Drugs ten years later: Epirubicin

Annals of Oncology 4: 359-369, 1993. © 1993 Kluwer Academic Publishers. Printed in the Netherlands.

Review Drugs ten years later: Epirubicin G. Bonadonna, L. Gianni, A. Santoro, V. Bonfante, P. Bidoli, P. Casali, R. Demicheli, & P. Valagussa Istiluto Nazionale Tumori, Milan, Italy

Key words: epirubicin, anthracyclines, chemotherapy

Introduction

Preclinical and clinical pharmacology

The first clinical report on adriamycin (doxorubicin), the 14-hydroxydaunomycin derivative, dates back to almost a quarter of a century ago [1]. The initial findings [2], indicating a broad spectrum of antitumor activity as well as cardiac effects, were confirmed within a few years in large case series all over the world [3—7]. During the past two decades, doxorubicin represented the single most effective agent in the treatment of breast cancer, malignant lymphomas, bone and soft tissue sarcomas, and an essential component in numerous drug combinations for the management of various neoplastic diseases of adults and children. The search in experimental animal systems for other anthracycline analogs with better therapeutic index than daunorubicin and doxorubicin has been tenaciously pursued since early 1970s by investigators of Farmitalia Carlo Erba and the Milan Cancer Institute, namely F. Arcamone, A. Di Marco, A. M. Casazza and their coworkers [8-10). Epirubicin (4'-epiadriamycin, epidoxorubicin, Pharmorubicin*) was the first compound of the new series which became available during the late 1970s for phase I and II studies [11, 12]. In 1984, an international symposium held in Milan [13] concluded that there was preliminary evidence in favor of similar treatment efficacy between epirubicin and doxorubicin but the new analog was moderately less toxic than the parent compound in terms of immediate and delayed side effects. During the past decade, epirubicin underwent extensive pharmacologic, toxicologic and therapeutic investigations. It is now the proper time to briefly review what has been achieved in terms of pharmacology, optimal dose, schedule, and cost-benefit ratio in the most important human malignancies. Primarily because of space, this review will not attempt to dissect every single study reported in the medical literature. Readers interested in exhaustive details are kindly referred to the recent papers of Mouridsen et al. [14] and Delozier etal.[15|.

Chemistry and biochemical reactions of epirubicin Epirubicin (EPI) is a semisynthetic L-arabino derivative of doxorubicin (DOX) in which the aminosugar daunosamine is replaced by acosamine (Fig.l). The equatorial orientation of the hydroxyl group at the carbon 4' position of the sugar has several consequences on the physical-chemical properties of EPI. In particular, EPI is a weaker base and is more lipophilic than DOX [9]. These features might be relevant to the favorable pharmacological profile of the analog [8,10]. Intercalation of DNA, inhibition of topoisomerase II activity, generation of oxygen and drug free radical, as well as chelation of transition metal-ions are the biochemical reactions most commonly implicated in the mechanism of action and cardiac toxicity of DOX and other anthracyclines [16-19]. Largely, the same reac-

<

CH2OH 'OH

OCHi

O

OH

NH 2 daunosaminc

NH 2 acosamine

Doxorubicin : R = daunosaminc Epirubicin : R = acosamine

Fig. I. Chemical structure of doxorubicin (adriamycin) and 4'-epidoxorubicin (epirubicin).

360

tions have also been demonstrated for or are credited to EPI [14, 16, 19]. Here suffice it to say that the steric modification introduced in EPI's structure affects to some degree the stability of DNA-anthracyclines complex [9], causes a faster influx and a lower retention of the analog inside tumor and normal cells [14], determines a different structure of the anthracycline coordination complex with metal ions [8], reduces the ability of the epirubicin-iron complex to peroxidize lipids [20], and favors a unique glucuronidation pathway that is not observed with DOX and other anthracyclines [21]. These biochemical peculiarities may explain the better therapeutic index of EPI than DOX (see below), but do not sustain any substantial difference of the type of reactions implicated in the mechanism of action and toxicity of the analog. It is then not surprising that the two anthracyclines display complete cross-resistance in all in vitro systems [22], and that EPI is also implicated in the phenomenon of multidrug resistance [23]. Pharmacokinetics and metabolism The pharmacokinetics of EPI after intravenous bolus administration can be fitted to an open three compartment model [24-27]. The comparative analysis of the plasma decay of DOX and EPI consistently show that the analog is more rapidly eliminated than the parent compound. The observation is confirmed not only by the cross-over study performed by Camaggi et al. [24], but also by the more stringent experimental approach with simultaneous administration of the two anthracyclines [26]. While the initial and intermediate halflives of DOX and EPI are very similar, the terminal half-life is substantially shorter for EPI than for DOX [26]. This difference may be responsible for the lower toxicity, in particular cardiac toxicity, of the analog, and is generally attributed to the unique metabolism of this anthracycline in humans [19]. In fact, EPI is the only anthracycline that serves as a substrate for fj-glucuronidation [21]. It is now clear that this metabolic pathway plays a key role in the mechanism of EPI's disposition in humans. According to different authors, the glucuronides of EPI and epirubicinol can account for about one third to more than half of EPI's area under the plasma concentration x time curve (AUC) [25, 26, 28]. Since glucuronidation is a classical pathway of drug detoxification and facilitates the excretion process, the large extent and the inter-individual variability of this EPI's metabolism may bear important consequences on its tolerability in humans. As already known for DOX, elevated serum bilirubin levels seem associated with very long terminal half lives of EPI and its metabolites [25]. Thus, caution should be used in the administration of EPI to patients with altered liver function, elevated serum bilirubin, and/or liver metastases. EPI has a wide range of therapeutic doses in humans (50-180 mg/m2). One report in the medical literature [28] suggested that EPI's pharmacokinetics deviated

from linearity at very high doses (135-150 mg/m2). This study was not specifically designed to address the question of pharmacokinetic linearity, and according to the authors the observed dose-dependency did not bear a predictive value for more severe acute toxicities. A more recent study [29] has specifically addressed the problem of dose-dependency of EPI pharmacokinetics in the range of 40 to 135 mg/m2 in a large patient population. The results indicate that the plasma exposure to EPI and its metabolites is linearly related to the dose. Clinical toxicology and pharmacology The spectrum of toxicity associated with EPI is qualitatively identical to that caused by DOX. The most prominent acute side effect is hematological. Nausea and vomiting, hair loss and oral mucositis represent other common side effects [11,12]. The key point about toxicity in humans is whether EPI is less toxic than DOX as it is in animals. The initial phase I evaluation of EPI defined a range of tolerable doses at 70 to 90 mg/m2 every three weeks, and suggested that EPI was better tolerated because it produced less myelosuppression, less nausea and vomiting as well as less hair loss than DOX [11, 12]. Since preclinical studies showed that EPI had a better therapeutic index than DOX [22], the two anthracyclines were administered at equimolar doses to exploit the tolerability advantage of EPI in a subsequent phase II study [30]. In view of the structure similarities of EPI and DOX, this approach led to the conclusion that the two anthracyclines should be used at equal doses on a mg per m2 basis. This interpretation did not take into account that the preliminary data of our phase I study fell short of formally defining the maximum tolerated dose (MTD) of EPI [11]. Thus, an appropriate evaluation of the comparative human toxicology of EPI and DOX requires further clarification. It is now established that the reference dose for the use of EPI as single agent is about 120 mg/m2 by i.v. bolus administration every three weeks, while MTD ranges from 165 to 180 mg/m2 [31-33]. The administration of EPI at the new reference dose is associated with hematological and extra hematological side effects which are not only qualitatively but quantitatively close to those documented following the administration of 90 mg/m2 of DOX every three weeks. At higher doses, oral and intestinal mucositis induced by EPI become frequent and severe enough to concur with bone marrow suppression to the definition of the limiting toxicity [33]. A meaningful comparison of both toxicity and antitumor activity between the two anthracyclines should then take into account whether the doses selected for any given trial are equimolar or equitoxic. When the two drugs are compared at equimyelotoxic doses, the incidence and severity of non cardiac toxicities become superimposable [34, 35].

