`Primed
`I” Great Britam.
`
`M77-5379/89$3.00+0.00
`0 1983 Pergamon Press Ltd.
`
`Carminomycin vs Adriamycin in Advanced
`Tissue Sarcomas: an EORTC Randomised
`Phase II Study*
`
`Soft
`
`VIVIEN H. C. BRAMWELL,t$ H. T. MOURIDSEN,§ J. H. MULDER,II R. SOMERS, n A. T. VAN
`OOSTEROM,** A. SANTORO,?? D. THOMAS,$$ R. SYLVESTERIS and D. MARKHAM?
`tCancer Research Campaign, Department of Medical Oncology, Manchester Uniuersity and Christie Hospital Q
`Holt Radium Institute, Wilmslow Road, Manchester M20 9BX, U.K., §Finseninstitutet, Copenhagen, Denmark,
`(IR.R.T.I., Rotterdam, The Netherlands, TAntoni Van Leeuwenhoek Ziekenhuis, Amsterdam, The Netherlands,
`**Akademisch Ziekenhuis, Leiden, The Netherlands,
`ttlstituto Nazionale Per La Studio Dei Tumori, Milan,
`Italy and $$EORTC Data Centre, Brussels, Belgium
`
`patients with aduanced soft tissue sarcoma who had
`Abstract-Eighty-three
`received no prior chemotherapy
`entered a randomised phase II study comparing
`20 mg/m’ with adriamycin
`(ADM) 75 mg/m2, both
`carminomycin
`(CMM)
`i.v. bolus euery 3 weeks. Six patients were ineligible and response
`administered
`could not be evaluated in 6 additional patients. Thirty-eight evaluable patients
`received ADM and 33 received CMM. There was one complete response to ADM and
`10 partial responses, an overall response rate of 29%. CMM showed significantly
`(P = 0.01) lower antitumour activity-one
`partial response (3%). Stabilisation of
`disease was similar in both arms (47 and 45%), but fewer patients progressed on
`ADM
`(24 compared
`to 52%). Durations of response dating from the start of
`chemotherapy were 5 months for complete
`remission on ADM, a median oj 7
`months (range 4-17)for partial response on ADM and 14 monthsfor the one partial
`remission on CMM. Although
`the median
`time to progression
`for all patients
`(P = 0.001) shorter than for those
`receiving CMM (2 months) was significantly
`receiving ADM (5 months), patients on ADM had only a marginally significant
`longer duration of survival (P = 0.06) than the patients receiving CMM. A leucocyte
`nadir <2.0 X log/1 occurred in 38% of patients receiving ADM and 43% receiving
`CMM. Anaemia and thrombocytopoenia were uncommon. Other toxicitiessuch as
`nausea, vomiting, anorexia and alopecia were more severe in the ADM group.
`in the ADM arm, and one
`There was one infective death secondary to leucopoenia
`patient who had
`received
`109 mg/m 2 CMM +255 mg/m2 ADM developed
`progressively
`fatal cardiomyopathy.
`
`INTRODUCTION
`Son- tissue sarcomas are sometimes responsive to
`chemotherapy and many single agents and drug
`combinations have been tested in these compara-
`tively rare tumours. Adriamycin remains the most
`active
`single agent,
`although
`the range of
`response rates, 9-70%,
`is wide [l]. Variability in
`the dose and schedule of drug administration and
`heterogeneity of the patient population with
`
`Accepted 11 April 1983.
`*This investigation was supported by grant Nos 5RIO
`CAI1488-12 andZUlOCA11488-13, awarded by the National
`Cancer Institute, DHEW, Bethesda, MD, U.S.A.
`$To whom requests for reprint should be addressed.
`
`LJC 19:8-”
`
`1097
`
`regard to factors such as prior chemotherapy or
`radiotherapy, performance status and site, volume
`and histology of the tumour may account for this
`disparity.
`In 1975 two literature reviews [Z, 31,
`each with more than 350 patients, found average
`response rates of 25 and 27%. Collected data from
`more recent series suggest a slightly lower rate of
`remission-23%
`in 154 patients [4,5]. In most of
`these series large intermittent doses of adriamycin,
`in the range 60-75 mg/m2 every 3-4 weeks, have
`produced
`the highest
`response
`rates.
`the past 7 yr a large number of drug
`Over
`[4-61. Those
`combinations
`have been evaluated
`lacking adriamycin
`are rarely active, but the range
`for those containing
`of response
`rates (15-74s)
`
`Ex. 1043-0001
`
`
`
`1098
`
`V. H. C. Bramwell et al.
