Progress in Chemotherapy for Metastatic Breast Cancer
`
`George W~ Sledge, Jr and Karen H. Antman
`
`APPROXIMATELY 45,000 women died of
`breast cancer in the United States in 1991,
`virtually all as a consequence of distant meta(cid:173)
`static disease. Women with metastatic breast
`cancer are essentially incurable with standard
`therapy with a median survival of about 2 years
`after documentation of metastases. 1,2 The me(cid:173)
`dian survival of women with metastatic disease
`has not changed in the 5 decades for which
`statistics are available. While generally sensitive
`to initial chemotherapy regimens, metastatic
`breast cancer virtually always progresses with
`shorter and less complete remissions with subse(cid:173)
`quent regimens. Women with estrogen receptor
`positive tumors (median survival, 2.3 years), and
`those who achieve a complete response with
`standard dose therapy (median, 2.5 years) or
`who have only small amounts of local disease
`(median, > 4 years) have a somewhat better
`prognosis. 2
`Metastatic breast cancer therefore represents
`a major public health problem as well as a
`frightening personal dilemma for women af(cid:173)
`flicted with the disease. For the physician caring
`for the patient with metastatic breast cancer,
`the disease represents a separate set of prob(cid:173)
`lems. What treatment goals should the physi(cid:173)
`cian strive for? Should medical efforts focus on
`the production of high clinical response rates?
`On prolongation of survival? On palliation of
`symptoms? On improvements in quantitatively
`elusive yet eminently real quality of life consid(cid:173)
`erations?
`'The clinical researcher evaluating new cyto(cid:173)
`toxic therapy for metastatic breast cancer is
`faced with a parallel set of concerns. Should the
`physician-scientist (as well as cooperative oncol(cid:173)
`ogy groups and research institutions) focus on
`issues affecting the therapeutic ratio of chemo(cid:173)
`therapy for metastatic breast cancer (maximize
`tumor response and minimize toxicity)? Or
`should the researcher view metastatic breast
`cancer as a model for regimens that might prove
`useful in an adjuvant (and potentially curative)
`setting? Finally, should the physician-scientist
`aim for the therapeutic "home run "-the cure
`of metastatic disease? Each of these approaches
`results in different, and sometimes mutually
`
`exclusive, research strategies. The problems
`and concerns listed above, frequently formu(cid:173)
`lated as testable scientific hypotheses, have
`guided much of clinical research in the chemo(cid:173)
`therapy of metastatic breast cancer in recent
`years.
`
`NEW AGENTS
`In the past decade no new chemotherapeutic
`agents received FDA approval for the treat(cid:173)
`ment of metastatic breast cancer. A positive
`new trend in phase II drug testing in patients
`with metastatic breast cancer is worth noting.
`Phase II agents had been routinely evaluated in
`heavily pretreated patients. Impaired perfor(cid:173)
`mance status and previous marrow damage
`frequently limited administration of effective
`doses of drug to tumors with significant levels of
`multi-drug and other mechanisms of resistance.
`However several promising new agents (Table
`1) have demonstrated significant clinical activity
`in phase II trials. 3-5,7-9 These new agents were
`tested in patients with no previous chemother(cid:173)
`apy for metastatic disease. This clinical environ(cid:173)
`ment represents a more realistic setting for
`evaluation of new agents.
`
`Anthracycline and Anthracene Analogs
`Doxorubicin, generally considered the single
`most active agent in metastatic breast cancer, is
`limited primarily by its hemopoietic and cardiac
`toxicity. Two novel agents, the anthracycline
`epirubicin and the anthracene mitoxantrone,
`appear to have less cardiotoxicity than doxorubi(cid:173)
`cin. Although mitoxantrone appears to be mod(cid:173)
`estly less active with regard to response rate
`than doxorubicin in doses with equivalent myelo-
`
`From' the Department of Medicine, Indiana University
`School of Medicine, Indianapolis, IN; and the Department of
`Medicine, Dana-Farber Cancer Institute, HOIvard Medical
`School, Boston, MA.
`Supported in part by US Public Health Sell'ice Grant No.
`P01CA-38493 and a grant from the Mather's Foundation.
`Address reprint requests to George W Sledge, !I; MD,
`Indiana University Hospital, 926 W Michigan St, Room 1730,
`Indianapolis, IN 46202.
