkin disease achieve complete remission when they are treated with alternating cyclic combination chemotherapy; however,⬇33% of them will experience disease recurrence.1– 4The monthly or rapid alternation ofmechlor-ethamine, vincristine, procarbazine, and prednisone (MOPP)5 with
doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD),6 – 8two
noncross-resistant and highly effective combinations for the treat-ment of patients with Hodgkin disease, represents the application of a rationale suggested by Goldie et al.,9who predicted that a higher © 2000 American Cancer Society
cure rate could be achieved with the earliest possible introduction and the most rapid alternation of all ac-tive agents in a therapeutic program.
When using the so-called “hybrid” MOPP/ABVD regimen, myelosuppression is a frequent cause of the drug delivery delays that lead to a suboptimal dose intensity.2,3In their original study, Viviani et al.3found
that, despite optimal drug delivery during the first 6 cycles of hybrid MOPP/ABVD, the median relative dose intensity (expressed as the doses received during the first 6 cycles and the actual days of treatment divided by the planned doses and duration of treat-ment) was 0.78. Retrospective analyses have indicated that dose intensity or the cumulative dose are impor-tant factors that may affect outcome for patients with malignancies.10 –15
Expectations have been raised by the introduction of colony stimulating factors, which may allow the maintenance of full dose chemotherapy and thereby increase the cure rate for patients with chemosensitive neoplasms.16,17Hematopoietic growth factors usually
are administered during the phase of early bone mar-row reconstitution after chemotherapy to enhance progenitor kinetic activity and to allow faster recovery.18
An assessment of the effects of granulocyte-mac-rophage– colony stimulating factor (GM-CSF) on pro-genitor cells has suggested more rational scheduling.19
The in vivo administration of GM-CSF causes an ap-proximately three-fold increase in the production of bone marrow cells as a result of the recruitment of resting cells and a shorter cell cycle.20,21After
discon-tinuation, the proliferation of bone marrow cells falls below baseline levels, probably as a result of feedback phenomena. The noncycling status of bone marrow progenitor cells after the discontinuation of GM-CSF possibly indicates a refractory response to the killing action of cytostatic drugs, and this, together with an increase in the myeloid reservoir, may lead to myelo-protection if GM-CSF is discontinued shortly before chemotherapy.22–24On the basis of this rationale, we
conducted a randomized, placebo-controlled study to assess whether the prechemotherapy administration of GM-CSF can reduce myelotoxicity in Hodgkin lym-phoma patients who receive the hybrid MOPP/ABVD regimen.
MATERIALS AND METHODS
Patients
Newly diagnosed patients with Hodgkin disease ages 18 – 80 years were considered eligible for the study. At presentation, they were clinically staged on the basis of a physical examination, complete blood cell, differ-ential and platelet counts, the erythrocyte sedimenta-tion rate, serum biochemistry, bone marrow biopsy,
standard chest X-rays, and computed tomography scans of the thorax and abdomen. The enrolled pa-tients gave their informed consent before they were assigned to treatment, and the protocol was approved by the Ethics Committees of the participating institu-tions.
Trial Design
This was a prospective, multicenter, double-blind, placebo-controlled trial in which each patient was randomized to receive GM-CSF or placebo before the starting of chemotherapy.
Chemotherapy
It was planned that all of the patients would receive 6 cycles of the MA/MA chemotherapy regimen, which consists of mechlorethamine-6 mg/m2 intravenously
(IV) on Day 1, vincristine 1.4 mg/m2 IV (maximum
dose, 2 mg) on Day 1, procarbazine 100 mg/m2orally
on Days 1–7, prednisone 40 mg/m2orally on Days 1–7,
doxorubicin 25 mg/m2 IV on Day 15, bleomycin 10
mg/m2IV on Day 15, vinblastine 6 mg/m2IV on Day
15, and dacarbazine 375 mg/m2 IV on Day 15. The
cycles were repeated at 28-day intervals. In the event of myelosuppression on the planned day of adminis-tration, the drug schedule was modified as follows: a reduction to 66% of the planned dose for each drug (except vincristine, prednisone, and bleomycin) if the neutrophil count was 0.5–1.0⫻ 109/L or if the platelet
count was 25–50⫻ 109/L. If the neutrophil count was
⬍0.5 ⫻ 109/L or if the platelet count⬍25 ⫻ 109/L, then
the chemotherapy was delayed for 1 week.