`
`tested
`is wide. The most extensively
`adriamycin
`is CYVADIC
`(cyclophosphamide,
`combination
`vincristine, adriamycin and DTIC), but although
`an earlier study documented an overall response
`rate as high as 59% [Z], re-analysis of the data
`reduced
`this
`to 47%
`[7],
`and
`subsequent
`experience with
`this
`combination
`has been
`disappointing
`[g, 91. In all series the complete
`remission
`rate rarely exceeds 15%. In advanced
`soft
`tissue
`sarcoma
`chemotherapy
`is almost
`invariably palliative, and long-term survivors are
`rare
`[lo]. As
`the
`toxicity
`of combination
`chemotherapy usually exceeds that encountered
`with
`single-agent
`adriamycin
`and
`there
`is
`considerable
`overlap
`in
`response
`rates,
`the
`advantages of multiple drug therapy are far from
`clear.
`Adriamycin has a number of undesirable side-
`effects,
`including
`vomiting, mucositis, myelo-
`suppression,
`alopecia and,
`in high cumulative
`doses, cardiotoxicity.
`Intensive research has been
`directed
`towards the development of analogues
`lacking some of these toxicities. Carminomycin,
`is an
`isolated
`from Actinomadura
`carminata,
`interesting analogue which has been developed in
`Russia.
`It is chemically
`related to daunorubicin,
`but
`its biochemical
`activity
`resembles
`that of
`adriamycin
`[Ill.
`Pre-clinical
`studies
`in
`the
`Zbinden rat model suggested a lower potential for
`cardiotoxicity
`than adriamycin. Myelosuppres-
`sion was the dose-limiting
`toxicity and data from
`studies, which
`employed
`two dose
`Russian
`schedules
`(7.5 mg/m* twice weekly and 5.5 mg/
`m*/day X 5), suggested significant activity in soft
`tissue
`sarcomas-13
`responses
`in 48 patients
`(27%)- as well as in other tumours such as breast
`carcinoma,
`lymphomas and leukaemia
`[12,13].
`Nausea and vomiting
`seemed
`less severe and
`alopecia was not reported. A phase I study in
`Brussels,
`conducted
`by Rozencweig
`and co-
`workers using a single dose schedule, found that
`leucopoenia, with a nadir at day 12, was the dose-
`limiting
`toxicity, and the recommended dose for
`phase II study in good-risk patients was 20 mg/m2
`i.v. every 3 weeks [14].
`At the end of 1979 the EORTC Soft Tissue and
`Bone Sarcoma Group decided to conduct a series
`of randomised phase II studies in patients who
`had received no prior chemotherapy,
`comparing
`newer anthracyclines with adriamycin. The aim
`was to identify an analogue with less toxicity but
`equal or better antitumour activity
`in soft tissue
`sarcoma.
`In a randomised phase II study a true
`comparison of the relative efficacies of two active
`analogues cannot be made without considerable
`expansion of patient numbers, but an inactive
`drug may be discarded if the response rate in the
`adriamycin arm indicates a representative patient
`
`this
`advantage of
`population. An additional
`policy was that new phase II agents could be
`evaluated
`in patients who had only received one
`or two anthracyclines.
`The results of the first
`study assessing carminomycin
`are reported here,
`and a second study evaluating 4’-epiadriamycin
`is
`currently
`in progress.
`
`MATERIALS AND METHODS
`Criteria for eligibility
`Patients,
`15-80 yr of age, with histologically
`proven advanced and/or metastatic
`soft tissue
`sarcoma were eligible
`for this study. Patients were
`required
`to have measurable progressive disease
`and a Karnofsky performance
`status of at least
`50%. Recurrent
`tumour
`in irradiated areas was
`not permitted as the sole evaluable
`lesion, and
`pleural effusions or bony metastases were not
`considered
`to be measurable. Other criteria
`for
`exclusion were prior
`treatment with cytotoxic
`agents,
`a previous
`or concomitant
`different
`malignant
`tumour, congestive cardiac failure or
`other
`serious
`concurrent
`disease, and central
`nervous system metastases. Prior to entry patients
`were required
`to have adequate hepatic excretory
`function
`(serum bilirubin <50 pmol/l) and bone
`marrow reserve (leucocytes <4.0 X log/l, platelets
`>lOO X 109/l).