`Copyright © 1992 by WB. Saunders Company
`0093-7754/92/1903-0001$05.00/0
`
`Seminars in Oncology, Vol 19, No 3 (June), 1992: pp 317-332
`
`317
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 1 of 16
`
`

`
`318
`
`SLEDGE AND ANTMAN
`
`Table 1. Novel Agents for the Treatment of Metastatic Breast
`Cancer: Trials in Previously Untreated Patients
`
`Agent
`
`Dose
`
`Refer·
`Response Rate ence
`
`Mitoxantrone 14 mglm 2 every 3 wk
`120 mg/m 2 every 3 wk
`Epirubicin
`30 mg/m 2 qd X4 every 3 wk
`Cisplatin
`Carboplatin 400 mg/m 2 every 3 wk
`Navelbine
`30 mg/m 2 every wk
`Amonafide
`CL-941
`
`35/99 (35%)
`4
`15/22 (69%)
`8
`9/19 (47%) 10
`4/20 (20%)
`7
`10/19 (52%)
`3
`6/26 (23%)
`5
`20/31 (64%) 11
`
`toxoicity, both as a single agent and in combina(cid:173)
`tion therapy, overall survival in prospective,
`randomized trials has not been compromised. 12-15
`This lack of correlation between response rates
`and overall survival may represent the more
`general inability of standard breast cancer che(cid:173)
`motherapy to significantly affect overall survival
`in patients with metastatic disease. Epirubicin is
`certainly at least equivalent to doxorubicin with
`regard to response rate and overall survival in
`prospective randomized trials when adminis(cid:173)
`tered in doses with equivalent myelotoxicity.!6-20
`Therefore, both agents represent reasonable
`alternative to doxorubicin in the patient with
`metastatic breast cancer, but are not commer(cid:173)
`cially available in the United States.
`CI-941 is an anthrapyrazole synthesized in an
`attempt to produce an agent with less cardiotox(cid:173)
`icity than doxorubicin. CI-941 has a chro(cid:173)
`mophore modification of the anthracenedione
`nucleus. The bis-hydroxyethylaminoalkyl side
`chains are identical to mitoxantrone replacing
`the glycone structure of doxorubicin. A phase II
`clinical trial of single agent CI-941 at a dose of
`50 mg/m2 given every 21 days in 31 patients with
`advanced breast cancer who had had no previ(cid:173)
`ous anthracycline or mitoxantrone. Fifteen had
`no previous cytotoxic chemotherapy and the
`remainder had had mainly cyclophosphamide,
`methotrexate, fluorouracil (CMF). Thirty pa(cid:173)
`tients were evaluable for response. Two pa(cid:173)
`tients (7%) had complete responses (of bulky
`intra-abdominal disease on computed tomogra(cid:173)
`phy and of a soft tissue deposit on the anterior
`thoracic wall). The response rate in patients
`with and without previous chemotherapy was
`63% and 64% respectively (95% confidence
`interval 46% to 81%). The median response
`
`duration was 28 weeks from start of treatment
`(range,4 to 70+ weeks).!!
`
`Cisplatin and Platinum Analogs
`Cisplatin, tested extensively in the 1970s and
`early 1980s in heavily pretreated metastatic
`breast cancer, demonstrated no appreciable
`clinical activity. In more recent trials in patients
`with no previous chemotherapy for metastatic
`disease, cisplatin has had significant activity,
`with response rates equivalent or superior to
`those of currently used agents in metastatic
`breast cancer.9,21 In combination therapy, it has
`been demonstrated to achieve response rates
`and overall survival times comparable to those
`of other standard chemotherapy regimens. 22.25
`Therefore, it can be considered a "new drug" in
`metastatic breast cancer. Its peculiar toxicities
`and relative inconvenience have limited its use
`in metastatic breast cancer, a disease tradition(cid:173)
`ally treated in the outpatient clinic, although
`not in the setting of high-dose chemotherapy
`and autologous bone marrow transplantation
`(as discussed below). There is a somewhat
`smaller body of experience with the platinum
`analog carboplatin. As front-line chemother(cid:173)
`apy, responses were observed in four of eight
`patients in one study and in four of 20 patients
`in the other.7,26 As carboplatin's predominant
`toxicity is hematopoietic, it is a potential candi(cid:173)
`date for dose intensification with autologous
`stem-cell transplantation (as discussed below).
