compliance of physicians and patients with a consensus protocol for treatment of advanced breast...

J Cancer Res Clin Oncol (1989) 115:564-570 C f/ :er esearch Clinical �9 �9 Springer-Verlag 1989

Compliance of physicians and patients with a consensus protocol for treatment of advanced breast cancer * Franz Porzsolt 1, 2, Gerhard Meuret 3, Ernst D. Kreuser 1, 2, Siegmar Mende a, Longin Buchelt 4, Peter Strigl s, Martin Redenbacher 6, Friedrich Klumpp 7, Michael Schmelz s, Richard Kni~chelmann 9, Volker Hiemeyer 10, Kurt Fleiseher 11, Wilhelm Gans 12, Rosi Leichtle 1.12, Claudia Popp 1,12, Renate Kloiber 12, and Wolfgang Schreml ~' 2

Regional Study Group: 1 Tumorzentrum and 2 Abt. Innere Medizin III, Universit/it Ulm, 30nkologischer Schwerpunkt Ravensburg, 40nkologischer Schwerpunkt G6ppingen, 5 Diakoniekrankenhaus Schw/ibisch Hall, 6 Krankenhaus Schw/ibisch Gmiind, 7 Krankenhaus Heidenheim, 8 Krankenhaus Niirtingen, 9 Filderklinik Filderstadt, 10 Stadtkrankenhaus Kempten, 11 Kraukenhaus Geislingen, 12 Klinische Dokumentation Universitfit Ulm, Federal Republic of Germany

Summary. In a multicenter study we used a consensus protocol including more than five subsequent therapeutic steps for treatment of patients with advanced breast cancer. A total of 335 evaluable patients from 27 participating hospitals were allocated to a low- or high-risk group, receiving different therapies during the initial phase of treatment. About half of these patients were treated without protocol violations (corn- pliers). The protocol non-compliers were divided into three groups: those receiving more intensive therapy than recommended, those with similarly intensive, and those with less intensive therapy. The reasons for protocol violations were analysed. The intensity of the therapy given actually was correlated with the survival of subgroups. Median survival times were significantly longer in 208 low-risk than in 127 high-risk patients (P < 0.0001), marginally longer in 165 compilers than in 170 non-compilers (P<0.04), significantly longer in low-risk compliers than in low-risk non- compilers (P=0.002), and significantly shorter in high-risk compilers than in high-risk non-compilers (P = 0.007). Survival of all subgroups of low-risk non- compilers was the same regardless of the actual therapies given. The survival of high-risk patients who received less intensive therapy was significantly longer than that of high-risk compilers (P = 0.015). After six Cycles of successful chemotherapy there was no differ-

* This study was supported in part by the Bundesministerium f/Jr Ar- beit und Sozialordnung and by the German subsidiaries of Asta-De- gussa, Ciba-Geigy, Cyanamid-Lederle, Farmitalia, Pharma-Leo, Rhone-Poulenc, Upjohn

Offprint requests to: F. Porzsolt, Tumor Center University of Ulm, Robert-Koch-Strasse 8, D-7900 Ulm, FRG Abbreviations used." CAF, cyclophosphamide, Adriblastin, 5- fluorouracil; CMF, cyclophosphamide, metothrexate, 5- fluorouracil

ence, either in time to progresion or in survival, between patients who had received either maintenance therapy or no therapy. We postulate that the groups of low-risk and high-risk patients comprised patients with different prognoses. Among low-risk patients, survival of the subgroup with poor prognosis (low- risk non-compliers) was not influenced by therapy. Among high-risk patients, a subgroup with poor prognosis may have been overtreated by using standard chemotherapies as recommended in our consensus protocol.

Key words: Compliance - Consensus protocol - Ad- vanced breast cancer - Survival - Overtreatment

Introduction

In the course of advanced breast cancer physicians are frequently confronted with the problem that an ongoing therapy has to be changed because of progression of the disease. As these therapeutic decisions have to be made repeatedly in most patients with advanced breast cancer it might be useful to orientate the decisions by a protocol whose steps had been evalu- ated in a clinical study. Like others (Canellos 1987) our regional study group developed a consensus protocol with a sequence of different therapeutic steps for groups of patients of defined risk, This concept was based on commonly accepted treatment schedules and was carried out as a prospective multicenter phase IV study starting in 1983. The aim of this phase IV study was to describe the compliance of physicians and patients with a consensus protocol for treatment of advanced breast cancer.

