High-Dose Chemotherapy and
`Autologous Bone Marrow or Stem
`Cell Reconstitution for Solid Tumors
`
`Abstract.—High-dose chemotherapy—in conjunction
`with the transplantation of either mononuclear cells
`harvested from the marrow or CD 34+ cells harvested
`from the peripheral blood—has proved effective in
`curing certain patients with leukemia, lymphoma, and,
`to a lesser extent, multiple myeloma. Though the CD
`34+ therapy is a relatively new treatment and the
`mononuclearcell therapy is more standard, both have
`been successfully used to reconstitute lethally damaged
`hematopoietic stem cells. Allogeneic transplants have
`been more effective than autologous transplants against
`tumors, but they also pose a greater hazard of death
`from complications, graft-versus-host disease, and
`infections. More currently, this approach has been used
`in patients with certain solid tumors, either in a
`metastatic or recurrent disease setting or as an adjuvant
`to surgery and/or standard doses of chemotherapy in
`patients with a known high risk of recurrence.
`Unfortunately, the majority of the studies about the
`impact of this therapy have been small and nonran-
`domized against standard therapy, and they have
`encompassed diverse populations of patients. This
`makes comparisons with contemporary standard—
`dose approaches—already problematic from a statis-
`tical point of view—even more dangerous because ofthe
`dissimilarity of the groups being compared. Particularly
`in the high-risk adjuvantsetting, data suggest that those
`patients that meet the eligibility criteria for high-dose
`therapy and transplantation exhibit the prognostic fac-
`tors for a positive outcome. When one compares these
`results with those of a more heterogeneous group of
`patients treated with conventional therapy, the conclu-
`sion might be drawn that high-dose therapy is superior
`to standard therapy, when a longer follow-up of the
`Curr Probl Cancer, May/June 1998
`
`138
`
`(cid:43)(cid:82)(cid:86)(cid:83)(cid:76)(cid:85)(cid:68)(cid:3)(cid:89)(cid:17)(cid:3)(cid:42)(cid:72)(cid:81)(cid:72)(cid:81)(cid:87)(cid:72)(cid:70)(cid:75)(cid:3)
`Hospira v. Genentech
`(cid:44)(cid:51)(cid:53)(cid:21)(cid:19)(cid:20)(cid:26)(cid:16)(cid:19)(cid:19)(cid:27)(cid:19)(cid:24)(cid:3)
`IPR2017-00805
`(cid:42)(cid:72)(cid:81)(cid:72)(cid:81)(cid:87)(cid:72)(cid:70)(cid:75)(cid:3)(cid:40)(cid:91)(cid:75)(cid:76)(cid:69)(cid:76)(cid:87)(cid:3)(cid:21)(cid:19)(cid:25)(cid:22)
`Genentech Exhibit 2063
`
`

`

`patients in the study will show this to be untrue. Thus
`there is a plea from clinicians and physicians conducting
`trials for prospective, randomized trials that would
`allow a fair comparison between high-dose therapy in
`combination with transplant procedures and a more
`conventional, standard chemotherapy, which is often
`less toxic and definitely less expensive.
`This article reviews the data for transplantation in
`four tumors: breast cancer, ovarian cancer, small-cell
`lung cancer, and germ cell testis cancer. There is such a
`small number of randomized trials that an attempt
`must be made to compare these small high-dose therapy
`studies with similar, though not identical, large studies
`of conventional therapy. This article attempts to make
`those comparisons, and several conclusions are drawn,
`which are detailed below.
`First, few data support the use of high-dose chemo-
`therapy in any patient with recurrent and drug-
`resistant breast cancer or ovarian cancer. Similarly, few
`data support the use of high-dose approaches for pa-
`tients with extensive small-cell lung cancer. For patients
`with metastatic breast cancer that has responded com-
`pletely to conventional chemotherapy, no data suggest a
`survival advantage for the immediate consolidation of
`that response with high-dose chemotherapy. The only
`trial addressing this issue found that immediate trans-
`plantation led to a better disease-free survival rate, but
`overall survival, as compared with that of patients who
`received transplants at relapse, was not affected, and the
`study did not address the issue of the relative merits of
`conventional chemotherapy in either case. The only
`study of high-dose versus conventional chemotherapy
`was Statistically underpowered, and it showed poorer-
`than-anticipated outcomes in the patients who received
`conventional therapy. Ongoing or recently completed
`trials will, it is hoped, address the many unanswered
`questions in this area.
`For patients with high-risk, non-metastatic breast
`cancer, no completed and analyzed phase III ran-
`domized studies address the relative merits of conven-
`tional versus high-dose therapy. Early results from
`
`Curr Probl Cancer, May/June 1998
`
`139
`
`