`Three other agents, navelbene, amonafide,
`and taxol, have shorter pedigrees than those
`mentioned above, but are potentially quite prom(cid:173)
`ising. Navelbene, a vinca alkaloid, has been
`reported to have an objective response rate of
`52% in previously untreated patients, higher
`than that of other vinca alkaloids. 3,27 Amona(cid:173)
`fide, a new imide derivative of napthalic acid
`and a DNA intercalator, has significant activity
`in preyiously untreated patients, with a re(cid:173)
`sponse rate of 23%.5 As this agent has primarily
`hematologic toxicity, it is a potentially interest(cid:173)
`ing candidate for dose escalation. Taxol, an
`antimicrotubule agent, has produced a response
`rate of 56%.6 Some of these responsive patients
`had prior anthracycline therapy. These agents
`will require confirmatory studies before routine
`application.
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 2 of 16
`
`

`
`CHEMOTHERAPY FOR METASTATIC BREAST CANCER
`
`319
`
`CONVENTIONAL CHEMOTHERAPY FOR
`METASTATIC BREAST CANCER
`From the late 1960s through the mid 1970s
`clinical researchers developed chemotherapy
`regimens for metastatic breast cancer. While
`these regimens differed in terms both of num(cid:173)
`ber and type of chemotherapeutic agents em(cid:173)
`ployed, they shared common characteristics.
`(Representative combinations are shown in Ta(cid:173)
`ble 1.) These regimens were based on the
`superiority of combinations over single agents
`in the laboratory to decrease the emergence of
`drug resistance,28-31 and used agents with non(cid:173)
`overlapping toxicities. Generally administered
`in an outpatient setting, regimens were de(cid:173)
`signed to achieve maximal objective clinical
`response rates with acceptable toxicity.
`Standard dose chemotherapy regimens,
`whether doxorubicin (eg, 5-flourouracil, adri(cid:173)
`amycin, cyclophosphamide [F AC]) or metho(cid:173)
`trexate (eg, cyclophosphamide methotrexate
`5-flourouracil [CMF] or CMF vincristine pred(cid:173)
`nisone [VP]) based, have more similarities than
`differences. In previously untreated patients,
`these regimens produce 40% to 75% objective
`response rates complete response [CR] and
`partial response [PRJ, with median durations of
`response and survival of 6 to 12 months and 12
`to 24 months, respectively. These regimens
`frequently palliate the symptoms of metastatic
`breast cancer, but do not substantially extend
`the median survival and virtually never result in
`the cure of patients with metastatic breast
`cancer. Doxorubicin-based regimens generally
`have somewhat higher overall response rate
`than methotrexate-based regimens, although at
`greater cost in toxicity.32
`The results reported for standard chemother(cid:173)
`apy should be considered in context. The major(cid:173)
`ity of trials included a defined subset of patients
`with metastatic breast cancer. Clinical trials
`regularly exclude patients with coexisting medi(cid:173)
`cal and psychological illnesses, poor perfor(cid:173)
`mance status, organ system dysfunction, or older
`age. Furthermore, reported results for standard
`regimens initially were, and largely still are,
`derived in populations with no previous adju(cid:173)
`vant chemotherapy. Thus, many patients with
`predictably poor response rates or survival are
`excluded from reported clinical trials, but
`
`present very real therapeutic challenges in the
`physician's office. The real world of metastatic
`breast cancer treatment is therefore frequently
`even more disheartening than the already dis(cid:173)
`mal statistics reported in clinical trials.
`
`OLD WINE IN NEW BOTTLES
`Although there is general, if incomplete agree- .
`ment as to what constitutes conventional chemo(cid:173)
`therapy for metastatic breast cancer, there is
`little agreement as to how such therapy should
`be used. Recent research has studied the best
`means of using standard chemotherapy regi(cid:173)
`mens. Such research has focused on two salient
`questions in cancer chemotherapy: duration of
`therapy and dose intensity.
`
`Duration of Therapy
`An important question regarding standard
`therapy is the optimal duration of chemother(cid:173)
`apy. Should chemotherapy be administered con(cid:173)
`tinuously (ie, until disease progression), or inter(cid:173)
`mittently (with treatment only until maximum
`response, followed by retreatment at time of
`progression)? Opponents of continuous therapy
`argue that treatment past a predefined goal
`(such as an objective remission, or palliation of
`symptoms) results in impairment of quality of
`life secondary to cumulative drug toxicity. Be(cid:173)
`cause metastatic breast cancer is incurable with
`standard chemotherapy, intermittent treatment
`(at such time as there is clear disease progres(cid:173)
`sion or symptomatic worsening) would avoid
`such toxicity. Conversely, opponents of intermit(cid:173)
`tent therapy argue that continuous therapy
`might delay relapses, increase overall response
`rates (hence improve palliation of disease), and
`potentially prolong survival.