F. Porzsolt et al.: Compliance with consensus protocol 565

A recent report on our results (Porzsolt et al. 1987) demonstrated that about 50% of all advanced breast cancer patients of the service area of our Tumor Center had been recruited for the study. Of these patients 40% received three consecutive types of therapy and 21% received four therapies within a period of 3 years. Protocol violations (defined in Materials and methods) increased from 12% in first-line therapies to 47% in fourth-line therapies (Porzsolt et al. 1987), the overall violation rate being 51% (170 of 335 patients). These patients are considered as non-compliers.

In this paper, we investigate the reasons for protocol violations, and the intensities of therapies given to compliers and non-compliers are compared. Finally we correlate the survival of compliers and non-corn- pliers with the intensities of given therapies.

Materials and methods

Patients. Between January 1983 and December 1985, 347 patients with advanced breast cancer were recruited from 27 participating in- stitutions. Of these, 335 patients (96%) were evaluable. Exclusion criteria were secondary malignancies and preceding adjuvant chemotherapy.

Patients were allocated to a low- or high-risk group. High-risk conditions were fulfilled if at least one of the following criteria was present: CNS or liver metastases, lymphangiosis carcinomatosa of the lungs, fast progression of metastatic disease, negative hormone receptor status, hypercalcemia, inflammatory cancer. According to these criteria 208 patients were allocated to the low-risk group and 127 to the high-risk group. The receptor status of the primary tumor was unknown in 40%. The median age in the low-risk group was 63.8 years, and 57.2 years in the high-risk group; 26 out of 208 low-risk patients (13 %) and 36 out of 127 high-risk patients (28 %) were premenopausal (interval to last menstruation shorter than 1 year). The incidence of metastatic sites and of risk factors is shown in Table 1.

Treatment. Treatment started after detection of the recurrence. Each treatment schedule was continued until disease progression occurred. Then the next therapeutic step was applied.

Table 1. Incidences of metastatic sites and of risk factors in low- and high-risk patients; high-risk factors are indicated by (x)

Sites of metastases/riskfactors Low-risk High-risk patients patients n=208 n=127

Bone metastases 123 Skin metastases 48 Lymph node metastases 40 Pulmonal metastases 30 Pleural metastases 25 Ovarial metastases 4 Liver metastases (x) 0 Lymphangiosis carcinomatosa of lungs (x) 0 CNS metastases (x) 0 Other metastases 9

Fast progression of metastatic disease (x) 0 Neg. hormone receptor (x) 0 Inflammatory cancer (x) 0 Hypercalcemia (x) 0

56 26 30 17 18

1 44 23

9 8

72 37 15 9

hiffh RISK low

pre- post meno )ausal menopausal

. . . . . . . . .

(6 cycl)plus (6 cycl)plus ovarect.

[ 1

pre- post- menopausal menopausal

o+ .......

~h~ if no progress after 6 cycles

[ P.ANDOMI ZATION

r

3 cycl. CAF or CMF, then CMF till progression of disease, No additional hormon.th.