`
`Trial design
`After stratification by institution, patients were
`randomised
`to receive either adriamycin
`(NSC-
`123127) or carminomycin
`(NSC-180024). Patients
`showing disease progression after two courses of
`carminomycin were crossed
`to the adriamycin
`arm, while those failing on adriamycin went off-
`study.
`In
`the event of disease
`stabilisation,
`patients continued
`therapy until progression or
`the maximum
`cumulative dose of the drug was
`received.
`
`regime
`Therapeutic
`Adriamycin
`(ADM) 75 mg/m’ was given as an
`i.v. bolus once every 3 weeks. Continuation
`of
`therapy beyond a cumulative dose of 550 mg/m2
`was not
`recommended,
`but was
`left at
`the
`discretion
`of
`the
`individual
`investigator.
`Carminomycin
`(CMM) 20 mg/m2 was given as an
`i.v. bolus once every 3 weeks. The potential
`for
`cardiotoxicity was unknown
`and no specific
`recommendations were made about cumulative
`dose. (Carminomycin was supplied by Bristol-
`Myers, New York.)
`
`treatment
`Dose modifications during
`The dose was reduced by 50%
`if the serum
`bilirubin was between 35 and 50 pmol/l, and the
`drug was discontinued
`if the bilirubin was above
`
`Ex. 1043-0002
`
`
`
`Carminomycin
`
`us Adriamycin
`
`in Advanced Soft Tissue Sarcomas
`
`1099
`
`count was below
`the WBC
`If
`50 I.rmol/l.
`3.0 X log/1 or the platelets below 100 X 10’11 3
`weeks
`after
`the
`last
`course,
`treatment was
`postponed
`for 1 week. At this time, if the WBC
`were between
`2.0 and 2.9 X log/1 or platelets
`75-99 X log/l, therapy was continued at 50% dose.
`Counts below
`these levels precluded
`treatment.
`Adjustments
`for
`the nadir count
`in previous
`courses were: WBC 2.0-2.9 X log/1 or platelets
`50-74 X log/l, adriamycin
`75% dose, carmino-
`mycin 90% dose; WBC <2.0 X log/1 or platelets
`<50 X 109/1, adriamycin
`50% dose, carmino-
`mycin
`75% dose. Dose escalation was not
`permitted. Patients went off-study
`if haemato-
`logical toxicity delayed retreatment
`for more than
`3 weeks.
`
`investigations
`Pretreatment and follow-up
`Baseline studies included history and physical
`examination,
`Karnofsky
`performance
`status,
`tumour measurements,
`complete
`blood count
`including differential white count, biochemical
`profile, chest radiograph and ECG. Echocardio-
`graphy and radionucleide
`cardiac
`scans were
`performed
`in some centres. Nadir blood counts
`between
`1 and 2 weeks after
`treatment were
`performed
`in half of the cases, and all baseline
`investigations were repeated after two courses of
`chemotherapy and at the time of discontinuation
`or cross-over
`to alternative
`therapy.
`
`Definition of response
`Patients were considered evaluable for response
`if they had received two courses of chemotherapy
`and tumour measurements had been repeated at 6
`weeks. Response
`criteria were those defined by
`WHO [ 151. Any demise occurring before 6 weeks
`was classified as early death.
`
`Toxicity
`toxicity were
`cardiac
`and
`Haematological
`to WHO criteria
`[15]. Other
`graded according
`toxicities were graded according
`to the following
`scale: 0 = none; 1 = mild, not requiring modifica-
`tion of treatment; 2 = moderate, requiring modifi-
`cation of treatment; 3 = severe; 4 = life-threaten-
`ing; 5 = lethal. Alopecia was graded: 0 = none;
`1 = minimal;
`2 = mild, not
`requiring
`a wig;
`3 = moderate, requiring a wig; 4 = complete.
`
`Central fiathology review
`A central pathology review was carried out by 2
`panels consisting of 6 members each, one for the
`Northern European
`Institutes, chaired by Profes-
`sor Van Unnik, Utrecht, The Netherlands, and one
`for the Southern European
`Institutes, chaired by
`Dr Contesso, Villejuif, Paris, France. If 2 members
`of a panel independently made the same diagnosis
`
`this diagnosis was
`as the referring pathologist,
`accepted.
`If
`there was disagreement,
`other
`members of the panel examined
`the histological
`sections and a consensus diagnosis was reached.
`
`Statistical design
`An initial patient entry of 29 evaluable patients
`in each arm was required, with termination of the
`study if 3 or fewer responses were observed
`in
`either arm. There was provision
`to stop the study
`if no responses were reported
`in
`the first 19
`patients
`in either arm. This plan ensured that if
`the anthracycline
`analogue had a true response
`rate of at least 25%, the probability of rejecting
`it
`from further study was cO.05.