`Three recently-reported trials have addressed
`this question (Table 2). The Australia-New
`Zealand Breast Cancer Trials Group random(cid:173)
`ized patients to receive standard chemotherapy
`regimens (doxorubicin and cyclophosphamide
`or cyclophosphamide, methotrexate, fluorou(cid:173)
`racil, and prednisone) either intermittently or
`continuously.33 Intermittent therapy, comprised
`of three cycles of therapy, with re-treatment
`upon progression, proved inferior to continuous
`therapy with regard to overall response rates,
`although survival was not adversely affected.
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 3 of 16
`
`

`
`320
`
`Regimen
`
`Table 2. Randomized Trials of Treatment Duration in Metastatic Breast Cancer
`as
`
`Rando'mization
`
`TIP
`
`PValue
`
`SLEDGE AND ANTMAN
`
`Reference
`
`33
`
`34
`
`35
`
`D 50 mg/m2
`C 750 mg/m 2 every 21 days
`or
`C 100 mg/m2 daily for 14 days
`M 40 mg/m2 days 1 and 8
`F 600 mg/m2 days 1 and 8
`Pr 40 mg/m2 daily for 14 days
`every 28 days
`
`C 600 mg/m 2
`E 60 mg/m2 every 3 wk
`F 600 mg/m2
`
`+
`T 30 mg orally each day
`
`Mx 14 mg/m2 every 3 wk for 4
`courses
`
`Continue to prog ression
`v
`3 cycles; re-treat upon
`progression
`
`·6 mo
`
`10.7 mo
`
`4mo
`
`9.4mo
`
`18mo
`
`6mo
`
`v
`
`52wk
`
`39wk
`
`67wk
`
`58wk
`
`Retreat upon progression
`
`26wk
`
`52wk
`
`v
`Continue to progression
`
`22wk
`
`49wk
`
`0.19
`
`0.068
`
`NS
`
`Abbreviations: D, doxorubicin; C, cyclophosphamide; M, methotrexate; F, 5-fluorouracil; Pr, prednisone; T, tamoxifen; Mx,
`mitoxantrone; TTP, time to progression; as, overall survival.
`
`However, quality of life was superior in patients
`receiving continuous therapy. A similar trial by
`the Danish Breast Cancer Cooperative Group
`randomized patients with estrogen and progest(cid:173)
`erone-receptor-negative cancers to receive cyclo(cid:173)
`phosphamide, epirubicin, and fluorouracil ev(cid:173)
`ery 3 weeks for either 6 or 18 cycles.34 All
`patients also received tamoxifen. The median
`time to progression was significantly longer (and
`survival marginally longer) in the group receiv(cid:173)
`ing continuous therapy.
`In a similar trial, Harris et at treated patients
`with four cycles of single-agent mitoxantrone.35
`Responding patients were then randomized to
`either continuous therapy with mitoxantrone, or
`to re-treatment upon progression. Patients on
`continuous mitoxantrone averaged seven cycles
`of therapy. There was no survival advantage for
`patients receiving continuous therapy; quality of
`life was not assessed in any systematic fashion.
`This result must be viewed cautiously because
`the overall response rate to four cycles of
`single-agent mitoxantrone (30%) and the num(cid:173)
`ber of additional cycles was sufficiently low that
`any survival benefit or palliation of symptoms
`might be difficult to demonstrate. Similarly,
`Glaholm et al have pointed out that this trial
`lacked the statistical power to demonstrate even
`moderate survival benefits. 36
`
`These three studies suggest that prolonging
`duration of therapy is likely to have at best a
`marginal effect on overall length of survival for
`women with metastatic breast cancer, but may
`improve the quality of that time, by diminishing
`breast-cancer-related symptoms and delaying
`relapse. The appropriate duration of therapy
`must be considered still an open issue. The
`Eastern Cooperative Oncology Group is com(cid:173)
`pleting protocols in which maintenance therapy
`after establishment of complete remission is
`randomized. These protocols should provide
`additional information regarding the duration
`of therapy.
`
`Dose Intensity
`A relationship between dose and clinical
`response has long been recognized, for both
`individual and combination chemotherapeutic
`agents.37 Laboratory models have been recently
`reviewed.38 Systematic analysis of dose-re(cid:173)
`sponse relationships in the clinic has proven
`problematic, and until recent years few at(cid:173)
`tempts had been made to quantify dose(cid:173)
`response relationships in patients with meta(cid:173)
`static breast cancer.