with ~ w i t h o u t high risk ~ high risk criteria

I C-- pretreated pretreated with CAF with CMF

I I

criteria

i i pretreated pretreated with CAF with CMF

Fig. 1. Protocol for treatment of advanced breast cancer patients. The initial therapies and the consecutive therapies after progression of the disease are shown. The therapies had to be applied for at least 2 months or two cycles before evaluation of response. If the disease was still progressive the next therapeutic step was applied. If the evaluation confirmed remission or stable disease the therapy was continued till progression. TAM, tamoxifen p.o. 30 rag/day. MPA, medroxyprogesterone acetate p.o. 1000mg/day. CAF, cyclophosphamide i.v. 500 mg/m 2 day 1 and Adriablastin i.v. 50 mg/m 2, day 1 and 5-fluorouracil i.v. 500 mg/m 2, days 1 + 8. Repeated day 28. CMF, cyclophosphamide p.o. 100 mg/m 2, days 1-14 and methotrexate i.v. 40 mg/m 2, days 1 + 8 and 5-fluorouracil i.v. 600 mg/m 2, days 1 + 8. Repeated day 28. CHL + MTX, chlorambucil p.o. 0.1 mg kg - 1 day - i and methotrexate p.o. 15mg weekly. MTX+5FU+L, methotrexate i.v. 200 mg/m 2, day 1 and 5-fluorouracil i.v. 600 rag/ m 2, day 1, 1 h after MTX, and Leucovorin p.o. 10 mg/m 2, 6 times every 6 h starting 24 h after MTX. Repeated day 21. ldAo low-dose Adriablastin i.v. 15 mg/m 2 weekly. A V, Adriablastin i.v. 40 mg/m 2 day 1 and vincristine i.v. 1.4 mg/m 2, days 1 + 8. Repeated day 28. PM, prednimustin p.o. 160 mg/m 2, days 1-5. Repeated day 21. * Therapy is skipped if patient had not responded to previous ap- plication of this drug.

The initial treatment for high-risk patients consisted of six cycles of chemotherapy with or without hormonal therapy (Fig. 1). Patients with negative receptor status received chemotherapy only. The CAF (Bull et al. 1978; Tranum et al. 1978) regimen was used if there were no contraindications against Adriamycin. Otherwise CMF (Canellos et al. 1974, 1976) was given. Chemotherapy with or without hormonal therapy consisted in chemotherapy plus ovarectomy in pre-

566 F. Porzsolt et al.: Compliance with consensus protocol

menopausal patients or 6 months of tamoxifen in postmenopausal patients.

The initial treatment for low-risk patients was ovarectomy in premenopausal and tamoxifen in postmenopausal patients till disease progression. If disease progression occurred, hormone re- sponders received a second hormonal treatment consisting in tamoxifen for patients after ovarectomy and medroxyprogesterone acetate in patients pretreated with tamoxifen. Patients with primary and secondary resistance to endocrine therapies received six cycles of CAF. CMF was given as an alternative if there was a contraindica- tion against Adriamycin. At this point of the therapeutic flow dia- gram, patients with complete remission, partial remission or stable disease, from the high- and low-risk groups, were pooled and ran- domly allocated to two groups: observation only or chemotherapy using chlorambucil plus methotrexate.

If subsequent disease progression occurred three cycles of CAF were applied to patients without contraindications for Adriamycin treatment. Thereafter responding patients continued with CMF treatment till progression. Patients with contraindications against Adriamycin received CMF.

Patients with progressive disease, during either initial or consecutive chemotherapy, were divided into two groups according to the presence or absence of high-risk criteria.

High-risk patients, pretreated with CAF, received methotrexate/ 5-fluoruoracil/Leucovorin (Allegra 1983), those pretreated with CMF received low-dose Adriamycin weekly or Adriamycin plus vincristine. The therapy of failures was continued in several consecutive steps using tamoxifen, medroxyprogesterone acetate and prednimustin.

Low-risk patients received the same therapy as high-risk patients when progression was detected during CAF or CMF combination chemotherapy. The sequence of the therapeutic steps, however, was different. Low-risk patients were treated by endocrine therapies before using the methotrexate/5-fiuorouracil/Leucovorin or Adria- mycin/vincristine combinations.

The physicians were requested to document all therapeutic mea- sures, effects, and side-effects for each visit of the patient. The study center continuously recorded the results.

Method of data exchange. In order to facilitate the realization of the complex protocol, the physicians were guided by the study center for each therapeutic step provided in the protocol (Porzsolt et al. 1987). They reported the results to the study center after each examination of a patient.