361

Cardiac toxicity Preclinical and clinical evaluations of new anthracyclines have traditionally given special emphasis to the problem of cardiac toxicity [7, 36]. Initial studies in animals have indicated that EPI caused similar but less myocardial damage than DOX [22]. The latter observation was actually the major reason for starting the clinical evaluation of the analog. The clinical experience has confirmed that EPI causes cumulative dose dependent heart failure and cardiac morphological alterations to the myocardium which appear identical to those described for DOX, and mainly consist of dilatation of sarcoplasmic reticulum, myofibrillar loss and interstitial fibrosis [37, 38]. However, a number of reports have now established that the cardiotoxic potency of EPI is significantly lower than that of the parent anthracycline [37-39]. The initial appreciation of EPI's cardiotoxic potential has come from functional studies of left ventricular ejection fraction (LVEF) and electrocardiographs monitoring. These procedures bear low sensitivity and specificity [21] and consequently little value as predictors of subsequent clinical manifestations due to cumulative myocardial toxicity [37]. Morphologic evaluation from endomyocardial biopsies remains the most sensitive, specific and unquestioned method of assessing heart damage by anthracyclines [38,40, 41]. For doxorubicin, a close similarity exists between the arbitrary dose limitations estimated from morphological studies and the cumulative dose limits deduced from the prevalence of patients who actually developed congestive heart failure [41]. Two accurate morphologic studies on small series of patients confirmed the lower cardiotoxic potential of EPI compared to DOX [38, 39]. On morphologic examination, the degree of cardiac damage produced by EPI starts to increase with total cumulative doses exceeding 450 mg/m2. These studies concluded that the relative cardiotoxic potency of EPI is about 0.6 of that of DOX. In other words, an equal risk of developing clinical cardiomyopathy is associated with the administration of cumulative doses of 550 mg/m2 of DOX and 900 mg/ m2 of EPI, respectively. However, on clinical grounds the 3% probability of developing congestive heart failure is at 450 mg/m2 for DOX and 900 mg/m2 for EPI, respectively [42] (Fig.2). The data on large series of patients would indicate the relative cardiotoxic potency of EPI is 0.5 of that of DOX. Thus, the latter figure will be held as the reference equivalence for comparing the clinical cardiotoxic potency of the two anthracyclines. From our previous discussion on equimyelotoxic doses, the relative myelotoxic potency of EPI is about 0.75 of that of DOX. Taking into account the lower cardiotoxic potency of EPI, the overall toxicological profile in humans indicates that the analog has a limited but significant advantage over DOX. Available data showing comparable antitumor activity of EPI and DOX at equimolar low doses in combination chemo-

J>frwMs4» PmiiMlillirfl—« (mg/tf » J DOX OT Fig. 2. Cumulative probability of developing congestive heart failure versus total cumulative dose of DOX ( — ) and EPI ( ), respectively. From ref. 42, with permission.

therapy would then suggest a significantly better therapeutic index for the analog [43, 44]. This conclusion must be accepted with caution in view of conflicting reports from some randomized studies that show equal activity at equimyelotoxic rather than at equimolar doses of the two anthracyclines [34, 35]. The latter studies would support the concept that the better therapeutic index of EPI would be limited only to its comparatively lower cardiac toxicity in humans. Such an advantage could be even greater when this anthracycline is administered concurrently or in sequence with radiotherapy. The cardiotoxic synergism of DOX and chest wall radiation is well known [45, 46]. For EPI, instead, a recent report in patients with small cell lung cancer seems to indicate that such a synergism is not clinically significant even at high cumulative doses (660 mg/m2) of the anthracycline [47]. Clinical activity in various malignancies Breast cancer

Due to its clinical importance, breast cancer, as in the past acute leukemias and malignant lymphomas, has now become one of the tumor models to test the relative merits of new drugs in solid tumors. Tables 1 and 2 report the essential therapeutic results of treatments delivered at equimolar or at equimyelotoxic doses when EPI was tested vs DOX either as single agent every three weeks or as part of a drug combination containing fluorouracil and cyclophosphamide (FAC and FEC). The findings showed a discrepancy consistently observed in various clinical trials in which the analog was used at doses ranging from 50 to 90 mg/m2. Although in advanced breast cancer DOX and EPI appeared equally active, the identical

362 Table I. Regimens with doxorubicin vs. epirubicin in advanced breast cancer. Results in selected randomized studies of anthracyclines delivered at equimolar doses. Dose in mg/m2

Patients evaluable

DOX Bonfante |30J French Group |43] Italian Group |44|

42 230' 443 b

EPI

DOX

75 50 50

75 50 50

Median duiration in months

CR + PRin% EPI

52 52 56.5

62 50.4 53.6

DOX

EPI

13 10 9

11 9 7

• 20% had received prior adjuvant CMF. h 28% had received prior adjuvant CMF and 17% some chemotherapy for advanced disease. Table 2. Single agent epirubicin in advanced breast cancer. Results in selected randomized studies of anthracyclines delivered at equimyelotoxic doses. First author

Perez [34) Jain [35] Hortobagyi [48]

Patients evaluable

138 52 48

Prior chemotherapy No Yes Yes

Dose in mg/m2

CR + PRin%

DOX

DOX

60 60 60°

EPI 90 85 90°

47 25 28

Median duration in months EPI 49 25 26

DOX

EPI

10 7 5

8 12 4

" Through a 48 h. continuous infusion. CR: complete remission. PR: partial remission.

therapeutic result was achieved by equimolar doses in some case series (Table 1), and by equimyelotoxic doses in other series (Table 2). Despite treatment findings were clearly dependent on prior vs. no prior chemotherapy, it is worth of note that by reducing the dose of anthracycline in the combination regimens, multidrug treatment failed to yield superior outcome compared to either DOX or EPI given alone. At equimolar doses there was a rather consistent evidence that EPI induced a comparatively lesser degree of severe myelosuppression as well as nausea and vomiting. In contrast, the percentage of hair loss was almost superimposable for DOX and EPI regardless of equimolar or equimyelotoxic doses. Cardiac effects, expressed as percent fall in LVEF and congestive heart failure, were lesser for EPI than DOX when similar cumulative doses were given. In a pilot study, Bonadonna et al. [49] tested FAC and FEC as primary (neoadjuvant) chemotherapy. The comparative complete plus partial remission rates were 81% vs. 76% with no difference, at the doses administered (DOX 50 mg/m2, EPI 75 mg/m2), as far as acute and delayed toxicity was concerned. Recent reports have emphasized the delivery of EPI through high doses every 3 to 4 weeks as well as through low weekly or biweekly doses. Among the high-dose group, Bezwoda et al. [50] treated 18 patients previously given DOX but not necessarily resistant to DOX. The overall response rate (62% for a median duration of 5.8 months, CR 1) was related to the starting dose level (110 mg/m 2 :1 of 4 patients, 120140 mg/m2: 9 of 11, 150 mg/m2: 3 of 3). Grade 2-3 myelosuppression was universal and in one patient fatal neutropenia was documented. Despite prior DOX