`In 1984 Hryniuk and Bush introduced the
`concept of dose intensity for the purpose of
`quantifying dose-response effects, using meta-
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 4 of 16
`
`

`
`CHEMOTHERAPY FOR METASTATIC BREAST CANCER
`
`321
`
`static breast cancer as a clinical modeP9 They
`argued that quantitation of dose-response ef(cid:173)
`fects required that dose be viewed as a function
`of the time (ie, dose intensity). In their analyses,
`dose intensity is expressed arbitrarily as units of
`drug administered per square meter per week.
`In trials in which multiple agents were used,
`Hryniuk and Bush expressed the dose intensi(cid:173)
`ties of individual drugs compared with those of
`a standard regimen. These relative dose intensi(cid:173)
`ties are then added and divided by the number
`of drugs in the regimen to produce the average
`relative dose intensity for a regimen. These
`average dose intensities are then compared with
`those of an arbitrary "standard" regimen (eg,
`the Cooper regimen for methotrexate-based
`regimens, or Bull and Tormey cyclophospha(cid:173)
`mide, doxorubicin, fluorouracil (CAF) for doxo(cid:173)
`rubicin-based regimens.
`In metastatic breast cancer, the Hrynuik(cid:173)
`Bush dose intensity analysis suggested that dose
`intensity correlated strongly with response, and
`that response in turn correlated significantly
`with survival. This relationship was seen for
`both methotrexate- and doxorubicin-based regi(cid:173)
`mens. Based on this analysis, the authors sug(cid:173)
`gested that chemotherapy regimens should be
`designed to maximize dose intensity. This pro(cid:173)
`vocative thesis has had a profound influence on
`the design of both individual regimens and
`group trials. Emphasizing overall dose adminis(cid:173)
`tered over time, rather than the peak dose, has
`led to the development of regimens in which
`complex scheduling and repetitive therapy re(cid:173)
`place high-dose, single bolus infusions. This
`approach stands in contrast to the philosophy
`underlying high-dose chemotherapy and autolo(cid:173)
`gous bone marrow transplantation, as discussed
`below. Two such dose-intensive regimens (the
`Duke AFM regimen and the Johns Hopkins
`16-week regimen) are shown in Table 3.39,41
`The dose-intensity hypothesis and methodol(cid:173)
`ogy have been debated extensively based on
`both practical and theoretical concerns. 13,42 Dose
`intensity, as calculated by Hryniuk and Bush,
`necessarily oversimplifies a complex problem by
`making numerous assumptions. These include
`(1) that all drugs in a given regimen are thera(cid:173)
`peutically equivalent in dose-intensity calcula(cid:173)
`tions; (2) that synergy and cross-resistance be(cid:173)
`tween drugs play no role; (3) that peak drug
`
`Table 3. High Dose Intensity Chemotherapy Regimens
`
`AFM Regimen: Cycles repeated every 21 days
`Fluorouracil 750 mg/m 2 CI for 5 days
`Doxorubicin 25 mg/m 2 days 3-5 IV·
`Methotrexate 250 mg/m 2 day 151V
`Folinic acid 12.5 mg orally q6h X 6 D15
`Hopkins Regimen: Cycles repeated every 14 days
`Cyclophosphamide 100 mg/m 2 orally for 7 days
`Doxorubicin 40 mg/m 2 IV day 1
`Vincristine 1 mg IV day 1
`Methotrexate 100 mg/m 2 IV day 1
`Fluorouracil 600 mg/m 2 day 2
`Leucovorin rescue
`Fluorouracil 300 mg/ m 2 by CI days 8 and 9
`
`Abbreviations: CI, continuous infusion; IV, intravenously,
`
`concentrations are not so important as the area
`under the curve (AUC); (4) that scheduling has
`no importance other than as it relates to total
`dose intensity; and (5) that duration of therapy
`is inconsequential. Furthermore, the Htyniuk(cid:173)
`Bush retrospective analysis relies heavily on the
`assumption that all reported studies are compa(cid:173)
`rable with regard to entrance criteria, prognos(cid:173)
`tic'variables, and analysis of response and sur(cid:173)
`vival.
`Given the concerns regarding the methodol(cid:173)
`ogy in Htyniuk-Bush type of retrospective anal(cid:173)
`ysis, it seems appropriate to consider results
`from randomized clinical trials in which dose
`intensity is the sole or most important vari(cid:173)
`able.43-51 Some of these trials are shown in Table
`4.