Analysis of protocol violations. Protocol violations were defined as follows: continuing an ongoing regimen despite progressive disease, premature termination of a therapeutic step, or changing the therapy to an unscheduled regimen. Protocol violations were classified as "less intensive", "similarly intensive", or "more intensive". "Less intensive" stands for (a) no therapy or unproven therapy being given instead of the proposed hormonal therapy; (b) hormonal therapy being given instead of the proposed chemotherapy with of without hormonal therapy. "Similar" signifies (a) an unproven type of therapy without cytoreductive potential being applied instead of observation only; (b) hormonal therapy other than the proposed hormonal treatment being applied; (c) another chemotherapy with or without hormonal therapy being used rather than the proposed chemotherapy with or without hormonal therapy. "More intensive" means (a) hormonal or chemotherapy with or without hormonal therapy being given instead of the proposed observation phase; (b) chemotherapy with or without hormonal therapy being given instead of the proposed hormonal therapy. Discrepancies of dosage were not considered.

Statistical methods. Survival rates were calculated according to Kaplan-Meier (BMDP Statistical Software Inc., Los Angeles, Calif).

The Mantel-Cox test was used to compare survival of different subgroups. The Wilcoxon rank sum test was used for comparing the disease-free intervals of low- and high-risk patients as there was no nor- mal distribution of the disease-free intervals of groups.

Results

Median disease-free survival after primary surgery was 32.2 months in 208 low-risk patients and 13.0 months in 127 high-risk patients (P < 0.0001). After detection of metastases, the median survival of all 335 patients was 26.9 months. Median survival was 32.1 months in 208 low-risk patients, and 13.1 months in 127 high-risk patients (P < 0.0001).

No significant differences were observed between responding patients whose chemotherapy was stopped after six cycles and those who received maintenance

Table 2. Results of the randomized part of the study

Group" Median time Median to progression b survival b (months) (months)

Untreated controls (n = 35) 6.2 23.6 Maintenance therapy (n=33) 7.6 17.4

P 0.42 0.65

" Patients with complete response, partial response or stable disease after six cycles of chemotherapy were randomized to an untreated control group or to maintenance therapy. Patients with progressive disease during or after completion of six cycles of chemotherapy are not included. The groups are balanced for age, menopausal status, allocation to low/high-risk groups, and CAF/CMF pretreatment (data not shown) b Time to progression and survival time from randomization

(n t - O

0 >

0 fJ 2 2 O.

60

40

low risk patients

20 II N 48 19 30 ~o ~ 91 ~7

0 I. 2. 3. 4. I.

line of therapy

Fig. 2. Protocol violations of low-risk and high-risk patients in first- to fourth-line therapies. In each patient only the first deviation from protocol was recorded. The quotients give the numbers of patients deviating from protocol over the numbers of patients at risk

high risk patients

28 14 6

2. 3. 4.

F. Porzsolt et al.: Compliance with consensus protocol 567

> L _

o = :E & 2

1.00 ~ _

0.75

0 50

025

I 0.00 6

L__\__ I L 1

c 3 - i

I i I

12 18 2t+

survival (months)

Fig. 3. Survival of low-risk compliers (n = 117; pliers ( n = 9 1 ; - - - ) . P < 0 . 0 0 2

- L _ l " - - ! _

c . . . . . . . . . :

t 0 i

36

) and non-corn-

1.00 L - ........... : : : : i . L- - I ~ . . . . . . . . . . . . .

. . . . I : ' "~

~, 025 '-~ i"'.., l :_~ ---"L:;~=~ -', l > - ' ~ - 1 ~- :,.. i

"~4~ . . . . . . . . . . . . . . . . . .

g 0.50 -~

2 . . . . . . . . . . . . !I =. ii

0 l 2 5 ] [

0.00 6 12 I 24 36

survival (months)

Fig. ft. Survival of low-risk compliers (n = 1 ] 7; - - - ) and non-corn- pliers with deviation to more intensive (n = 18; - - . . - - ) or less intensive ( n = 33; - - - ) or therapies with similar ( n = 4 0 ; . . . ) intensity. P < 0.0009

S w~

g

o c ~

1.00

0.75

0.50

0.25

I I I I

0.00 6 12 18 24

surviw[ {monthsl

Fig.4 . Survival of high-risk compilers (n = 48; pliers (n = 79; - - - ) . P = 0.007

I --- I

L~

I

30 i

36

) and non-com-

chemotherapy. The median time to disease progression was 6.2 months for patients on observation only and 7.6 months for those on maintenance chemotherapy (P=0,42). The corresponding median survival times were 23.6 months and 17.4 months respec- tively (P = 0.65) (Table 2).