chemotherapy, in only one case there was an important LVEF fall after a cumulative dose of 550 mg/m2 of DOX followed by 900 mg/m2 of EPI. Carmo Pereira et al. [31] tested 120 mg/m2 for 10 cycles in 40 patients five of whom had received prior adjuvant chemotherapy. The response rate was 65% (CR 22.5%) for a median duration of 7 months. Grade 2-3 leukopenia was 35%, congestive heart failure was documented in 2 of 5 patients showing LVEF fall > 15%. Treatment efficacy of low dose epirubicin was evaluated in two studies. After the administration of approximately 12 mg/m2/week, Jones [51] achieved a 43% response rate for about five months. Toxicity was mild and 35% of cases never experienced side effects. Gundersen et al. [52] compared the biweekly infusion of EPI over three hours (50 mg per dose to 68 patients) to the weekly i.v. bolus administration of DOX (20 mg per dose to 81 patients). The response rate was 22% vs. 36% for a median duration of 10 and 9 months, respectively. In a prospective randomized study, Scottish investigators [53| tested 'low' (50 mg/m2) vs. 'high' dose EPI (100 mg/m2) every three weeks plus prednisolone (50 mg for five days) in a total of 202 evaluable patients. The response rate and its median duration were superior in the 'high' dose arm (42% vs. 23% for 7.5 vs. 4 months, respectively); total survival was superimposable between the treatment groups (10.6 vs. 10.5 months). At the time of this review, only a few data are available on the efficacy of EPI-containing regimens in an adjuvant situation [54, 55]. At present, the only randomized comparison of standard CMF (cyclophospha-

363

mide, methotrexate, fluorouracil) vs. FEC (fluorouracil, epirubicin, cyclophosphamide) was undertaken by the International Collaborative Cancer Group [54]. In a total of 585 evaluable premenopausal node positive women, the five year results for the entire patient series failed to detect a difference between treated groups. However, there was a superiority of FEC over CMF limited to the subset of women with 1 to 3 positive axillary nodes. Hematological toxicity was about the same in both treatment arms.

remission before radiotherapy was achieved in 76%. The toxic effects of EBVP also included grade 3 alopecia in half patients. A German group [65] has developed the EBOEP regimen consisting of epirubicin (30 mg/m2 on days 1 and 8), bleomycin, vincristine, etoposide and prednisone, all recycled every three weeks. After four cycles followed by radiotherapy, the complete remission rate in 44 patients with stage I-IIIB was 100% with a 3-year relapse free survival of 90% in previously untreated patients. The authors also reported no iatrogenic damage on spermatogenesis. Another three weekly regimen is the Milan VEBEP: etoposide, Malignant lymphomas epirubicin (40 mg/m2 on days 1 and 2), bleomycin, During Phase I-II studies in non-Hodgkin's lymphomas cyclophosphamide and prednisone. After eight cycles, (all histologies) refractory to many conventional agents, radiotherapy is delivered through involved fields (36 but not necessarily resistant to DOX, EPI showed a re- Gy). In 43 evaluable patients (stage IIB-IV) the commarkable response rate (58%—75%) in limited case plete remission rate was 95%. With a median follow-up series given 40 up to 180 mg/m2 [12, 56, 57]. Of par- of 12 months, three patients have relapsed. The results ticular interest are the data reported by Case et al. [57]. of all the above mentioned studies clearly indicate that Nineteen previously treated patients (10 with DOX) in Hodgkin's disease EPI can be an effective substitute were given three dose levels: 120, 150, 180 mg/m2. for DOX in the achievement of complete remission. As Total response rate occurred in 58% (CR 2) regardless of today, the follow-up remains too short to provide of prior DOX. A >10% fall in LVEF was documented meaningful conclusions in terms of curative rate and in seven patients, four of whom had previously received long-term iatrogenic morbidity. DOX. As far as combination chemotherapy is concerned, Soft tissue sarcomas the only two randomized studies providing some detailed information are those which tested CHOP For more than 20 years DOX has been considered the (cyclophosphamide, doxorubicin, vincristine and pred- single most effective drug for all histologic subtypes of nisone) vs. CEOP, a similar dose regimen substituting soft tissue sarcomas [3, 5, 6]. In 1987 the EORTC Soft EPI for DOX at almost equimolar doses [58, 59]. Al Tissue and Bone Sarcoma Group [66] has reported the Ismail et al. [58J randomized 50 high grade lymphomas results of a randomized study of DOX vs. EPI delivered with stage III and IV without detecting significant dif- at the dose of 75 mg/m2 every 3 weeks in a total of 167 ference in the comparative complete and partial remis- evaluable patients. Using equimolar doses the response sion rates, mean duration of response and survival. De rate was slightly superior in the DOX-treated group Lena et al. [59] included in the study 60 patients (all (25%) vs. that of EPI-treated group (18%, P - 0.33), stages) with 'poor prognosis' and added bleomycin to although there was an equivalent time to disease proboth regimens as well as radiotherapy to chemotherapy gression (3.5 vs. 3.0 months, respectively). The comin stages I and II. Tumor response and treatment out- parative toxicity showed that leukopenia and hair loss come were essentially the same but decrease in the were significantly more pronounced after DOX. In LVEF was more frequently documented in the DOX- more recent years, EPI replaced DOX in various comcontaining regimen. Similar therapeutic results with bination regimens such as those containing ifosfamide, CEOP in advanced lymphomas with 'poor histology' another drug effective in soft tissue sarcomas. The were also reported in 28 patients treated by Cornelia et highest response rate was documented in patients given al. [60]. More recently, EPI has replaced DOX in vari- EPI at doses >90 mg/m2 for a median duration ranging ous combination regimens, such as MECOP-B [61] and from 4 to 6 months (Table 3). Since in many cases of POCE [62], and yielded similar therapeutic results as soft tissue sarcomas the time to achieve the maximum previously reported for the parent compound. tumor shrinkage is long, and thus requiring high cumudrug doses, EPI (at the appropriate doses in mg/ In Hodgkin's disease, the efficacy of alternating lative 2 m ) appears preferable to DOX. MOPP and ABVD [63] has somewhat delayed the design of new treatment regimens. During the last few years, a novel combination consisting of epirubicin (70 mg/m2), bleomycin, vinblastine and prednisone (EBVP) has been tested by French groups to reduce toxicity. The most recent and documented experience is that reported by Hoerni et al. [64] in 100 consecutive patients with favorable or unfavorable stage I-IIIA. EBVP was delivered in full dosage every three weeks; after the third cycle patients were irradiated. Complete

Lung cancer With the exception of the small cell subtype, lung cancer is a group of malignancies which are notoriously chemoresistant to the vast majority of available drugs. Actually, some clinicians are still in doubt whether to treat or not to treat most patients presenting with advanced non-small cell bronchogenic carcinoma.