`
`These trials are difficult to interpret because
`the increased doses planned varied from 10% to
`to threefold over the low dose arms.
`two-
`Because the serum levels for a given dose of
`drug commonly vary fivefold, the serum levels of
`drug achieved on these trials must overlap
`considerably. In addition, the actually delivered
`dose (frequently not included in the manu(cid:173)
`script) is often not significantly different from
`that delivered on the lower dose arm.
`In many cases, these trials have shown an
`incre'ase in response rates for regimens with
`greater dose intensity. Only two trials show a
`significantly increased overall survival. In a trial
`by Tannock et aI, randomization resulted in an
`excess of patients with brief durations between
`initial diagnosis and relapse on the low-dose(cid:173)
`intensity arm causing the authors to advise
`caution in the interpretation of their observa(cid:173)
`tions.49 Carmo-Pereira et al demonstrated a
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 5 of 16
`
`

`
`Table 4. Randomized Dose Intensity Trials in Metastatic Breast Cancer
`
`ORR
`
`PValue
`
`PValue
`
`Reference
`
`Regimens
`
`RDI
`
`D 70 mg/m2 every 3 wk
`v
`D 35 mg/m2 every 3 wk
`
`E 40 mg/wk
`
`E 20 mg/w
`
`"Good Risk" patients
`D 75 mgt m2 every 3 wk
`v
`D 60 mg/m2 every 3 wk
`v
`D 40 mgt m2 every 3 wk
`
`"Poor Risk" patients
`D 50 mg/m2 every 3 wk
`v
`D 25 mg/m2 every 3 wk
`
`C 600 mg/m2 every 3 wk
`E 60 mg/m2 every 3 wk
`v
`C 600 mg/m2 every 3 wk
`E 40 mg/m2 every 3 wk
`
`1.87
`
`1.33
`
`1.5
`
`N
`
`24
`
`24
`
`26
`
`27
`
`44
`
`27
`
`32
`
`34
`
`34
`
`58
`
`67
`
`102
`
`58
`
`25
`
`34
`
`37
`
`25
`
`37
`
`32
`
`24
`
`6
`
`40
`
`40
`
`41
`
`<0.02
`
`NS
`
`OS
`
`20 mo
`
`8mo
`
`42wk
`
`84wk
`
`<0.01
`
`NS
`
`NS
`
`NR
`
`NR
`
`NR
`
`NS
`
`NR
`
`NR
`
`NR
`
`NS
`
`44
`
`45
`
`48
`
`43
`
`E 100 mg/m2 every 3 wk
`Pr 50 mg 5 times a day every 3 wk
`v
`E 50 mg/m2 every 3 wk
`Pr 50 mg 5 times a day every 3 wk
`
`P 120 mg/m2 every 3 wk
`v
`P 60 mg/m2 every 3 wk
`
`C 600 mg/m2
`M 40 mg/m2 every 3 wk
`F 600 mg/m2
`
`v
`
`C 300 mg/m2
`M 20 mg/m2 every 3 wk
`F 300 mg/m2
`
`F 500 mg/m2 days 1-5
`D 70 mg/m2 day 1
`C 1200 mg/m2 day 1
`v
`F 500 mg/m2 days 1 & 8
`D 50 mg/m2 day 1
`C 500 mgt m2 day 1
`
`E 50 mg/m2 days 1 & 8
`C 500 mg/m2 every 3 wk
`F 500 mg/m2
`
`v
`E 50 mg/m2 day 1
`C 500 mg/m2 every 3 wk
`F 500 mg/m2
`
`V 0.625 mg/m2/wk
`M 15 mg/m2/wk
`F 300 mg/m2/wk
`C 60 mg/m2 orally every day
`Pr 30 mg/m once a day for 14 days
`20 mg/m2 once a day for 14 days
`10 mg/m2 daily
`v
`C 120 mg/m2 IV 5 times
`M 4 mg/m2 IV 5 times
`F 180 mg/m2 IV 5 times
`V 0.625 IV days 1 & 5
`Pr 40 mg/m2 5 times
`Repeat every 28 days
`
`0.006
`
`44wk
`
`46wk
`
`NS
`
`'NS
`
`NR
`
`NR
`
`15.6mo
`
`0.03
`
`0.026
`
`12.8mo
`
`20mo
`
`20 mo
`
`NS
`
`NS
`
`<0.02
`
`NR
`
`NS
`
`100
`
`19
`
`18
`
`67
`
`66
`
`32
`
`27
`
`160
`
`23
`
`21
`
`0
`
`30
`
`11
`
`78
`
`78
`
`67.2
`
`43.1
`
`106
`
`59
`
`14mo
`
`<0.05
`
`NS
`
`98
`
`40
`
`14mo
`
`2.5
`1.4
`2.4
`
`2.0
`1.0
`1.0
`
`1.33
`2.8
`
`46
`
`52
`
`49
`
`47
`
`50
`
`51
`
`Abbreviations: C, cyclophosphamide; D, doxorubicin; F, fluorouracil; M, methotrexate; E, epirubicin; P, cisplatin; Pr, prednisone; NS, not significant; NR,
`not reported; ROi, relative dose intensity of more dose intense: less dose intense regimen for individual drugs in regir.len; ORR, objective response rate; as,
`overall survival (in weeks or months); IV, intravenously.