Protocol violations were observed in 170 of 335 patients (50.8%). They occurred earlier and were more frequent in high-risk patients than in low-risk patients (Fig.2). There was only a marginal difference (P < 0.04) between survival of 165 patients treated according to the protocol (median not yet reached) and 170 patients with protocol violations (22.2 months). However, significant differences in survival emerged when the risk groups were analysed separately. Low- risk patients without protocol violations (compliers)

1.00

0.75

g 0.50 L_ & o c l

0.25

i i ~176176 survival (mo~thsl

Fig.6 . Survival of high-risk compliers ( n = 4 8 ; - -

1_ . . . . . . . . . .

L~

i . . . . . . . . 1 ,.....~ I- 1

i ,

........ "-"~ ]- . . . . . . . 7

.............. i .......................................... I .........

- t l I

30 i

36

-) and non-cam- pliers with deviation to less intensive (n = 37; - - - ) or therapies with similar (n = 38; . . . ) intensity. P < 0.016. Survival of four n o n - c a m - pliers with deviation to more intensive therapy is not shown

had a longer (P= 0.002) overall survival (median not yet reached) than those with protocol violations (non- compliers) (median survival 24.1 months) (Fig. 3). In contrast, the survival of high-risk compliers was significantly (P=0.007) shorter (median survival being 11.4 months) than that of non-compliers (median survival being 16.8 months) (Fig. 4). The group of low- risk non-compliers (n=91) comprised three subgroups of patients: those who received less intensive therapies than scheduled in the protocol (n=33), those receiving more intensive therapies (n = ] 8), or those receiving therapies of similar intensity (n = 40). In the group of high-risk non-compliers (n = 79) 37 patients received less intensive therapies, 4 patients received more intensive therapies, and 38 patients received therapies of similar intensity.


P a

t i e

n

t S

No deviation from protocol P

r

o

P 0

r

t i 0

n

Less intensive therapies

More intensive therapies

o 1 4 7 10 1 4 7 10 1 4 7 10 f Months Months Months

Fig. 7. Intensities of therapies applied to low-risk patients during the first 12 months of treatment. The intensities of treatments according to protocol, of less intensive therapies, and of more intensive therapies are shown. The intensities of similar therapies (not shown) were identical to those of the first group. Black, proportion of patients who received chemotherapy with or without hormonal therapy. Grey, proportion of patients who received hormonal therapy. White, proportion of patients who received neither chemo- nor hormonal therapy

No deviation from Less intensive P

protocol therapies r o p l O 1

P a

r , t i i e

n o t n

s

o 1 4 7 1 0 1 4 7 1 0 f Months Months

Fig. 8. Intensities of therapies applied to high-risk patients during the first 12 months of treatment. The intensities of treatments according to protocol and of less intensive therapies are shown. The data of our high-risk patients treated with more intensive therapies and of patients treated with similar therapies are not shown. For symbols see legend to Fig. 7

It is shown in Fig. 5 that the mean survival o f low- risk compliers was significantly longer ( P = 0 . 0 0 8 ) (median no t yet reached) than that o f all subgroups o f low-risk non-compliers regardless o f the applied therapy. In contrast , high-risk patients with pro tocol deviations f rom less intensive therapy lived longer (median survival 22.6 months) t h a n high-risk compliers (median survival 14.1 months) (P = 0.015) (Fig. 6).

The analysis o f p ro toco l violations demonst ra ted that low-risk patients with less intensive therapy were mainta ined longer on observat ion only than those treated according to the protocol . Those being treated more intensively received more chemotherapy with or wi thout ho rmona l therapy (Fig. 7). In high-risk patients being treated less intensively than recommended by the protocol , chemotherapy was usually replaced by either endocrine t rea tment or no therapy (Fig. 8).