364 Table 3. Epirubicin in combination regimens with ifosfamide (IFO) for advanced soft tissue sarcomas. First author

Patients

Dose in mg/m2

Toma [67)

15

EPI IFO

60-75 6000

40

CR plus PR (%)

Chevallier [68|

27

EPI IFO

100-130 5000

48

Elli [69|

22

EPI IFO

75

23

4500

Casali |70]

48

EPI IFO

90

48

7500

DTIC" 900 u

Dacarbazine.

by EORTC group [79] using the dose of 110 mg/m2 every three weeks. Gastrointestinal cancer

In the various malignancies originated in the gastrointestinal tract (esophagus, stomach, colon-rectum, pancreas and liver) DOX, alone or in combination, has always yielded modest results. The review of the medical literature allows to conclude that, in general, when EPI has replaced DOX the therapeutic results remained superimposable. The most relevant experience published in esophageal cancer is that of Kolaric et a). [80|. The authors have treated 33 patients combining EPI (150 mg/m2 in two days) with irradiation. The overall response rate was 70% (CR 11, PR 12) at the expense of moderate toxicity. In gastric cancer, single agent chemotherapy with EPI was tested at the dose of 75-100 mg/m2 every 3 weeks [14]. Partial response was documented in 17%27% of cases with a median duration of 3 to 4 months. The administration of EPI through continuous intravenous infusion (6 mg/m2 for 3 weeks) has yielded in 27 evaluable patients a response rate of 15% (CR 2) with a median duration of about one year at the expense of minimal toxicity [81]. Table 5 reports the therapeutic results of FEM (fluorouracil, epirubicin, mitomycin C) in advanced gastric cancer. The findings were similar to those reported with FAM, but the authors concurred that FEM was well tolerated. In colorectal cancer the overall response rate following EPI alone (75-150 mg/m2) was less than 20% [14]. The drug (20 mg/m2/week) was also combined with other agents such as methotrexate, fluorouracil and folinic acid [86]. In 90 patients the response rate was

Reported findings are also often contradictory for a number of reasons, i.e. patient selection and doses of all drugs included in given combination regimens. In this review, we shall concentrate our efforts on a few specific points, namely antitumor activity of EPI as single agent, the feasibility of high doses (i.e. > 100 mg/ m2) when EPI was delivered alone or in combination, and cardiac toxicity. In small cell carcinomas, the response rate to EPI as single agent is doubled when the anthracycline is given at doses ranging from 100 to 140 mg/m2 J71-74]. When high doses of EPI were delivered, severe granulocytopenia was documented in approximately 80% of patients and hair loss occurred in almost all patients. By contrast, oral mucositis was observed only in a minority of patients (16%) and in no cases was a decline in LVEF > 20% from baseline documented. EPIcontaining combinations (with EPI usually delivered at the dose of 50 mg/m2) are yielding a response rate similar to that known for DOX-containing polydrug regimens (Table 4). In non-small cell carcinomas, single agent EPI yield- Table 4. Epirubicin in untreated small cell lung cancer. Cumulative ed a response rate in less than 10% of patients when data from representative studies. doses were < 90 mg/m2 [75]. When administered at No. CR CR+PR Median higher doses (105-180 mg/m2), EPI induced a supericases % % survival (mo) or response rate (20%-25%) but no complete remissions |33, 76, 77]. Because of poor chemosensitivity of EPI as single agent non-small cell carcinomas, also the various drug com4 doses <100 mg/m2 101 21 4-15 1 I 48 doses 100-140 mg/m2 155 7-18 binations containing EPI failed to definitely improve total response rate (18%-56%) and median survival EPl-containing duration (5—11 months) over other combination regicombinations 419" 30 82 10-18 mens [14]. 1 Case series with >50 evaluable patients.

Pleural mesothelioma Although the optimal treatment of this malignancy remains to be defined, DOX has been reported among the most effective single agents. Two studies have been recently published in Europe on the efficacy of EPI in advanced mesothelioma. Italian investigators [78] have achieved one partial response in 21 evaluable patients given 75 mg/m2. A higher response rate (7 of 48 or 15%) for a median duration of 9 months was reported

Table 5. Fluorouracil, epirubicin and mitomycin C (FEM) in advanced gastric cancer. First author

Patients evaluable

CR + PR

Median survival (mo)

Flechtner (82| Queisser |83| Iton |84| Roth |85]

25 29 12 39

28 28 25 33

5.3 6.2 6.3 5.3

365

29% (CR 6%). However, it is difficult to dissect in this study the real contribution of EPI. In pancreatic cancer, EPI alone at doses ranging from 60 to 90 mg/m2 yielded a response rate of less than 10%. Wils [87] reported complete and partial remission in 25% of cases given EPI at 120 mg/m2 every 4 weeks. The EORTC group documented partial response in 12% to 14% of cases when EPI was combined with either fluorouracil [88] or ifosfamide [89]. In advanced hepatocellular carcinoma single agent EPI (75-90 mg/m2) yielded a response rate ranging from 9% to 17%. No definite improvement was observed following hepatic intra-arterial infusion at 40 to 90 mg/m2 every 3-4 weeks [90]. Head and neck cancer With the exception of undifferentiated nasopharyngeal cancer, DOX has never been reported to be particularly active in head and neck cancer. The same applies to EPI. In particular, in 27 patients given 75 mg/m2 every three weeks Hu et al. [91] have documented a response rate of 52% (CR 6). EPI was also delivered through various combination regimens, the most important of which is that reported by the Institut Gustave Roussy [92]. The regimen BEC (bleomycin, epirubicin (80 mg/ m2), cisplatin) induced a 78% complete plus partial response in 47 evaluable patients (27 previously treated with chemotherapy) presenting with advanced disease. As neoadjuvant chemotherapy, BEC achieved a 98% total response with 63% complete remission in 51 patients. Of interest, after a mean follow-up of 19 months, 42 patients (78%) remain alive and disease free. Grade 3-4 hematologic toxicity was observed in 32% of cases after the third cycle. In the ongoing study at the Milan Cancer Institute after primary chemotherapy with EPI (70 mg/m2) plus cisplatin (100 mg/m2) we have documented response in 17 of 21 patients (CR 4).

Table 6. Epirubicin combined with cisplatin in advanced ovarian cancer. Representative case series. First aulhor

Palients evaluable

Dose in mg/m 2

CR%

CR+PR %

EPI 60 CDDP 50

19

61

EPI 60 C D D P 60. C T X 600

36

59

Martoni [93]

59

Eckhardt [y4|

1(17

Homesley |95|

62

EPI 75 C D D P 60

15

42

54

D O X 60 C D D P 60

31

55

24

EPI 50 CDDP 60, CTX 600

58

79

27

D O X 40 CDDP 60. CTX 600

33

63

Bezwoda [96|

CDDP — cisplatin, CTX — cyclophosphamide.

the National Prostate Cancer Project (NPCP) classifies stabilization as favorable response. EORTC investigators have tested EPI weekly (12 mg/m2) in 33 patients refractory to endocrine therapy. The response rate was 12% according to the WHO criteria and 54% according to the NPCP criteria, respectively [97]. In a prospective randomized study, Tveter et al. [98] compared estramustine phosphate (280 mg twice daily) vs. EPI (20 mg/week) either alone or combined with medroxyprogesterone acetate (1000 mg/day) in patients resistant to conventional hormone treatment. Patients receiving EPI showed decreased bone pain; EPI plus medroxyprogesterone also showed improved performance status and response duration. Another important study is that of Pummer [99]. Following orchiectomy, 117 patients with stage D disease were randomized to receive flutamide alone or flutamide plus EPI (25 mg/m2 for 18 weeks). Combined treatment with EPI significantly improved percent of response, remission duration and total survival.

Ovarian cancer Single agent EPI (75-150 mg/m2) in advanced ovarian carcinoma refractory to a number of first line regimens with or without cisplatin yielded objective tumor response in about 20% of cases [14|. Probably due to patient heterogeneity, a dose-response effect could not be assessed. Table 6 summarizes the response rate in four patient series treated with EPI and cisplatin. The results are roughly similar to those being achieved with cisplatin plus cyclophosphamide. Furthermore, in two series randomizing EPI vs. DOX [95, 96] treatment findings were almost superimposable. However, less frequent cardiac effects were reported in the EPI treated group. Prostatic carcinoma In this disease the response to chemotherapy is notoriously difficult to assess in most patients. Furthermore,

Transitional bladder carcinoma The data on the endocavitary administration of EPI are, at present, difficult to assess mainly because of patients and treatment heterogeneity [14]. Single agent intravenous chemotherapy with EPI at 90 mg/m2 every three weeks yielded in the EORTC experience a 15% response rate in 33 patients [100]. In the various drug combinations utilized, it is important to quote the five studies in which in the M-VAC regimens EPI substituted for DOX (Table 7). The response rate, ranging from 37% to 83%, probably reflects disease extent prior to chemotherapy. It is important to point out that even when EPI was administered at the dose of 60 mg/m2 [105] the frequency of oral mucositis was lower (4%) compared to DOX (27%).