`
`RESULTS
`Over a g-month period from March 1980, 83
`patients were entered
`into
`this study by 13
`European centres. Six patients were considered to
`be ineligible,
`five of these after histological
`review
`(xanthogranuloma,
`chondrosarcoma,
`benign
`schwannoma,
`non-Hodgkin’s
`and Hodgkin’s
`lymphomas)
`and one patient who had a
`Karnofsky performance
`status below 50% and no
`measurable disease. Another six patients were not
`evaluable
`for tumour response. These
`included
`three early deaths (<6 weeks), two due to malignant
`disease and one due to toxicity. One patient was
`lost to follow-up after the second course, prior to
`evaluation,
`and
`in
`two patients
`the followed
`lesion was treated by either surgery or radio-
`therapy after one course of therapy. The last two
`patients were considered
`to be evaluable
`for
`toxicity. Thus 71 patients, 38 in ADM and 33 on
`CMM, could be evaluated for response, and their
`characteristics
`are shown in Table 1. Apart from a
`reversal of the sex ratio, such that females were
`commoner
`in the ADM arm, the groups were well
`balanced. Histological material has been reviewed
`in 62 of the 71 evaluable cases, and the cell types
`are shown in Table 2. ADM produced 1 CR and 10
`PR, an overall response rate of 29% (Table 3).
`Disease
`stabilised
`in 18 patients
`(47%)
`and
`progressed
`in 9 (24%). CMM showed minimal
`antitumour activity-l
`PR (3%) disease stabilis-
`ing in 15 patients
`(45%) and progressing
`in 17
`(52%).
`The
`difference
`in
`response
`rate
`is
`at P = 0.01. Sixteen patients crossed
`significant
`from CMM
`to ADM, and one patient with
`progressive disease achieved a PR. Nine of the 11
`patients responding
`to ADM had lung metastases,
`although one also had lymph node and another
`had subcutaneous metastases. Two patients had
`local residual/recurrent
`disease only. The one
`patient who responded
`to CMM had a complete
`clinical
`remission of extensive
`intra-abdominal,
`
`Ex. 1043-0003
`
`
`
`II00
`
`V. H. C. Bramwell et al.
`
`Table 1. Patient characteristics
`
`ADM (38 patients) CMM (33 patients)
`
`Sex
`male
`female
`Age
`median (yr)
`range
`Kamofsky PS (%)
`median
`range
`Sites of disease
`locoregional only
`metastases only
`both
`Distribution of disease
`locoregional*
`abdominal
`trunk
`limbs
`metastasest
`lung
`liver
`subcutaneous
`lymph nodes
`intra-abdominal
`No. of courses
`mean
`range
`Dose per course (mg/m*)
`mean
`range
`
`15 (39%)
`23 (61%)
`
`56.5
`22-73
`
`90
`60-100
`
`7
`19
`12
`
`10
`1
`2
`
`25
`3
`1
`1
`1
`
`6
`2-11
`
`70
`45-77
`
`20 (61%)
`13 (39%)
`
`54
`28-74
`
`90
`50-100
`
`10
`10
`13
`
`13
`2
`3
`
`14
`6
`5
`2
`3
`
`4
`2-13
`
`19
`15-21
`
`irradiated lesions not included-followed
`*Previously
`meters only.
`tSome patients-more
`
`than one site.
`
`para-
`
`a
`from
`node metastases
`and axillary
`liver
`liposarcoma of the thigh, although residual intra-
`abdominal disease was evident on CT scan.
`Durations of response dating from the start of
`chemotherapy were 5 months
`for the 1 CR (1
`month
`from documentation
`of CR), median 7
`months
`(range 4-17 months)
`for 6 PR on ADM
`
`and 14 months for the 1 PR on CMM. One partial
`responder to ADM was lost to follow-up after four
`courses of chemotherapy, but with this exception
`all responding patients ultimately progressed. A
`significant difference
`(P = 0.001)
`in median time
`to progression
`(Fig. 1) in favour of ADM (5 vs 2
`months) was observed. Based on all eligible
`patients for whom follow-up data wereavailable, a
`marginally
`significant
`longer duration of survival
`on ADM was noted (P = O.OS), with a median of
`approximately
`9 months
`in both groups (Fig. 2).