GM-CSF
The purified and lyophilized human recombinant E.
coli-derived GM-CSF was supplied by Novartis
Pharma Ltd. (CH 4002; Basel, Switzerland). The pa-tients who were allocated randomly to the GM-CSF arm received GM-CSF subcutaneously at a dose of 5 g/kg/day from Day 7 to Day 4 before the first cycle of chemotherapy and then on Days 8 –11 and Days 22–25 of each subsequent cycle (except Days 22–25 of the last cycle). In the placebo arm, the placebo was ad-ministered following the same schedule. The total study observation period covered the duration of the complete chemotherapy regimen.
Primary Efficacy End Point
The primary efficacy end point was adherence to the planned delivery rate of the cytostatic drugs: Success was defined as a delivery rate over Cycles 1–3 ofⱖ90% together with a delivery rate over Cycles 1– 6 ofⱖ80%. The delivery rate was calculated as the arithmetic mean of the percentages of the planned dose
intensi-ties of each of the drugs of the MA/MA protocol (ex-cluding prednisone) actually administered during each period. The dose intensity of each drug was cal-culated by dividing the total amount administered during each period (mg/m2) by the total number of
days considered. For Cycles 1–3, the number of days was counted from the first day of Cycle 1 to the day before starting Cycle 4; for Cycles 1– 6, the number of days was counted from the first day of Cycle 1 to the last day of administration of the second part of the MA/MA protocol in Cycle 6. Any patient who failed to complete the trial was considered a treatment failure. The planned dose intensity for each drug was calcu-lated as the total scheduled dose divided by 84 (3 28-day cycles) for Cycles 1–3 and by 155 (5 28-day cycles plus 15 days until the last administration of the sixth cycle) for Cycles 1– 6. In the case of vincristine, the protocol stated that the maximum administered dose had to be 2 mg regardless of body surface area (scheduled dose, 1.4 mg/m2); therefore, the
adminis-tered dose was considered full provided that it was 1.4 mg/m2 or 2 mg. When calculating delivery rates, a
dose of 2 mg was taken considered equivalent to a dose of 1.4 mg/m2.
Secondary End Points
The secondary efficacy end points were the nadir neu-trophil counts, the total duration of neutropenia, and the total number of days of antibiotic or antifungal treatment.
Tolerability and Safety
The safety analyses were conducted for the set of randomized patients who were treated with at least 1 dose of GM-CSF (the SAF set). Each adverse event was considered as a single event from the date of its onset to the date of its disappearance (or from the date of onset if it persisted), regardless of whether its duration involved one or more cycles.
Statistical Methods
The sample size of 56 patients was calculated to achieve an 80% power of detecting an increase in the success rate (i.e., maintenance of the target delivery rate after 3 cycles and 6 cycles) from 40% (placebo) to 80% (GM-CSF) using the Fisher one-sided exact test at a significance level of P ⫽ 5%. The randomized pa-tients who were treated with at least 1 dose of GM-CSF and who also provided information regarding the pri-mary variable were considered for the main efficacy analysis according to the intention to treat principle (the ITT set). The per protocol (PP) set of patients included all of the ITT patients who completed the study without any major protocol violation (defined as
delay of chemotherapy for ⬎ 15 days for nonclinical reasons).
The patients who discontinued the study before the end of the sixth cycle for any reason were classified as treatment failures with respect to the primary effi-cacy variable, which was evaluated by means of the Fisher one-tailed exact test at a significance level of
P⫽ 0.05. The results of the primary efficacy evaluation
were confirmed by repeating the analysis on the PP set of patients. Because the statistics reported for the sec-ondary efficacy variables were biased due to the in-complete records of the patients who discontinued the study prematurely, the results obtained from the ITT set of patients were rank transformed, with the worst rank assigned to the noncompleters. Statistics based on these ranks were then produced, and the Student t test for unpaired data was used to make between-group comparisons of the ranks of the quantitative variables.
The differences in nonhematologic adverse events between the two protocol arms were analyzed with the Fisher exact test at a significance level of P⫽ 0.05. The data were entered using version 3.4 CLIMED software (SIMED, Lyon, France) and were analyzed with SAS software (version 6.11 for Windows; SAS Institute, Cary, NC).
RESULTS
Patient Disposition and Outcome
Between September 1991 and December 1994, 56 pa-tients from eight centers were enrolled in the study: Thirty patients were randomized to receive GM-CSF, and 26 patients were randomized to receive placebo. All of the randomized patients met the inclusion cri-teria and were eligible for protocol treatment, and all were eligible for both the SAF and ITT data sets. Their characteristics are shown in Table 1. A total of 16 patients (8 patients in the GM-CSF arm and 8 patients in the placebo arm) did not complete either the study or the scheduled treatment period for the reasons listed in Table 2. Three patients died during the study: two in the placebo group (heart failure and cerebral apoplexy) and one in the GM-CSF group (septic shock). None of the deaths was found to be related to the study drug.