366 Table 7. M-VAC-like regimens containing epirubicin in advanced transitional bladder cancer. First author

EPI mg/m 2

Patients evaluable

Disease extent

CR+PR %

Riither [101]

30

58

72

Frassoldati [102] Goldfarb 1103]

30 50

23 43

Lorusso 110-4] Passalacqua |105|

30 60 vs DOX 30

16 54

Locally advanced and disseminated Locally advanced Locally advanced and disseminated Disseminated Locally advanced

83 72 37 44 54

Conclusions When epirubicin appeared in the horizon of cancer chemotherapy a major conceptual revision was underway about the interrelation of dose, dose intensity and tumor response. Steep dose-response curves were clearly confirmed from laboratory models: a small increase in drug dose resulted in a disproportionally increase in tumor cell kill, and in many experimental animal systems the log kill was greater for the higher dose [106, 107]. Moreover, the considerable improvement achieved in numerous malignancies through combination chemotherapy, including systemic adjuvant treatment in bone sarcomas and in stage n breast cancer, stressed that dose was a critical factor in cancer chemotherapy [108-111]. The process of revision of doses and schedules for individual chemotherapeutic agents benefited both from new effective supportive measures to treat acute drug toxicity and the wide application of monitor systems to detect early organ dysfunction. More recently, the strategy to better exploit the basic pharmacological principle of dose response effect also included the administration of mega doses of chemotherapy with hematopoietic growth factors and autologous bone marrow support. The above mentioned conceptual evolution in the medical treatment of cancer has profoundly influenced the concurrent clinical development of epirubicin. In fact, the first generation of trials was based upon doses defined in 1979 and 1980 when dose escalation during phase I studies was discontinued at 90-100 mg/m2, i.e. in accordance to the safety criteria adopted at that time for ambulatory treatment without intensive monitory of patients. Early in 1980s, the analog was more frequently compared to doxorubicin on equimolar rather than on equimyelotoxic bases. The redefinition of maximal tumor dose according to modern concepts of acceptable risks and manageable complications allowed to initiate a number of new studies with epirubicin. As a consequence, these studies eventually led to a more appropriate use of this anthracycline. The review of representative medical literature of the past decade allows to conclude that in animal models as well as in various malignancies of adults epirubicin possesses a broad spectrum of antitumor activ-

ity as doxorubicin but a reduced potential for cardiotoxicity. As documented for its parent compound, EPI proved to be particularly effective in all stages of breast cancer, malignant lymphomas, bone and soft tissue sarcomas, undifferentiated nasopharyngeal carcinoma and small cell lung cancer. No consistent benefit could be demonstrable in malignancies usually resistant to doxorubicin such as intestinal and renal carcinomas. It is well known that the prolonged administration of doxorubicin is limited by cumulative dose-dependent chronic cardiotoxicity. Several approaches have been proposed to reduce the myocardial injury. The use of potentially cardioprotective measures such as delivery of vitamin E, N-acetylcysteine, a bispiperazinedion as ICRF-187, and even liposomal encapsulation of DOX have been shown to decrease the cardiotoxicity in experimental animals and in humans [112]. However, present data have not been sufficiently promising to warrant their large scale clinical application. Thus, empiric dose restriction not to exceed 450 mg/m2 is still widely practiced to limit within 3% the risk of symptomatic left ventricular or biventricular failure. In fact, the incidence of congestive heart failure raises steeply after higher cumulative doses. As far as epirubicin is concerned, published data indicate that this analog produces less severe morphologic and functional changes in the myocardium than does doxorubicin. This represents a major advantage since modern chemotherapy is frequently delivered with curative intent. Cumulative doses of EPI above 900 mg/m2 potentiate a considerable risk of cardiotoxicity and death from congestive heart failure but there is a minimal risk of developing heart failure when patients are given EPI at cumulative doses of 450 to 900 mg/m2 (see Fig.2). Since subclinical myocardial damage has been documented at doses in excess of 450 mg/m2 [38], the cumulative safe dose of EPI should not exceed this value in an adjuvant or neoadjuvant situation and when radiation therapy to mediastinum or left breast is part of the sequential combined modality strategy. Although LVEF is probably the most reliable non invasive method to measure the myocardial contractile function, the clinical implications of its changes after anthracycline therapy are still controversial because of rather low sensitivity and specificity in predicting congestive heart failure [37,112]. Radionuclide ventriculography should be obtained as part of the diagnostic work-up before delivering the drug in patients clinically suspected of cardiac disease and when the clinical situation requires the administration of EPI at cumulative doses higher than 450 mg/m2. A key point in the clinical management with EPI is its optimal dose regimen, particularly in untreated ambulatory patients with no prior extensive irradiation and/or age less than 65 years. When administered as single agent chemotherapy, 110 to 120 mg/m2 can be delivered in a standard every three week schedule. This dose is approximately the equivalent of 90 mg/m2 of DOX. Higher doses could be reserved to carefully

367

selected hospitalized patients. In a combination regimen including other myelosuppressive drugs the optimal dose of EPI is 75 mg/m2 in ambulatory patients. Presumably, since also peak plasma levels have an important relationship to both DOX and EPI cardiotoxicity, the standard dose of 110 to 120 mg/m2 should probably be divided in two to three fractions. As previously demonstrated for DOX, administration of EPI by prolonged intravenous infusion (48 to 96 hours) can be relatively cardiac sparing. Also the weekly drug administration (20—25 mg per patient) can be minimally toxic to the bone marrow without decreasing significantly its antitumor activity. Prolonged infusion and weekly administration can be reserved to elderly patients, in the presence of reduced marrow reserve from prior radiation or chemotherapy as well as in the presence of impaired liver function. In conclusion, epirubicin is a steroisomer of doxorubicin which retains broad antitumor activity but, at equimolar doses, is a less toxic compound with regard to both general toxicity and cardiac injury. Thus, in future studies there is the possibility for further refinements in developing other anthracycline analogs with improved cost-benefit ratio, particularly as far as potential cardiac damage is concerned. Preclinical and clinical studies have indicated that this can be possible by structural alterations of the anthracycline molecule. References 1. Bonadonna G, Monfardini S, De Lena M et al. Clinical evaluation of adriamycin, a new antitumor antibiotic. Br Med J 1969; 3: 503-6. 2. Bonadonna G, Monfardini S, De Lena M et al. Phase I and preliminary Phase II evaluation of adriamycin (NSC 123127). Cancer Res 1970; 30:2572-82. 3. Bonadonna G, Beretta G, Tancini G et al. Adriamycin (NSC123127) studies at the Istituto Nazionale Tumon, Milan. Cancer Chemotherapy Rep 1975; 6: 231-45. 4. Tbrmey D. Adriamycin (NSC-123127) in breast cancer. An overview of studies. Cancer Chemoth Rep 1975;6: 319-27. 5. O'Bryan RM, Luce JK, Talley RW et al. Phase II evaluation of adriamycin in human neoplasia. Cancer 1973; 32: 1 - 8 . 6. Blum RH, Carter SK. Adriamycin. A new anticancer drug with significant clinical activity. Ann Intern Med 1974; 80: 249-59. 7. Praga C, Beretta G, Vigo PL et al. Adriamycin cardiotoxicity: A survey of 1273 patients. Cancer Treat Rep 1979; 63: 827-34. 8. Arcamone F. Properties of antitumor anthracyclines and new developments in their application: Cain Memorial Award Lecture. Cancer Res 1985; 45: 5995-9. 9. Arcamone F. Analogs modified in the aminosugar residue. In Arcamone F (ed): Doxorubicin. New York: Academic Press; 1981: 196-258. 10. Di Marco A, Casazza AM, Dasdia T et al. Changes in the activity of daunorubicin, adriamycin and stereoisomers following the introduction or removal of hydroxyl groups in the amino sugar moiety. Chem Biol Interact 1977; 19:291-302. 11. Bonfante V, Bonadonna G, Villani F et al. Preliminary phase I study of 4'-epi-adriamycin. Cancer Treat Rep 1979; 63: 915-18. 12. Bonfante V, Bonadonna G, Villani F et aJ. Preliminary clinical experience with 4-epi-doxorubicin in advanced human neoplasia. Recent Results Cancer Res 1980; 74:192-9.