`If only evaluable patients were analysed,
`the
`difference was even
`less significant
`(P = 0.13).
`Thirty-one patients on the CMM arm have died
`and 28 on the ADM arm. There was one toxic
`death due to anthracycline
`cardiomyopathy,
`and
`apart
`from
`the
`four
`intercurrent
`deaths,
`the
`remaining patients died of malignant disease.
`Haematological
`toxicity encountered during
`the first
`two courses
`is illustrated
`in Table 4.
`Nadir
`(days 7-14
`after chemotherapy)
`blood
`counts were available
`for 21 patients in each arm.
`Leucopoenia <2.0X
`10’11 was noted in 38% of
`patients
`receiving ADM and 43% of patients
`receiving
`CMM.
`Thrombocytopaenia
`and
`anaemia were uncommon. Data were insufficient
`to calculate
`cumulative myelotoxicity. Other
`toxicities
`are summarised
`in Table 5. Nausea/
`vomiting,
`anorexia
`and alopecia were signi-
`ficantly more pronounced
`in the ADM group.
`Diarrhoea
`and mucositis were
`infrequent
`and
`mild in the ADM arm and rare in the CMM arm.
`Table
`6 summarises
`the cumulative
`dose of
`anthracycline. The patient who died of progres-
`sive cardiomyopathy
`had received 109 mg/m’ of
`CMM and 255 mg/m’ of ADM. Three other
`patients had crossed from CMM to ADM and had
`received high cumulative doses of both drugs. The
`highest
`total dose of ADM alone was 772 mg/m’
`
`Table 2. Histological subtypes
`
`ADM
`
`CMM
`
`Total (%)
`
`Leiomyosarcoma
`Malignant fibrous histiocytoma
`Liposarcoma
`Neurofibrosarcoma
`Fibrosarcoma
`Synovial sarcoma
`Angiosarcoma
`Rhabdomyosarcoma
`Unclassified
`Miscellaneous*
`Not reviewed
`
`Total
`
`10
`5
`2
`4
`3
`2
`0
`0
`4
`4
`4
`
`38
`
`6
`6
`8
`2
`1
`1
`2
`0
`1
`1
`5
`
`16 (23)
`11 (15)
`10 (14)
`6 (8)
`4 (6)
`3 (4)
`2 (3)
`0 (0)
`5 (7)
`5 (7)
`9 (13)
`
`33
`
`71 (100)
`
`cell, mixed mesodermal, alveolar soft parts, round
`*Includes: ADM-clear
`cell undifferentiated; CMM-endometrial
`stromal.
`
`Ex. 1043-0004
`
`
`
`Carminomycin
`
`us Adriamycin
`
`in Advanced Soft Tissue Sarcomas
`
`1101
`
`Table 5. Response to chemotherapy
`
`%
`
`3
`26
`47
`24
`
`100
`
`ADM
`
`No.
`
`1
`10
`18
`9
`
`38
`
`2
`1
`1
`
`42
`
`CR
`PR
`NC
`PD
`
`Total evaluable
`
`Ineligible
`Early death
`Not evaluable
`
`Total entered
`
`CMM
`
`Total
`
`%
`
`0
`3
`45
`52
`
`100
`
`No.
`
`0
`1
`15
`17
`
`33
`
`4
`2
`2
`
`41
`
`1
`11
`33
`26
`
`71
`
`6
`3
`3
`
`83
`
`Difference
`P=O.OOl.
`
`in response
`
`rate, 29 vs 3%, P =O.Ol. Difference
`
`in avpage
`
`respofse,
`
`E.O.R.T.L_Duta
`Probab I 1 lty
`
`Center.
`TIME
`
`TO PROGRESSION
`
`I
`
`I
`
`I____
`
`I
`I
`I
`I
`I
`I
`8
`1
`I
`Lb___
`
`I
`I
`I
`I
`
`-T-
`.so-
`
`.BO-
`
`.70-
`
`.60-
`
`.50-
`
`.40-
`
`.30-
`
`.20-
`
`.10-
`
`I_____________________
`
`3
`0
`Number of patients
`36
`28
`33
`16
`
`ot
`
`risk
`
`6
`at
`16
`3
`
`9
`the corresponding
`7
`2
`
`TOTRL
`
`FAIL.
`
`TRERTIWIT
`
`30
`33
`
`::
`
`ml
`Cllll
`
`P=
`
`.OOl
`
`<LOGRANK
`
`)
`
`I
`
`I
`k________
`,
`12
`timess
`5
`1
`
`15
`
`5
`0
`
`1
`
`‘* MONTHS
`
`1 RDIl
`0 elm
`
`Fig. 1. Time to progression from randomisation, ADM vs CMM.