In the placebo arm, 1 patient was lost to follow-up, and another patient discontinued treatment be-fore the end of the study because complete remission was achieved; 3 patients were withdrawn from the study because of administrative problems, and 1 pa-tient was excluded from the PP set because a delay in chemotherapy⬎ 15 days for nonmedical reasons was considered a major protocol violation. The clinical
outcome of the two groups after 6 cycles of chemo-therapy is shown in Table 3.
Toxicity
Two patients in the GM-CSF group experienced seri-ous adverse events leading to the discontinuation of the study treatment: one transient ischemic attack with abdominal pain and emesis and one severe skin rash. Another 4 patients in the same group discontin-ued the treatment because of adverse events occurring before the 6 scheduled chemotherapy cycles had been completed (1 due to skin rash and chest pain, 1 due to skin rash alone, 1 due to pericarditis, and 1 due to hypotension), and another patient dropped out of the study because of an allergic reaction to procarbazine.
Nineteen of the 30 patients (63%) in the GM-CSF group experienced adverse events compared with 10 of the 26 patients (38%) in the placebo group (P ⫽ 0.06). The most frequent adverse events in the GM-CSF group were fever, skin rash, myalgia, and skeletal and abdominal pain; in the placebo group, they were fever, skin rash, and infection (Table 4). Only skin rash was significantly more frequent in the GM-CSF group (P⫽ 0.02).
Primary Efficacy Evaluation
Chemotherapy dose delivery rates of at least 90% over Cycles 1–3 and 80% over Cycles 1– 6 (see Materials and Methods) were attained in 18 patients (60%) and 22 patients (73.3%), respectively, in the GM-CSF arm, and in 14 patients (53.9%) and 17 patients (65.4%), respec-tively, in the placebo arm, for overall success rates of 57% and 50%, respectively. The differences between the two treatment arms were not statistically signifi-cant (Table 5). The results of the primary analysis were confirmed in the PP set of patients (data not shown).
Secondary Efficacy End Points
The mean nadir neutrophil count at each cycle ranged from 1.79 ⫻ 109/L to 0.81 ⫻ 109/L in the GM-CSF
group and from 1.53 ⫻ 109/L to 0.92 ⫻ 109/L in the
placebo group, with no statistically significant differ-ence between any of the 6 chemotherapy cycles (Table 6). The mean duration of neutropenia with an abso-lute neutrophil count (ANC) ⬍ 0.2 ⫻ 109/L and an
ANC ⬍ 0.5 ⫻ 109/L was 0.63 days and 9.77 days,
respectively, in the treatment arm and 1.12 days and 11.88 days in the placebo arm, respectively (Table 7) (not significant). Two patients in the placebo arm experienced thrombocytopenia (platelet levels⬍ 50 ⫻ 109/L). Seven patients in the GM-CSF group and nine
in the placebo group were treated with antibiotic or antifungal agents for a mean duration, respectively, of
TABLE 1 Patient Characteristics Characteristic GM-CSF (nⴝ 30) Placebo (nⴝ 26) No. % No. % Gender Male 19 63 14 54 Female 11 37 12 46 Age (yrs) Mean 37 — 34 — Range 18–77 — 18–64 — Disease stage II 7 23 7 27 III 8 27 11 42 IV 15 50 8 31 Bulky Yes 7 23 8 31 No 23 77 18 69
GM-CSF: granulocyte-macrophage–colony stimulating factor.
TABLE 2
Drop-Outs and Withdrawals
Causes GM-CSF Placebo All
Patients not completing the study (drop-outs) 2 4 6 Reason
Death 1 2 3
Lost to follow-up — 1 1
Early complete remission — 1 1
Allergic reaction to procarbazine 1 — 1
Premature withdrawal of study treatment
(withdrawals) 6 3 9
Reason
Adverse events 6 — 6
Administrative problems — 3 3
Major protocol violations — 1 1
All 8 8 16
GM-CSF: granulocyte-macrophage–colony stimulating factor.