13. Bonadonna G (Ed). Advances in anthracycline chemotherapy: Epirubicin. Masson, Milan 1984. 14. Mouridsen HT, Alfthan C, Bastholt L et al. Current status of epirubicin (Farmorubicin) in the treatment of solid tumors. Acta Oncol 1990; 29:257-85. 15. Delozier T, Vernhes JC. Etude comparative de l'adriamycine, Tepirubicine e la mitoxantrone dans le cancer du sein. Revue de la litterature. Bull Cancer 1991; 78: 1013-25. 16. Myers CE, Gianni L, Zweier J et al. The role of iron in adriamycin biochemistry. Fed Proc 1986; 45:2792-7. 17. Capranico G, Zunino F, Kohn KW et al. Sequence-selective topoisomerase II inhibition by anthracyclines derivatives is SV40DNA: Relationship with DNA binding affinity and cytotoxicity. Biochemistry 1990; 29: 562-9. 18. Gianni L, Corden BJ, Myers CE. The biochemical basis of anthracycline toxicity and antitumor action. In Hodgson E, Bend JR, Philpot RM (eds): Reviews in biochemical toxicology 5. New York, Amsterdam, Oxford: Elsevier Biomedical; 1983: 1-82. 19. Myers CE, Chabner BA. Anthracyclines. In Chabner BA, Collins JM (eds): Cancer Chemotherapy. Principles & Practice. 2nd ed. Philadelphia: J.B. Lippincott Co.; 1990: 356-81. 20. Gianni L, Vigano L, Lanzi C et al. Role of daunosamine and hydroxyacetyl side chain in reaction with iron and lipid peroxidation by anthracyclines. J Natl Cancer Inst 1988; 80: 1104-11. 21. Weenen H, Lankelma J, Penders PGM et al. Pharmacokinetics of 4'-epidoxorubicin in man. Invest New Drugs 1983; 1: 598604. 22. Casazza AM, Giuliani FC. Preclinical properties of epirubicin. In Bonadonna G (ed): Advances in anthracycline chemotherapy: Epirubicin. Milano: Masson; 1984: 31—40. 23. Moscow JA, Cowan KH. Multidrug resistance. J Natl Cancer Inst 1988; 80: 14-20. 24. Camaggi CM, Comparsi R, Strocchi E et al. Epirubicin and doxorubicin comparative metabolism and pharmacokinetics. A cross-over study. Cancer Chemother Pharmacol 1988; 21: 221-8. 25. Mross K, Maessen P, Van der Vijgh WJF et al. Pharmacokinetics and metabolism of epidoxorubicin and doxorubicin in humans. J Clin Oncol 1988; 6: 517-26. 26. Eksborg S, Stendahl U, Lqnroth U. Comparative pharmacokinetics of adriamycin and 4'-epiadriamycin after their simultaneous administration. Eur J Clin Pharmacol 1986; 30: 629-35. 27. Robert J, Vrignaud P, Nguyen-Ngoc T et al. Comparative pharmacokinetics and metabolism of doxorubicin and epirubicin in patients with metastatic breast cancer. Cancer Treat Rep 1985; 69:633-40. 28. Tjuljandin SA, Doig RG, Sobol MM et al. Pharmacokinetics and toxicity of two schedules of high dose epirubicin. Cancer Res 1990; 50: 5095-101. 29. Jakobsen P, Steiness E, Bastholt L et al. Multiple-dose pharmacokinetics of epirubicin at four different dose levels: Studies in patients with metastatic breast cancer. Cancer Chemother Pharmacol 1991; 28:63-8. 30. Bonfante V, Ferrari L, Brambilla C et al. New anthracycline analogs in advanced breast cancer. Eur J Cancer Clin Oncol 1986; 22: 1379-85. 31. Carmo-Pereira J, Costa FO, Miles DW et al. High-dose epirubicin as primary chemotherapy in advanced breast carcinoma: A Phase II study. Cancer Chemother Pharmacol 1991; 27: 394-6. 32. Hickish T, Cunningham D, Haydock A et al. Experience with intermediate-dose (110-120 mg/m2) epirubicin. Cancer Chemother Rep 1989; 24:61-4. 33. Martoni A, Melotti B, Guaraldi M et al. Activity of high-dose epirubicin in advanced non-small cell lung cancer. Eur J Cancer 1991; 27: 1231-4. 34. Perez DJ, Harvey VJ, Robinson BA et al. A randomized comparison of single-agent doxorubicin and epirubicin as first-line

368

35. 36. 37. 38. 39. 40. 41. 42.

43.

44.

45. 46. 47.

48. 49.

50.

51. 52. 53.

54.

55.

cytotoxic therapy in advanced breast cancer. J Clin Oncol l99l;9:2148-52. Jain KK, Casper ES, Geller NL et al. A prospective randomized comparison of epirubicin and doxorubicin in patients with advanced breast cancer. J Clin Oncol 1985; 3: 818-26. Von Hoff DD, Layard MW, Basa P et al. Risk factors for doxorubicin-induced congestive heart failure. Ann Intern Med 1979; 91: 710-7. Nielsen D, Jensen JB, Dombemowsky P et al. Epirubicin cardiotoxicity: A study of 135 patients with advanced breast cancer. J Clin Oncol 1990; 8: 1806-10. Dardir MD, Ferraris VJ, Mikhael YS et al. Cardiac morphology and functional changes induced by epirubicin chemotherapy. J Clin Oncol 1989; 7: 947-58. Torti FM, Bristow MR, Lum BL et al. Cardiotoxicity of epirubicin and doxorubicin: Assessment by endomyocardial biopsy. Cancer Res 1986; 46: 3722-7. Billingham ME, Mason JW, Bristow MR et al. Anthracycline cardiomyopathy monitored by morphologic changes. Cancer Treat Rep. 1978; 62: 860-5. Bristow MR, Mason JW, Billingham ME et al. Dose-effect and structure-function relationships in doxorubicin cardiomyopathy. Am Heart J 1981; 102:709-18. Praga C, Trave F, Petroccione A. Anthracycline-induced cardiotoxicity and its relevance in cancer treatment. In Nimmo WS et al (eds): Clinical Measurement in Drug Evaluation. London: Wolfe Medical Publications Ltd; 1991: 131-42. The French Epirubicin Study Group. A prospective randomized phase III trial comparing combination chemotherapy with cyclophosphamide, fluorouracil and either doxorubicin or epirubicin. J Clin Oncol 1988; 6: 679-88. Italian Multicentre Breast Study with Epirubicin. Phase III randomized study of fluorouracil, epirubicin, and cyclophosphamide vs fluorouracil, doxorubicin, and cyclophosphamide in advanced breast cancer: An Italian multicentre trial. J Clin Oncol 1988; 6: 976-82. Fojardo CF, Eltnngham JR, Stewart JR. Combined cardiotoxicity of adriamycin and x-radiation. Lab Invest 1976; 34: 86-96. Merril J, Greco FA, Zimbler H. Adriamycin and radiation synergistic cardiotoxicity. Ann Intern Med 1975; 82:122-3 Macchiarini P, Chella A, Riva A et al. Phase II feasibility study of high dose epirubicin-based regimens for untreated patients with small-cell lung cancer. Am J Clin Oncol (CCT) 1990; 13: 495-500. Hortobagyi GN, Yap HY, Kau SW et al. A comparative study of doxorubicin and epirubicin in patients with metastatic breast cancer. Am J Clin Oncol (CCT) 1989; 12:57-62. Bonadonna G, Veronesi U, Brambilla C et al. Primary chemotherapy to avoid mastectomy in tumors with diameters of three centimeters or more. J Natl Cancer Inst 1990; 82: 1539-45. Bezwoda WR, Dansey R, Seymour L. High-dose 4'-epiadriamycin for treatment of breast cancer refractory to standard dose anthracycline chemotherapy: Achievement of second responses. Oncology 1990; 47: 4-8. Jones WG. Effective palliation of advanced breast cancer with weekly low dose epirubicin. Eur J Cancer Clin Oncol 1989; 25:357-60. Gundersen S, Kvinnsland S, Klepp O et al. Weekly adriamycin vs 4-epidoxorubicin every second week in advanced breast cancer. A randomized trial. Eur J Cancer 1990; 26:45-8. Habenshaw T, Paul J, Jones R et al. Epirubicin at two dose levels with prednisolone as treatment for advanced breast cancer The results of a randomized trial. J Clin Oncol 1991; 9: 295-304. Coombes RC, Bliss JM, Marty M et al. A randomised trial comparing adjuvant FEC with CMF in premenopausal patients with node positive resectable breast cancer. Proc Am Soc Clin Oncol 1991; 10:41 (Abstract). Sertoli MR, Pronzato P, Rubagotti A et al. A randomized