`
`patient
`and of CMM alone 238 mg/m2-neither
`patients
`developed
`cardiomyopathy.
`Three
`received >550 mg/m2 ADM and two received
`2150 mg/m2 CMM.
`
`DISCUSSION
`The results of this study are in agreement with
`earlier reports in the literature of a response rate of
`2540%
`for single-agent
`adriamycin used at
`
`in
`tolerated doses [l-5]. However,
`maximally
`contrast with Russian studies [13], we have shown
`very little activity for carminomycin used in this
`dose and schedule. It is possible
`that a twice
`weekly
`schedule,
`as used by
`the Russian
`investigators,
`is more effective, although
`this
`remains to be confirmed. The total dose delivered
`over a 6-week period does not differ significantly
`between the two regimens. It is interesting to note
`
`Ex. 1043-0005
`
`
`
`1102
`
`V. H. C. Bramwell et al.
`
`E.O.R.T.C.
`
`Data
`
`Probability
`
`Center.
`DURATION
`
`OF SURUIUAL
`
`- FlLL ELIG
`
`PATS
`
`TOTR FRtL.
`
`TRERTnENl
`
`:z
`
`::
`
`RDn
`cnn
`
`P=
`
`.ow
`
`(LOGMW
`
`)
`
`.80-
`
`.70-
`
`.60-
`
`--? -____
`
`1__,
`
`I__
`
`I
`I
`
`0
`Number
`33
`36
`
`6
`
`of
`
`patients
`
`at
`
`risk
`
`at
`
`28
`23
`
`corresponding
`
`12
`the
`14
`10
`
`18
`times8
`
`12
`4
`
`L___.
`
`24
`
`5
`2
`
`1
`
`3o
`
`I’IONTHS
`
`0 RDil
`0 cn
`
`Fig. 2. Survival from randomisation, ADM us CMM.
`
`found
`trials have
`two other EORTC
`that
`carminomycin,
`in a similar dose and schedule, to
`be inactive in breast cancer [16,17].
`In the present
`study the median duration of response (7 months)
`was inferior
`to that (14 months) on CYVADIC
`combination
`chemotherapy
`in our previous study
`[9]. Surprisingly,
`the one patient responding
`to
`CMM had the longest duration of response
`(14
`months).
`In the previous CYVADIC randomised
`trial and the present study, differences in response
`rate between the two arms did not translate
`into
`significant
`benefit
`in terms of overall survival.
`The low response rate, even in the ADM arm, and
`the relatively
`indolent nature of some tumours
`may account for this discrepancy. In contrast with
`studies of combination
`chemotherapy,
`in which
`occasional
`long-term
`(>3 yr) disease-free
`sur-
`vivors have been reported [lo, 181, all patients
`in
`this study treated by chemotherapy
`alone have
`relapsed. Only one partial remission was observed
`on crossing from CMM to ADR, but the numbers
`are too small to draw any conclusions.
`Although
`nadir
`blood
`counts were only
`available
`in
`just over half of the patients,
`the
`distribution
`of haematological
`toxicity suggests
`
`at maximal
`that both drugs were administered
`leucopoenia
`dose for the schedule chosen. The
`produced by CMM slightly exceeded that induced
`by ADM. In contrast, gastrointestinal
`intolerance
`and alopecia were significantly
`less severe with
`CMM. Cross-over
`from CMM
`to ADM was
`accompanied
`by a significant
`deterioration
`in
`patient
`tolerance.
`As the majority of patients did not receive doses
`of ADM in excess of 550 mg/m2, and few patients
`received high doses of CMM, very little comment
`can be made about cardiotoxicity. No patient who
`received CMM alone developed cardiomyopathy.
`The one fatal case of cardiomyopathy
`had only
`received 225 mg/m* of ADM, which alone has low
`potential
`for cardiotoxicity. Prior to this he had
`received
`the equivalent of 5% courses of CMM,
`which may have contributed
`to
`the cardiac
`damage. However,
`in view of the low activity of
`CMM, the question of cardiotoxicity
`is at present
`academic.
`although CMM was better
`In conclusion,
`tolerated than ADM, in this dose and schedule it
`demonstrated minimal
`antitumour
`activity
`in
`patients with advanced soft tissue sarcoma.