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 6 of 16
`
`

`
`CHEMOTHERAPY FOR METASTATIC BREAST CANCER
`
`323
`
`statistically significant difference in medial} sur(cid:173)
`vival (8 months versus 20 months) for patients
`receiving two doses of doxorubiCin. 44 However,
`this trial suffered from its relatively small num(cid:173)
`ber of patients, inviting the reasonable criticism
`that the results might represent a chance out(cid:173)
`come. The overall lack of evidence for an
`improvement in survival for the higher-dose(cid:173)
`intensity arms may reflect the relatively minor
`differences in dose administered, or the under(cid:173)
`lying-and more profound-lack of any major
`effect of standard chemotherapy regimens on
`median survival in metastatic breast cancer.
`If, as these randomized trials suggest, modest
`increments in dose intensity produce at best
`only modest effects on survival, then clinicians
`may reasonably ask what benefit is achieved by
`increasing the response rate. Prolongation of
`survival is an important, but not the only, goal of
`the medical oncologist. The trial by Tannock et
`at suggests that (within the limits of that trial)
`increased response rates are associated with an
`improved quality of life.49 Assessments of qual(cid:173)
`ity of life are lacking in most randomized
`metastatic breast cancer trials involving chemo(cid:173)
`therapy, an unfortunate omission given the
`toxicity of the regimens and the incur ability and
`morbidity of the disease.
`
`DOSE-INTENSIVE TREATMENT WITH
`HEMATOPOIETIC GROWTH FACTOR SUPPORT
`The introduction of hemopoietic growth fac(cid:173)
`tors may facilitate the clinical evaluation of the
`importance of dose intensity in a non-transplant
`setting. In a small pilot trial, Bronchud et at
`have reported that use of granulocyte colony(cid:173)
`stimulating factor (G-CSF) allows the adminis(cid:173)
`tration of outpatient doxorubicin at high doses
`of up to 120 mg/m2 on an every 12-14-day
`schedule, with a high complete and overall
`response rate in patients with metastatic breast
`cancer.53
`
`DOSE-INTENSIVE TREATMENT WITH
`HEMATOPOIETIC STEM CELL SUPPORT
`Rationale
`In laboratory models of breast cancer and
`other malignancies, the delivery of the highest
`possible doses of chemotherapy is essential to
`achieving curative therapy. In theory, experimen-
`
`tal and clinical data suggest that breast cancer
`recurs despite an initial response to chemother(cid:173)
`apy because of resistance to th{!chemotherapy
`drugs. Clinically, dose escalation of agents ac(cid:173)
`tive in breast cancer is often limited by myelosup(cid:173)
`pression, leading a number of investigators to
`study autotransplants to ensure prompt marrow
`recovery after high doses of chemotherapy.
`Criteria for curative marrow transplant include
`a malignancy responsive to cytoreductive ther(cid:173)
`apy, effective treatment whose limiting toxicity
`is marrow failure, performance of the proce(cid:173)
`dure early in the disease course when there is
`minimal tumor burden and drug resistance, and
`a source of hematopoietic stem cells free of
`clonogenic tumor cells.54
`Alkylating agents exhibit a steep linear-log
`dose-response curve, and non-hematologic tox(cid:173)
`icity varies among the different agents. While
`many antimetabolites and other cycle-specific
`agents are schedule dependent in experimental
`and clinical trials, alkylating agents are non(cid:173)
`cycle specific and less schedule dependent.