TABLE 3
Clinical Status of Patients at the End of the Study
Status GM-CSF Placebo No. % No. % Complete remission 21 70.0 16 61.5 Partial remission 7 23.3 6 23.1 Nonresponse — — 1 3.8 Death 1 3.3 2 7.7 Lost to follow-up 1 3.3 1 3.8
GM-CSF: granulocyte-macrophage—colony stimulating factor.
The difference in the percentage of complete remissions was not significant (Fisher one-tailed exact test; P⫽ 0.350).
14.9 days (standard deviation [SD], 61.5 days) and 18.5 days (SD, 47.7 days) (not significant).
DISCUSSION
The current study was designed to investigate whether the incidence of myelosuppression after chemother-apy can be reduced by administering GM-CSF before administering the hybrid MA/MA schedule without affecting the optimal dose intensity of each of the drugs included in the combination. The rationale for this study was based on previous findings of a rapid reduction in the percentage of proliferating bone mar-row progenitor-precursor cells to below baseline levels after the withdrawal of GM-CSF and the possibility that this may lead to a myeloprotective effect against cytostatic drugs.20A previous randomized study
dem-onstrated the effectiveness of a schedule of GM-CSF given shortly before standard chemotherapy to meta-static breast carcinoma patients: None of the chemo-therapy cycles was delayed because of myelotoxicity
in the GM-CSF arm compared with 22% in the placebo group.24
To investigate this myeloprotective effect further, patients with newly diagnosed Hodgkin lymphoma who received the hybrid MA/MA regimen were ran-domized to receive GM-CSF or placebo for 4 days, with the final dose administered 48 hours before the scheduled chemotherapy. The primary objective was to evaluate the chemotherapy delivery rate, which re-flects the real adherence of patients to the planned doses of each of the drugs in the MA/MA protocol. With respect to this end point, we found a slight but not statistically significant beneficial effect in favor of GM-CSF; the effects on the secondary end points were not relevant, although neutrophil nadirs during the first three cycles were higher in the GM-CSF group, and the mean duration of severe neutropenia was shorter. The use of antibiotics and antifungal agents was similar in the two treatment groups.
This study failed to demonstrate that any statisti-cally significant benefit can be obtained from using GM-CSF as a chemoprotective agent, although our results are not completely discouraging. In fact, a trend in favor of increased dose intensity in the GM-CSF group was found. A possible explanation for these results is that the study was statistically underpow-ered. The number of 56 patients was chosen on the basis of an expected improvement in the success rate of 40% (range, 40 – 80%) in the GM-CSF group (see Materials and Methods). A smaller although clinically relevant benefit may exist, but the trial design was not adequate to detect it.
The differences in the results from the secondary
TABLE 4
Most Frequent Adverse Events
Event GM-CSF (nⴝ 30) Placebo (nⴝ 26) Skin rasha 10 2 Fever 6 3 Myalgia 3 1 Abdominal pain 3 0 Skeletal pain 2 1 Infection 0 2
GM-CSF: granulocyte-macrophage–colony stimulating factor.
aP⫽ 0.02.
TABLE 5
Primary Efficacy Evaluation (Intention to Treat Analysis)a
Measure GM-CSF Placebo P valueb No. % No. % Over Cycles 1–3 Success 18 60.0 14 53.9 0.423 Failure 12 40.0 12 46.1 — Over Cycles 1–6 Success 22 73.3 17 65.4 0.361 Failure 8 26.7 9 34.6 — Overall Success 17 56.7 13 50.0 0.409 Failure 13 43.3 13 50.0 —
GM-CSF: granulocyte-macrophage–colony stimulating factor.
aSuccess in maintaining the planned delivery rate (90% for Cycles 1–3; 80% for Cycles 1–6) in the
GM-CSF and placebo groups.
bFisher one-tailed exact test.
TABLE 6
Mean Nadirs of Neutrophil Counts (109/L)
Treatment Cycle 1 2 3 4 5 6 GM-CSF 1.79 1.39 1.29 1.14 0.81 1.00 Placebo 1.49 1.11 1.01 1.14 0.92 1.53 TABLE 7
Mean Duration of Neutropenia (Days)
ANC threshold value Treatment Mean duration (SD) P value
0.2⫻ 109 /L GM-CSF 0.63 (2.50) NS Placebo 1.12 (3.41) NS 0.5⫻ 109 /L GM-CSF 9.77 (14.87) NS Placebo 11.88 (12.79) NS
ANC: absolute neutrophil count; GM-CSF: granulocyte-macrophage–colony stimulating factor; SD: standard deviation; NS: not significant.
efficacy variables suggest a more favorable course af-ter the short af-term administration of GM-CSF. Because this study was not sized to reveal differences in the secondary variables, the lack of statistical significance is not surprising. Larger trials and different GM-CSF doses and schedules (i.e. lower doses or twice daily administration) are warranted to demonstrate whether its possible that the myeloprotective effect can be exploited to the point of clinical benefit.