Evaluat ion o f relapse chrono logy during the first year o f t rea tment in high-risk patients showed high mortal i ty dur ing the first 3 months in compliers and lower morta l i ty in non-compliers deviating to less intensive t rea tment (Table 3). The search for prognost ic factors within the low- and high-risk groups demon-

Table 3. Incidences of disease progression in two subgroups of high-risk patients during the first 12 months of treatment

Time of treatment (months)

Without deviation from protocol Deviation to less aggressive therapy

Patients Disease Deaths Patients Disease Deaths at risk progressions at risk progressions

1 3 48 1 11 37 3 3 4- 6 35 8 8 34 9 1 7- 9 12 4 2 33 6 5

10-12 7 2 3 28 4 6

F. Porzsolt et al.: Compliance with consensus protocol

Table 4. Reasons for protocol violations

Reasons for protocol Low- High- Total violations risk risk

n n n %

Patients not compliant 27 26 53 31.3 Objective side-effects 7 9 16 9.4 Physician not compliant 21 a 23 44 25.9 Not given 36 a 21 57 33.5

Total 91 79 170 100.0

18 of 57 patients were derived from a single institution where physicians used chemotherapy instead of hormonal therapy for initial treatment of patients with pulmonal metastases

strated no influences on survival from the menopausal status, disease-free intervals or numbers of metastatic site.

The reasons for protocol violations are listed in Table 4. True therapeutic intolerance caused protocol violations in only 16 of 335 patients (5%). The proposed treatment had been refused by 15.8% patients. In 101 of 335 patients (30%) the reasons for protocol violations resulted from the physicians' own decisions, rejecting the therapeutic step scheduled by the protocol.

Discussion

The results of this study demonstrate that it was possible to obtain a consensus on the treatment of patients with advanced breast cancer between a Tumor Center and different non-specialized hospitals. The protocol considered low- and high-risk patients, scheduled several subsequent therapeutic steps, and tested the effects of maintenance therapy versus no therapy on time to progression and survival of re- sponders to the primary chemotherapy. The complex- ity of the protocol required a system guided by the Tu- mor Center to guarantee compliance with the protocol.

Nevertheless, deviations from the protocol were observed in 50% of all patients. When analysing the reasons for protocol violations we found that in the majority of cases no rational explanation was avail- able. Many physicians and patients obviously felt that the therapy recommended by the protocol was not ap- propriate for the present situation of a particular patient. Instead of the therapy recommended by the protocol, patients received either more intensive or less intensive therapy or therapy of similar intensity.

The prospective allocation of our patients to a low- and high-risk group is comparable with the alloca- tions in other studies. Our low-risk patients were comparable to estrogen-receptor-positive groups of patients from other studies, who survived 15-48 months

569

from first recurrence of the disease; our high-risk patients were comparable to estrogen-receptor-negative patients in other studies, who survived for 10-19 months (Possinger and Wilmanns 1986; Hahnel et al. 1979; Paterson et al. 1982; Stewart et al. 1981; Howell et al. 1984; Howat et al. 1985; Moseley et al. 1980). It was not possible to define further prognostic criteria for the description of subgroups within low- and high- risk groups.

Analysing the compliance to the protocol we found that survival was not different between corn- pliers and non-compliers. However, survival was in- versely correlated with the intensity of therapy. There was no subgroup either within the low- or high-risk group in which more intensive therapy correlated with longer survival. However, there were subgroups within the low- and high-risk groups that were treated with less intensive therapies but survived longer than others (low-risk non-compliers with more intensive therapies and high-risk compliers). This observation is in contrast to retrospective analyses (Swenerton et al. 1979; Hryniuk and Bush 1984) that suggested that higher doses of cytotoxic drugs and shorter intervals between treatment cycles may prolong survival of advanced breast cancer patients. However, prospective randomised trials as well as non-randomised trials failed to show that intensification of therapies en- hances life expectancy (Hortobagyi et al. 1987; Rosner et al. 1987; Henderson et al. 1981; Israel et al. 1980; Paterson et al. 1985). The more recent discussion on this problem is still controversial (Hryniuk 1988; Peters et al. 1988; Tannock et al. 1987; Henderson et al. 1988).