`
`Ex. 1043-0006
`
`
`
`at
`de-
`and
`
`cardiomyopathy
`cardiac
`failure
`
`520 mg/mz ADM-mild
`heart
`
`cardiotoxicity-progressive
`
`tumour;
`
`to
`
`due
`
`death
`
`autopsy.
`compensation;
`death. Also, O.P.
`fDefinite
`tl50,
`*579, 629, 772 mg/m’.
`
`238 mg/m2.
`
`355
`255
`546
`546
`
`(mg/m’)
`??ADM
`
`6%
`12%
`
`8%
`34%
`
`56
`109
`39
`100
`
`(mg/m’)
`CMM
`
`e 8+ courses
`
`Both drugs
`
`2/33 patients7
`4/33 patients
`
`alone
`3/38 patients*
`13138 patients
`alone
`
`Adriamycin
`
`Carminomycin
`
`D.B.
`J.V.S
`A.K.
`R.V.
`
`2150 mg/m2
`2120 mg/m2
`
`>550 mg/m*
`>450 mg/m’
`
`Table 6. Cumulative dose of anthracycline
`
`21
`12
`21
`21
`
`21
`11
`21
`20
`
`CMM
`
`ADM
`No. evaluable*
`
`10
`84
`95
`53
`
`19
`45
`66
`55
`
`(%)
`CMM
`
`(%)
`ADM
`
`Total
`
`l-4
`
`Grades
`
`5
`17
`10
`0
`
`0
`18
`5
`0
`
`5
`17
`33
`10
`
`0
`18
`33
`0
`
`(%)
`CMM
`
`(%)
`ADM
`
`4
`
`(%)
`CMM
`
`(%)
`ADM
`
`3
`
`WHO grade of toxicity
`
`two courses
`
`toxicity--first
`
`hot flushes,
`
`improvement.
`
`0
`0
`3
`0
`0
`0
`3
`0
`9*
`
`0
`45
`3
`0
`0
`3
`0
`0
`13
`
`(%)
`CMM
`
`(%)
`ADM
`
`3
`
`Severe
`
`0
`3
`0
`0
`6
`0
`3
`0
`18
`
`lungs.
`
`liver, but not
`
`and
`
`abdomen
`
`in upper
`
`had
`tumour
`constipation.
`
`dry mouth,
`
`dizzy, tired, wateringeyes,
`
`e.g. anxiety/depression,
`
`after 2 courses with
`
`infusion
`
`to 6-hr
`
`psychological,
`changed
`evaluable: ADM, 38; CMM, 34.
`
`toxicities with both drugs-mainly
`vomiting-patient
`life-threatening
`
`$Respiratory-patient
`headache,
`tMild
`*One
`No. of patients
`
`31
`34
`0
`3
`8
`3
`24
`3
`45
`
`29
`41
`0
`6
`18
`3
`50
`9
`53
`
`26t
`21
`5
`8
`11
`29
`42
`21
`37
`
`71
`56
`97
`94
`76
`97
`44
`91
`20
`
`71
`0
`92
`89
`81
`65
`34
`76
`2
`
`(%)
`CMM
`
`(%)
`ADM
`
`2
`
`Moderate
`
`(%)
`CMM
`
`(%)
`ADM
`
`(%)
`CMM
`
`(%)
`ADM
`
`1
`Mild
`
`0
`None
`
`Grade of toxicity
`
`courses
`
`toxicity-all
`
`Table 5. Non-haematological
`
`Other
`Alopecia
`Bleeding
`Fever with drug
`Weight
`Mucositis
`Anorexia
`Diarrhoea
`Nausea/vomiting
`
`loss
`
`Type of toxicity
`
`(days 7-14) blood counts.
`
`nadir
`
`having
`
`*No. of patients
`
`0
`33
`38
`10
`
`5
`0
`14
`10
`
`0
`17
`14
`33
`
`14
`9
`14
`45
`
`90
`16
`5
`47
`
`81
`55
`34
`45
`
`(%)
`CMM
`
`(%)
`ADM
`
`2
`
`(%)
`CMM
`
`(%)
`ADM
`
`1
`
`(%)
`CMM
`
`(%)
`ADM
`
`0
`
`Thrombocytopoenia
`Granulocytopoenia
`Leucopoenia
`Anaemia
`
`Type of toxicity
`
`Table 4. Haematological
`
`Ex. 1043-0007
`
`
`
`1104
`
`V. H. C. Bramwell
`
`et al.
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`18.