Nevertheless, the results of this study will be of importance when planning adequate protocols in which growth factors are given during the intervals between rapidly recycling chemotherapy regimens. In this setting, chemotherapy often is resumed shortly after the end of growth factor administration, and the presence of still circulating progenitors may lead to permanent hemopoietic damage. Broxmeyer et al. have shown that progenitor cell proliferation remains high several days after the discontinuation of G-CSF, and they suggested that an adequate interval should elapse before the resumption of chemotherapy.21
Two recent studies addressing the issue of G-CSF priming used a time interval of 48 hours between the discontinuation of G-CSF and the use of chemother-apy and led to similar conclusions: De Wit et al. ob-served severe thrombocytopenia in breast carcinoma patients who were treated with doxorubicin (75 mg/ m3) and cyclophosphamide (1000 mg/m3) when
G-CSF was administered before the cytostatic drugs, and they concluded that the proliferation of progenitor cells was still increased 48 hours after the last dose of G-CSF and that the administration of chemotherapy at or within this time period actually worsens the toxic effects on bone marrow.25 In the second study, Tjan
Heijnen et al. observed that G-CSF priming in lung carcinoma patients was associated significantly with increased leukopenia and was associated less closely with increased thrombocytopenia.26
GM-CSF toxicity during the current study was similar to that reported in the literature,27,28and,
al-though GM-CSF-related adverse events caused treat-ment discontinuation in six patients, they were seri-ous in only two patients. Our data clearly show that no detrimental effects occur when GM-CSF is discontin-ued 48 hours before administering chemotherapy. If intensive chemotherapy is planned with a short inter-val between cycles, then GM-CSF may be indicated.
REFERENCES
1. Canellos GP, Anderson JR, Propert KJ, Nissen N, Cooper MR, Henderson ES, et al. Chemotherapy of advanced Hodgkin’s disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 1992;327:1478 – 84.
2. Klimo P, Connors JM. MOPP/ABV hybrid program: combi-nation chemotherapy based on early introduction of seven
effective drugs for advanced Hodgkin’s disease. J Clin Oncol 1985;3:1174 – 82.
3. Viviani S, Bonadonna G, Santoro A, Zanini M, Zucali R, Negretti E, et al. Alternating versus hybrid MOPP-ABVD in Hodgkin’s disease: the Milan experience. Ann Oncol 1991; 2(Suppl 2):55– 62.
4. Viviani S, Bonadonna G, Santoro A, Bonfante V, Zanini M, Devizzi L, et al. Alternating versus hybrid MOPP and ABVD combinations in advanced Hodgkin’s disease: ten-year re-sults. J Clin Oncol 1996;14:1421–30.
5. Devita VTJ, Serpick AA, Carbone PP. Combination chemo-therapy in the treatment of advanced Hodgkin’ s disease. Ann Intern Med 1970;73:881–95.
6. Bonadonna G, Zucali R, Monfardini S, De Lena M, Uslenghi C. Combination chemotherapy of Hodgkin’s disease with Adriamycin, bleomycin, vinblastine, and imidazole carbox-amide versus MOPP. Cancer 1975;36:252–9.
7. Bonadonna G, Santoro A. ABVD chemotherapy in the treat-ment of Hodgkin’s disease. Cancer Treat Rev 1982;9:21–35. 8. Longo DL, Glatstein E, Duffey PL, Young RC, Ihde DC,
Bastian AW, et al. Alternating MOPP and ABVD chemother-apy plus mantle-field radiation therchemother-apy in patients with massive mediastinal Hodgkin’s disease. J Clin Oncol 1997; 15:3338 – 46.
9. Goldie JH, Coldman AJ, Gudauskas GA. Rationale for the use of alternating non-cross-resistant chemotherapy. Cancer Treat Rep 1982;66:439 – 49.
10. Hryniuk W, Bush H. The importance of dose intensity in chemotherapy of metastatic breast cancer. J Clin Oncol 1984;2:1281– 8.
11. Carde P, MacKintosh FR, Rosenberg SA. A dose and time response analysis of the treatment of Hodgkin’s disease with MOPP chemotherapy. J Clin Oncol 1983;1:146 –53. 12. Longo DL, Young RC, Wesley M, Hubbard SM, Duffey PL,
Jaffe ES, et al. Twenty years of MOPP therapy for Hodgkin’s disease. J Clin Oncol 1986;4:1295–306.