56. 57. 58.

59.

60.

61.

62.

63.

64.

65.

66.

67. 68. 69.

70. 71. 72.

73.

study of penoperative chemotherapy in primary breast cancer. In Salmon SE (ed): Adjuvant Therapy of Cancer VI. Orlando: WB Saunders Co.; 1990: 196-203. Lopez M, Di Lauro L, Ganzina F et al. Epirubicin in nonHodgkin's lymphomas. Am J Clin Oncol (CCT) 1985; 8: 151-3. Case DC Jr, Gams R, Ervin T et al. Phase I-II trial of highdose epirubicin in patients with lymphoma. Cancer Res 1987; 47:6393-6. Al-Ismail SAD, Whittaker JA, Gough J. Combination chemotherapy including epirubicin for the management of nonHodgkin's lymphoma. Eur J Cancer Clin Oncol 1987; 23: 1379-84. De Lena M, Maiello E, Lorusso V et al. Comparison of CHOP-B vs CEOP-B in 'poor prognosis' non-Hodgkin's lymphomas. A randomized trial. Med Oncol & Tumor Pharmacother 1989; 6: 163-9. Cornelia P, Abate G, Fiore M et al. Cyclophosphamide, epirubicin, vincristine and prednisone (CEOP) in the treatment of advanced non-Hodgkin's lymphomas with poor histology. Haematologica 1988; 73: 509-12. Ibrahim EM, Satti MB, Al-Idrissi HY et al. Non-Hodgkin's lymphoma in Saudi Arabia: Prognostic factors and an analysis of the outcome of combination chemotherapy only, for both localized and advanced disease. Eur J Cancer Clin Oncol 1988; 24: 391-401. O'Reilly SE, Hoskins P, Howdle S et al. POCE chemotherapy - Phase II trial in elderly patients with advanced stage diffuse large cell lymphoma. Proc Am Soc Clin Oncol 1992; II: 327 (Abstract). Bonadonna G, Valagussa P, Santoro A. Alternating non-crossresistant combination chemotherapy or mechlorethamine, vincnstine, procarbazine, and prednisone in stage IV Hodgkin's disease. A report of 8-year results. Ann Intern Med 1986; 104: 739-46. Hoerni B, Orgerie MB, Eghbali H et al. Novel combination of epirubicin, bleomycin, vinblastine and prednisone (EBVP II) before radical radiotherapy in localized stages (I-IIIA) of Hodgkin's disease. J Cancer Res Clin Oncol 1991; 117: 377-80. Mitrou PS, Klippstein T, Lautenschlager G et al. Etoposide and epirubicin containing chemotherapy plus radiotherapy in the treatment of early stages of Hodgkin's disease (HD). Proc Am Soc Clin Oncol 1992; 11:328 (Abstract). Mouridsen HT, Bastholt L, Somers R et al. Adriamycin versus epirubicin in advanced soft tissue sarcomas. A randomized phase ll/phase III study of the EORTC Soft Tissue and Bone SarcomaGroup. Eur J Cancer Clin Oncol 1987; 23: 1477-83. Toma S, Coialbu T, Biassoni L et al. Epidoxorubicin plus lfosfamide in advanced and/or metastatic soft-tissue sarcomas. Cancer Chemother Pharmacol 1990; 26: 453-6. Chevallier B, Leyvraz S, Oliver JP et al. Epirubicin and ifosfamide in advanced soft tissue sarcoma (ASTS): A phase II study. Eur J Cancer 1991; 27 (Suppl 2): S162 (Abstract) Elli A, Hernandez Moran JC, Pascoon G et al. Ifosfamide (IFO), 4 epidoxorubicin (4EPI) chemotherapy (CT) for advanced soft tissue sarcomas (STS). Proc Am Soc Clin Oncol 1991; 10: 350 (Abstract). Casali P, Pastorino U, Zucchinelli P et al. Epirubicin plus ifosfamide and dacarbazine (EID) in advanced soft tissue sarcomas. Ann Oncol 1992; 3 (Suppl 1 y. S125-S126. Banham SW, Henderson AF, Bicknell S et al. High dose epirubicin chemotherapy in untreated poorer prognosis small cell lung cancer. Respir Med 1990; 84: 241-4. Blackstein M, Eisenhauer EA, Wierzbicki R et al. Epirubicin in extensive small-cell lung cancer: A phase II study in previously untreated patients: A National Cancer Institute of Canada Clinical Trials Group Study. J Clin Oncol 1990; 8: 385-9. Eckhardt S, Kolaric K, Vukas D et al. Phase II study of 4'epidoxorubicin in patients with untreated, extensive small cell

369

74.