`
`REFERENCES
`PINEDO HM, KENIS Y. Chemotherapy
`of advanced soft tissue sarcoma in adults. Cancer
`Treat Rev 1977, 4, 67-86.
`GOTTLIEB JA, BAKER LH, O’BRYAN RM et al. Adriamycin
`(NSC-123127) used alone
`and in combination
`for soft tissue and bone sarcomas. Cancer Chemother Rep 1975,6,
`271-282.
`WASSERMAN TH, &MIS RL, GOLDSMITH M et al. Tabular
`analysis of the clinical
`chemotherapy
`of solid tumours. Cancer Chemother Rep 1975, 6, 399-419.
`BRAMWELL VHC, PINEDO HM. Bone and soft tissue sarcomas. In: F'INEW HM, ed.
`1.
`1979, The EORTC
`Cancer Chemotherapy
`Cancer Chemotherapy
`Annual
`Amsterdam, Excerpta Medica, 1979, Ch. 20, 424-450.
`BRAMWELL VHC, PINEDO HM. Bone and soft tissue sarcomas. In: PINEDO HM, ed.
`Cancer Chemotherapy
`1980, The EORTC
`Cancer Chemotherapy
`Annual
`2.
`Amsterdam, Excerpta Medica, 1980, Ch. 21, 393-414.
`BRAMWELL VHC, PINEDO HM. Bone and soft tissue sarcomas. In: PINEDO HM, ed.
`1981, The EORTC
`Cancer Chemotherapy
`Cancer Chemotherapy
`Annual
`3.
`Amsterdam, Exerpta Medica, 1981, Ch. 21, 409-424.
`adriamycin
`vincristine,
`YAP BS, BAKER LH, SINKOVICS JG et al. Cyclophosphamide,
`and DTIC
`(CYVADIC) combination
`chemotherapy
`for the treatment of advanced
`sarcomas. Cancer Treat Rep 1980, 64, 93-98.
`GIULIANO AE, LARKIN KL, EILBER FR, MORTON DL. Failure
`chemotherapy
`(CYVADIC)
`in metastatic
`soft
`tissue
`sarcomas:
`adjuvant
`studies. Proc Am Sot Clin Oncol 1978, 19, 359.
`PINEDO HM, BRAMWELL VHC, MOURIDSEN HTet al. CYVADIC in advanced soft tissue
`study with two schedules. Cancer In press.
`sarcoma: a randomised
`SINKOVICS JG. Long
`tumour-free
`survival of adult patients
`treated for metastatic
`sarcomas. Proc Am Assoc Cancer Res 1979, 20, 16.
`BAKER LH, KESSEL DH, C~MIS RL, REICH SD, DEFURIA MD, (XOOKS ST. American
`Cancer Treat Rep 1979,63, 899-902.
`experience with carminomycin.
`COOKE ST. A review of carminomycin-a
`new an thracycline developed in the USSR. 1
`Med 1977,8, 295-316.
`study of
`PEREVO DCHIKOVA NI, LICHINITSER MR, GORBUNOVA VA. Phase I clinical
`carminomycin:
`its activity against soft tissue sarcoma. Cancer TreatRep 1977,61,1705.
`ADELE R, ROZENCWEIC M, BODY JJ et al. Carminomycin
`(NSC-180024): a phase I study.
`Eur / Cancer 1980, 16, 1555-1559.
`WHO Handbook for Reporting Results of Cancer Treatment. WHO Offset Publication
`No. 48, Geneva, 1979.
`VAN OOSTEROM AT, MOURIDSEN HT, WILDIERS J et al. Randomised phase II trial of
`carminomycin
`(CMM) vs adriamycin
`(ADM) in advanced breast cancer. In: Proceedings
`of the Third EORTC Breast Cancer Conference, Amsterdam, 1983. In press.
`phase II
`ROZENCWEIC M, TEN BOKKEL HUININK WW, CAVALLI F et al. Randomised
`trial of carminomycin
`versus 4’epiadriamycin
`in advanced breast cancer. Proc Am Sot
`Clin Oncol 1982, 1, 89.
`treatment of advanced
`WILTSHAW E, HARMER C, MCKINNA A. Soft tissue sarcoma:
`disease in the Royal Marsden Hospital.
`In: VAN OOSTEROM AT, CLETON FJ, MUGGIA
`FM, eds. Therapeutic Progress in Ovarian Cancer, Testicular Cancer and the Sarcomas.
`Martinus Nijhoff, The Hague, 413-423.
`
`of combination
`implications
`for
`
`Ex. 1043-0008