There is an important difference between low- and high-risk patients as far as disease progression and compliance with the protocol are concerned. In contrast to the low-risk group, all high-risk patients demonstrated early disease progression irrespective of compliance or non-compliance with the protocol (Table 3). However, compliers survived for significantly shorter times than non-compliers who had received less intensive therapies. This raises the question of why the survival of high-risk compliers was shorter than that of high-risk non-compliers, who received less intensive therapies. As we do not have direct evi- dence for treatment-related deaths in high-risk corn- pliers, it is possible that the biology of the disease, i.e. unknown prognostic factors, rather than the applied therapy influenced the survival of high-risk patients. If this assumption is true we have to consider the overtreatment of high-risk patients who were compliant to the protocol.

This hypothesis is supported by the results of the randomised part of the study (Schreml et al. 1985). There it is shown that neither time to progression nor


survival was prolonged in patients who received maintenance therapy as opposed to no therapy after the successful completion of six cycles of chemotherapy.

As the protocol of this study was not designed to test the effects of therapies on survival, we should not draw conclusions but rather ask the question whether or not there are reliable data demonstrating that survival of advanced breast cancer patients is really influenced by therapy.

Acknowledgements. We thank Prof H. Heimpel, Head Dept. Medicine III and Chairman Tumor Center Ulm, for critical review of the manuscript. We are grateful to all physicians in the service area of Tumor Center Ulm who contributed to this study. The editorial help of Dr. D. Bunjes is appreciated. We thank our secretaries Ms. I. Brand, Ms. G. Geiger, and Ms. S. Goldmann.

References

Allegra JC (1983) Methotrexate and 5-fluorouracil following tamoxifen and premarin in advanced breast cancer. Sem Oncol 10 (suppl 2):23-28

Bull J, Tormey DC, Li SH, Carbone TP, Falkson G, Blom J, Perlin E, Simon R (1978) A randomized comparative trial of Adriamy- cin versus methotrexate in combination drug therapy. Cancer 41:1649-1657

Canellos GP (1987) Treatment of metastases. In: Harris JR, Hellman S, Henderson IC, Kinne DW (eds) Breast diseases. Lippincott, Philadelphia, pp 385-391

Canellos GB (1988) The dose dilemma. J Clin Oncol 6:1363-1364 Canellos GP, DeVita VT, Gold GL, Chabner BA, Schein PS, Young

RC (1974) Cyclical combination chemotherapy for advanced breast carcinoma. Br Med J 1:218-220

Canellos GP, Pocock SJ, Taylor SG, Sears ME, Klaasen DJ, Band PR (1976) Combination chemotherapy for metastatic breast carcinoma. Cancer 38:1882-1886

Hahnel R, Woodings T, Vivian AB (I979) Prognostic value of oestrogen receptors in primary breast cancer. Cancer 44:671-676

Henderson IC, Hayes DF, Gelman R (1988) Dose-response in the treatment of breast cancer: a critical review. J Clin Oncol 6:1501-1515

Henderson K, Gelman R, Cavellos SP, Frei E (1981) Prolonged disease free survival in advanced breast cancer treated with super- CMF Adriamycin: an alternating regimen employing high-dose methotrexate with citrovorum factor rescue. Cancer Treat Rep (suppl 1) 65:67-75

Hortobagyi GN, Bodey GP, Buzdar AU, Frye D, Legha SS, Malik R, Smith RL, Blumenschein GR, Yap HY, Rodriguez V (1987) Evaluation of high-dose versus standard FAC chemotherapy for advanced breast cancer in protected environment units. A prospective randomized study. J Clin Oncol 3:354-364

Howat JMT, Harris M, Swindell R, Barnes DM (1985) The effect of oestrogen and progesterone receptors on recurrence and survival in patients with carcinoma of the breast. Br J Cancer 51:263-270