13. Lagarde P, Bonichon F, Eghbali H, de MI, Chauvergne J, Hoerni B. Influence of dose intensity and density on thera-peutic and toxic effects in Hodgkin’s disease. Br J Cancer 1989;59:645–9.
14. van Rijswijk RE, Haanen C, Dekker AW, de Meijer AJ, Ver-beek J. Dose intensity of MOPP chemotherapy and survival in Hodgkin’s disease. J Clin Oncol 1989;7:1776 – 82. 15. Bezwoda WR, Dansey R, Bezwoda MA. Treatment of
Hodgkin’s disease with MOPP chemotherapy: effect of dose and schedule modification on treatment outcome. Oncology 1990;47:29 –36.
16. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating fac-tor (1). N Engl J Med 1992;327:28 –35.
17. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating fac-tor (2). N Engl J Med 1992;327:99 –106.
18. ASCO Ad Hoc Colony-Stimulating Factors Guidelines Expert Panel. American Society of Clinical Oncology. Recommen-dations for the use of hematopoietic colony-stimulating fac-tors: evidence-based, clinical practice guidelines. J Clin On-col 1994;12:2471–508.
19. Aglietta M, Piacibello W, Pasquino P, Sanavio F, Stacchini A, Volta C, et al. Rationale for the use of granulocyte-macro-phage colony- stimulating factor in oncology. Semin Oncol 1994;21:5–9.
20. Aglietta M, Piacibello W, Sanavio F, Stacchini A, Apra´ F, Schena M, et al. Kinetics of human hemopoietic cells after in vivo administration of granulocyte-macrophage colony-stimulating factor. J Clin Invest 1989;83:551–7.
21. Broxmeyer HE, Cooper S, Williams DE, Hangoc G, Gutter-man JU, Vadhan-Raj S. Growth characteristics of marrow hematopoietic progenitor/ precursor cells from patients on a Phase I clinical trial with purified recombinant human granulocyte-macrophage colony-stimulating factor. Exp He-matol 1988;16:594 – 602.
22. Vadhan-Raj S, Broxmeyer HE, Hittelman WN, Papadopoulos NE, Chawla SP, Fenoglio C, et al. Abrogating chemotherapy-induced myelosuppression by recombinant granulocyte-macrophage colony-stimulating factor in patients with sar-coma: protection at the progenitor cell level. J Clin Oncol 1992;10:1266 –77.
23. Theriault RL, Fraschini G, Holmes FA, Frye D, Gutterman JU, Hortobagyi GN. The myeloprotective effect of recombi-nant human granulocyte-macrophage colony-stimulating factor given sequentially with continuous infusion vinblas-tine in metastatic breast cancer patients. Am J Clin Oncol 1993;16:132– 6.
24. Aglietta M, Monzeglio C, Pasquino P, Carnino F, Stern AC,
Gavosto F. Short-term administration of granulocyte-mac-rophage colony stimulating factor decreases hematopoietic toxicity of cytostatic drugs. Cancer 1993;72:2970 –3. 25. de Wit R, Verweij J, Bontenbal M, Kruit WH, Seynaeve C,
Schmitz PI, et al. Adverse effect on bone marrow protection of prechemotherapy granulocyte colony-stimulating factor support. J Natl Cancer Inst 1996;88:1393– 8.
26. Tjan-Heijnen VC, Biesma B, Festen J, Splinter TA, Cox A, Wagener DJ, et al. Enhanced myelotoxicity due to granulo-cyte colony-stimulating factor administration until 48 hours before the next chemotherapy course in patients with small-cell lung carcinoma. J Clin Oncol 1998;16:2708 –14. 27. Antman KS, Griffin JD, Elias A, Socinski MA, Ryan L,
Can-nistra SA, et al. Effect of recombinant human granulocyte-macrophage colony-stimulating factor on chemotherapy-induced myelosuppression. N Engl J Med 1988;319:593– 8. 28. Gianni AM, Bregni M, Siena S, Orazi A, Stern AC, Gandola L,
et al. Recombinant human granulocyte-macrophage col-ony-stimulating factor reduces hematologic toxicity and widens clinical applicability of high-dose cyclophospha-mide treatment in breast cancer and non-Hodgkin’s lym-phoma. J Clin Oncol 1990;8:768 –78.