`Significant resistance to alkylating agents is
`believed not to result from either gene amplifi(cid:173)
`cation or pleiotropic drug resistance (docu(cid:173)
`mented mechanisms of resistance against anti(cid:173)
`metabolites, antibiotics, and vinca alkaloids,
`respectively).55 Initially alkylating agents were
`thought to have similar mechanisms of action
`and cross-resistance. However, in 1978 Schabel
`reported that in in vivo experimental systems,
`alkylating agents were for the most part non(cid:173)
`cross-resistant.56 He subsequently demonstrated
`substantial synergy of alkylating agents with
`other alkylating agents, antimetabolites, or anti(cid:173)
`biotics.56-59 Full use of the dose-response curve
`for many alkylating agents is limited by myelo(cid:173)
`suppression. Even low levels of resistance (ten(cid:173)
`fold) are difficult to produce in the laboratory
`(compared with 1,OOO-fold resistance for some
`nonalkylating agents).60 Resistance in labora(cid:173)
`tory models to alkylating agents can often be
`overcome by using a five-to-tenfold higher dose,
`possible with stem cell support. Although most
`alkylating agents at conventional doses share a
`common dose-limiting toxicity (myelosuppres(cid:173)
`sion), marrow autografting permits dose escala(cid:173)
`tion to the level of dose-limiting non-hemato(cid:173)
`logic toxicity. The dose-limiting organ toxicity of
`
`NOVARTIS EXHIBIT 2067
`Par v Novartis, IPR 2016-00084
`Page 7 of 16
`
`

`
`324
`
`SLEDGE AND ANTMAN
`
`the various alkylating agents used at high dose
`differs substantially (pulmonary fibrosis and
`hepatic toxicity for carmustine [BeNU], car(cid:173)
`diotoxicity for cyclophosphamide, nephrotoxic(cid:173)
`ity for cisplatin, and stomatitis and enterocolitis
`for melphalan). Thus nonoverlapping toxicity
`may allow a combination of alkylating agents at
`full or nearly full transplant doses.
`
`Results of Dose-Intense Trials
`Initial trials involving the use of high-dose
`chemotherapy and autologous bone marrow
`support parallelled in many ways the initial use
`of standard phase II agents in metastatic breast
`cancer, in that they frequently used single agents
`administered to heavily pre-treated patients.
`Subsequent trials used combination dose-in(cid:173)
`tense regimens, and were administered to pro(cid:173)
`gressively less heavily pre-treated patients.
`
`Failed or RefractOlY Breast Cancer
`Single agents. There are 12 studies of high
`doses of alkylating agents and five of non(cid:173)
`alkylating agents followed by autotransplanta(cid:173)
`tion in women with advanced breast cancer. The
`response rates for alkylating agents was 39%
`compared with 16% for non-alkylating agents.
`
`All complete responses were obtained with
`alkylating agents (melphalan or thiotepa), al(cid:173)
`though partial responses were obselved with all
`drugs evaluated (Table 5).
`Combinations of drugs with or without total
`body irradiation. Multiple combinations of
`high-dose chemotherapy with or without total
`body irradiation (TEl) have been reported in
`patients with failed or refractory breast cancer.
`The response rate for radiation containing regi(cid:173)
`mens (12-19, 63%) was not significantly dif(cid:173)
`ferent from that of regimens without radiation
`(152-219,69%) (Table 6).
`
`Metastatic Breast Cancer
`
`Autotransplants inpatients with no priorchemo(cid:173)
`therapy for metastatic breast cancer. There are
`four studies of combination chemotherapy in 53
`previously untreated patients with inflamma(cid:173)
`tory or metastatic breast cancer. (Patients may
`have received prior adjuvant therapy) (Table 7).
`Autotransplants in patients with breast cancer
`responding to standard dose chemotherapy. Mul(cid:173)
`tiple regimens have been used in patients re(cid:173)
`sponding to induction therapy. A total of 59%
`of these patients have achieved complete re(cid:173)
`sponses and 28% were in continuous complete
`response at the time of data analysis (Table 8).
`
`Table 5. High-Dose Trials of Single Agents in Failed or Refractory Breast Cancer
`
`Author
`
`Institution
`
`Drug
`
`Non-aikyiating Agents
`Fraschini61
`Mulder62
`Wolff63
`Total
`Ariel 64
`Tannir65
`Total
`Alkylating Agents
`Shea66
`Peters
`Tannir67
`Schilcher68.69
`LeMaistre70
`Lazarus72
`Slease73
`