Howell A, Harland RNL, Bramwell VHC, Swindell R, Barnes DM, Redford J, Wilkinson MJS, Crowther D, Sellwood RA (1984) Steroid hormone receptors and survival after first relapse in breast cancer. Lancet I:588-591

Hryniuk WM (1988) More is better. J Clin Oncol 6:1365-1367 Hryniuk WM, Bush H (1984) The importance of dose intensity in

chemotherapy of metastatic breast cancer. J Clin Oncol 2:1281- 1288

Israel L, Brean JL, Aguilera J (1980) High-dose cytoxan and high- dose fluorouracil: a new first regime for advanced breast cancer. Proc Am Assoc Cancer Res/Am Soc Clin Oncol 21:409

Moseley HS, Peetz ME, Keenan E J, Awrich AE, Fletcher WS (1980) Endocrine ablation for metastatic breast cancer: a reappraisal of hormone receptors. Am J Surg 140:164-172

Paterson AHG, Zuck VP, Szafran O, Lees AW, Hanson J (1982) In- fluence and significance of certain prognostic factors on survival in breast cancer. Eur J Cancer Clin Oncol 18:937-943

Paterson AHS, Szafran O, Hanson J, Cyr M, Lees AW (1985) Re- sponse to treatment and its influence on survival in metastatic breast cancer. Am J Clin Oncol 8:283-292

Peters WP, Shpall E J, Jones RB, Olsen GA, Bast RC, Gockermann JP, Moore JO (1988) High-dose combination alkylating agents with bone marrow support as initial treatment for metastatic breast cancer. J Clin Oncol 6:1368-1376

Porzsolt F, Schreml W, Buchelt L, Meuret G, Mende S, Strigl P, Re- denbacher M, Klumpp D, Schmelz M, KnSchelmann R, Hie- meyer V, Fleischer K, Kreuser ED, Leichtle R, Popp C, Kloiber R (1987) Konzept zur Behandlung metastasierender Mamma- karzinome auBerhalb von Universit/itskliniken: Beschreibung der Methode und Priifung der Effizienz. Onkologie 10:367-373

Possinger K, Wilmanns W (1986) AIO-Studien zur Behandlung des metastasierenden Mammakarzinoms. In: Nagel GA (ed) Mam- makarzinome. Springer, Berlin Heidelberg New York Tokyo, pp 84-91

Rosner D, Nemoto T, Lange WW (1987) A randomized study of intensive versus moderate chemotherapy programs in metastatic breast cancer. Cancer 59:874-883

Schreml W, Porzsolt F, Mende S, Buchelt L, Strigl P, Nessler A, Brass B (1985) Randomisierter Vergleich zwischen Beobachtung und Erhaltungstherapie nach Induktions-Chemotherapie bei metastasierendem Mammarkarzinom. Dtsch Ges Senol Ham- burg

Stewart JF, King RJB, Sexton SA, Millis RR, Rubens RD, Hayward JL (1981) Oestrogen receptors, sites of metastatic disease and survival in current breast cancer. Eur J Cancer 17:449-453

Swenerton KD, Legha SS, Smith T, Hortobagyi GN, Gehan EA, Yap HY, Gutterman JU, Blumenschein GR (1979) Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 39:1552-1562

Tannock IF, Boyd NF, DeBoer G, Erlichman C, Fine S, Larocque G, Mayers C, Perrault D, Sutherland H (1988) A randomized trial of two dose levels of cyclophosphamide, methotrexate, and fluorouracil chemotherhapy for patients with metastatic breast cancer. J Clin Oncol 6:137%1387

Tranum B, Hoogstraten B, Kennedy A, Vaughn CB, Samal B, Thigpen T, Rivkin S, Smith F, Palmer RL, Costanzi J, Tucker WG, Wilson H, Maloney TR (1978) Adriamycin in combination for the treatment of breast cancer. Cancer 41:2078-2083

Received 2 August 1989/Accepted 22 August 1989

compliance of physicians and patients with a consensus protocol for treatment of advanced breast...

Documents