75. 76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

lung cancer. South-East European Oncology Group (SEEOG). Med Oncol & Tumor Pharmacother 1990; 7: 19-23. Macchiarini P, Danesi R, Manotti R el al. Phase II study of high-dose epirubicin in untreated patients with small-cell lung cancer. Am J Clin Oncol (CCT) 1990; 13: 302-7. Ettinger DS. New drugs in the treatment of non small cell lung cancer. Lung Cancer 1991; 7: 113-22. Wils J, Utama I, Sala L et al. Phase II study of high-dose epirubicin in non-small cell lung cancer. Eur J Cancer 1990; 26: 1140-1. Feld R, Wierzbicki R, Walde PLD et al. Phase I-II study of high-dose epirubicin in advanced non-small cell lung cancer. J Clin Oncol 1992; 10:297-303. Magri MD, Veronesi A, Foladore S et al. Epirubicin in the treatment of malignant mesothelioma: A phase II cooperative study. The North-Eastern Italian Oncology Group (GOCCNE) Mesothelioma Committee. Tumori 1992; 77:49-51. Mattson K, Giaccone G, Kirkpatric A et al. Epirubicin in malignant mesothelioma: A phase II study of the European Organization for Research and Treatment of Cancer, Lung Cancer Cooperative Group. J Clin Oncol 1992; 10; 824-8. Kolaric K, Roth A, Bau J et al. Combination of 4-epi-doxorubicin and irradiation. A new approach in the treatment of locoregionally advanced inoperable esophageal cancer. Tumori 1986; 72: 89-94. De Vries EG, Nanninga AG, Greidanus J et al. A phase II study of a 21 day continuous infusion schedule with epirubicin in advanced gastric cancer. Eur J Cancer Clin Oncol 1989; 25: 1509-10. Flechtner H, Queisser W, Heim ME et al. 5-fluorouracil, 4-epidoxonibicin, and mitomycin C (FEM) combination chemotherapy for advanced gastric carcinoma. A phase-II trial by the 'Chemotherapiegruppe Gastrointestinaler Tumoren (CGT)'. Onkologie 1987; 10:67-71. Queisser W, Flechtner H, Heim ME et al. Dose escalation and split course of 4-epidoxorubicin in combination chemotherapy (FEM II) of advanced gastric carcinoma. A phase-II trial the 'Chemotherapiegruppe Gastro-intestinaler Tumoren (CGT)\ Onkologie 1989; 12:202-6. Iton T, Sakata Y, Tamura Y et al. Effect of combination of 5-fluorouracil, epirubicin and mitomycin C (FEM) on advanced gastric carcinoma. Nippon Gan Chiryo Gakkai Shi 1989; 24: 1551-6. Roth A, Zupanc D, Luetic J el al. Open phase II with 5-fluorouracil, 4-epidoxorubicin and mitomycin C (FEM) in advanced gastric cancer. Tumori 1990; 76: 51-53. Pyrhonen SO, Kouri MO. Phase II study of epirubicin sequential methotrexate and 5-fluorouracil for advanced colorectal cancer. Eur J Cancer 1992; 28: 1828-32. Wils J, Bleiberg H, Blijam G et al. Phase II study of epirubicin in advanced adenocarcinoma of the pancreas. Eur J Cancer Clin Oncol 1985; 21: 191-4. Wils J, Bleiberg H, Dalesio O et al. An EORTC Gastrointestinal Group phase II evaluation of epirubicin combined with 5-fluorouracil in advanced adenocarcinoma of the pancreas. Eur J Cancer Clin Oncol 1987; 23: 1017-8. Wils J, Bleiberg H, Buyse M et al. An EORTC Gastrointestinal Group phase II evaluation of epirubicin combined with ifosfamide in advanced adenocarcinoma of the pancreas. EORTC Gastrointestinal Group. Eur J Cancer Clin Oncol 1989,25: 1119-20. Kajanti M, Pyrhonen S, Mantyla M et al. Intra-arterial and intravenous use of 4'epidoxorubicin combined with 5-fluorouracil in primary hepatocellular carcinoma. A randomized comparison. Am J Clin Oncol (CCT) 1992; 15: 37-40. Hu OY, Chang SP, Jame JM. Pharmacokinetic and pharmacodynamic studies with 4'-epi-doxonibicin in nasopharyngeal carcinoma patients. Cancer Chemother Pharmacol 1989; 24: 332-7. Gasmi J, Bachouchi M, Cvitkovic E et al. Nasopharyngeal carcinoma: a medical oncology viewpoint. The Gustave Roussy experience. Ann Oncol 1990; 1:245-53.

93. Martoni A, Bellucco A, Canova N et al. Four-year analysis of platinum and anthracycline combination for ovarian cancer. Oncology 1989; 46: 109-16. 94. Eckhardt S, Szanto J. Activity of epirubicin in combination chemotherapy of advanced ovarian cancer. Oncology 1987; 44:69-72. 95. Homesley HD, Harry DS, OToole RV et al. Randomized comparison of cisplatin plus epirubicin or doxorubicin for advanced epithelial ovarian cancer. Am J Clin Oncol (CCT) 1992; 15: 129-34. 96. Bezwoda WR. Treatment of advanced ovarian cancer: A randomized trial comparing adriamycin or 4'-epi-adriamycin in combination with cisplatin and cyclophosphamide. Med Pediatr Oncol 1986; 14: 26-9. 97. Jones WG, Fossa SD, Bono AV et al. European Organization for Research and Treatment of Cancer (EORTC) phase II study of low-dose weekly epirubicin in metastatic prostate cancer. Cancer Treat Rep 1987; 71: 1317-8. 98. Tveter KJ, Hagen S, Holme I et al. A randomized study on hormone-resistant prostatic cancer Estramustine phosphate versus low dose epirubicin with or without medroxyprogesterone acetate. Scand J Urol Nephrol 1990; 24: 243-7. 99. Pummer K. Epirubicin plus flutamide and orchiectomy in previously untreated advanced prostatic cancer. Semin Oncol 1991, 18(Suppl6):26-8. 100. Fossa SD, Splinter T, Roozendaal KJ et al. A phase II study of 4'-epi-adriamycin in advanced urothelial transitional cell cancer. EORTC-GU Group Protocol 30867. Eur J Cancer Clin Oncol 1989;25:389-90. 101. Riither U, Rupp W, Bauerle K et al. Methotrexate, vinblastine, 4'-epirubicin and cisplatin in advanced urothelial tract tumors: Efficacy and toxicity. World J Urol 1990; 8: 51-4. 102. Frassoldati A, Federico M, Barbien F et al. Methotrexate, vinblastine, epidoxorubicin and cisplatin (M-VEC) in patients with locally advanced transitional bladder cancer. Med Oncol & Tumor Pharmacother 1991; 8: 99-103. 103. Goldfarb A, Negro A, Coimbra F et al. Methotrexate, vindesine, epidoxorubicin and cisplatin (M-VEC) for stage III-IV bladder cancer. Proc Am Soc Chn Oncol 1992; 11: 210 (Abstract). 104. Lorusso V, Berardi F, Catino A et al. Methotrexate, vinblastine, epidoxorubicin and carboplatin (M-VECA) in the treatment of advanced bladder cancer. Proc Am Soc Clin Oncol 1992; 11: 212 (Abstract). 105. Passalacqua R, Cocconi G, Macaluso G et al. M-VAC versus the same combination using epirubicin at double dose (M-VEEC) in locally invasive bladder cancer. A prospective study. Proc Am Soc Clin Oncol 1992; 11: 210 (Abstract). 106. Schabel FM, Griswold DP, Corbett TH et al. Increasing the therapeutic response rates to anticancer drugs by applying the basic principles of pharmacology. Cancer 1984; 54: 1160-7. 107. Skipper HE. Laboratory models: The historical perspectives. Cancer Treat Rep 1986; 70: 3-7. 108. Frei E III. Curative cancer chemotherapy. Cancer Res 1985; 45:6523-37. 109. Frei E III, Canellos GP. Dose: A critical factor in cancer chemotherapy. Am J Med 1980; 69: 585-93. 110. Bonadonna G, Valagussa P. Dose-response effect of adjuvant chemotherapy in breast cancer. N Engl Med 1981; 304: 10-5. 111. Hryniuk W, Levine MN. Analysis of dose intensity for adjuvant chemotherapy trials in stage II breast cancer. J Clin Oncol 1986; 4: 1162-70. 112. Allen A. The cardiotoxicity of chemotherapeutic drugs. Semin Oncol 1992; 19:529-42. Received 22 December 1992, accepted 23 December 1992 Correspondence to: Dr. Gianni Bonadonna Istituto Nazionale Tumori Via Venezian 1 Milan 20133, Italy