`

`high-dose approaches suggest a better disease-free
`survival than has typically been observed with standard
`chemotherapy, but the patients who have received
`transplants are often highly selected and may have
`favorable prognostic factors as compared with the more
`heterogeneous groups that have received conventional
`therapy over the years. Several trials underway address
`the value of high-dose therapyin early stage, high-risk
`breast cancer patients (the group mostlikely to benefit
`from this type of therapy, based on the knowledge
`gleaned from the studies of transplants in patients with
`hematologic tumors), and these results are anxiously
`awaited by patients and health care professionals alike.
`In ovarian cancer, where high-dose chemotherapy and
`transplantation is becoming increasingly more common,
`many small phase II trials have shown high response
`rates but short response durations, and the data cur-
`rently available make it difficult to maintain enthusiasm
`for this approach in patients with drug-resistant or
`large-volume disease. Several trials underway random-
`ize patients with demonstrated drug sensitivity and a low
`volume of tumors (achieved either surgically or with con-
`ventional chemotherapy) to receive either high-dose
`therapy or more conventional consolidation approaches.
`Until these trials are complete, the value of high-dose
`therapy will remain unknown.
`There are little data regarding transplantation in
`patients with limited-stage small-cell lung cancer. A sin-
`gle randomized trial demonstrated an outcome advan-
`tage for high-dose therapy, but the patient numbers
`were too small to allow any definitive conclusions to be
`made. Large randomized trials in this population are
`needed to address the value of high-dose approaches.
`The preferred population in which to perform these tri-
`als would be patients with limited disease who respond
`to initial conventional chemotherapy and are subse-
`quently randomized to receive conventional consolida-
`tion with chest and cranial radiation or high-dose ther-
`apy with the same radiation. Survival would be the only
`pertinent endpoint. No trials of this sort are currently
`underway.
`
`140
`
`Curr Probl Cancer, May/June 1998
`
`

`

`For patients with testis cancer, data suggest a clear
`survival benefit from high-dose chemotherapy and
`transplantation. Two groups of patients still have
`very poor outcomes—those with mediastinal primary
`disease and those who progress or relapse within 4
`weeks of completion of standard, cisplatin-based
`therapy. It is unclear if high-dose approaches can sal-
`vage these patients. However, in patients whose dis-
`ease recurs at a point more distant from conventional
`therapy, there is a potential for a long-term, disease-
`free survival rate that may be as much as or more
`than 50% better than that of patients who have been
`treated with conventional salvage approaches. Final-
`ly, it is now possible to identify patients who are at a
`high risk of a poor outcome and who are candidates
`for high-dose approaches as part of initial therapy.
`These patients are currently being studied in an
`intergroup, randomized trial.
`
`a: is a plethora of data documenting the value of high-dose
`
`chemotherapy followed by the rescue of lethally damaged
`hematopoietic stem cells with either autologous mononuclear cells
`harvested from the patient’s marrow or mononuclear cells derived from
`the patient’s blood, with the most success being apparentin cases of acute
`leukemias, chronic myelogenous leukemia, Hodgkin’s and non-
`Hodgkin’s lymphoma, or multiple myelomas. When transplantation
`occurs in patients with these hematologic malignant tumors, there is a
`variable—but real—curerate that is achieved over and above that which
`would occur in these patients if they did not receive transplants.
`In cases of chronic myelogenous leukemia,transplantation is the only
`knowncurative form of therapy. Similarly, in cases of acute leukemias
`and lymphomasfor which initial primary therapy has failed, transplan-
`tation offers the only real chance of long-term disease-free survival.
`Early data regarding myelomaare suggestive of a cure potential with
`transplantation. These topics have been recently reviewed.'! Some of
`these studies used allogeneic marrow whenautologous donation was not
`acceptable because of tumor involvement. Such transplants have typi-
`cally shown a better outcome than autologous donations, although there
`have been no comparativetrials. This better outcomeis postulated to be
`caused by a graft-versus-tumor effect,
`in which the engrafted cells
`
`Curr Probl Cancer, May/June 1998
`
`141
`
`

`

`recognize the host tumorcells as foreign and attempt to destroy them,
`and that can be demonstrated in vitro in mixing experiments. It has been
`demonstrated in patients with malignant hematologic disease that the
`patients with drug-sensitive disease and low tumor burden will enjoy the
`best outcome following either kind of transplantation.
`Patients who received transplants at the time of clinical remission but
`had a high risk of relapse or patients treated into a clinical complete
`remission with standard drug doses before undertaking transplantations
`have better outcomes than patients treated at the time of relapse, when
`there is clinically evident disease, or both. Thus the basic tenets of trans-
`plantation that were learned as a result of working with patients with
`malignant hematologic disease were to select patients for transplantation
`whose disease was drug-sensitive and in complete clinical remission.
`This difficult lesson was learned after unsuccessful outcomesofthe treat-
`ment regimen, high complication rates in those patients with extant dis-
`ease, or both. Unfortunately, it does not appear that these basic rules were
`followed whentransplantation beganto be used for the treatmentof solid
`tumors.
`
`Breast cancer is the solid tumor most commonly treated with trans-
`plantation. A recent study of breast cancer patients concluded that
`patients in whom the disease had metastisized and who also received
`transplants had 3-year probabilities of disease-free survival of 7%, 13%,
`and 32% if the transplant wascarried out in patients with progressive dis-
`ease, partially responding disease, or completely responding disease,
`respectively, after pre-transplant chemotherapy.’ The cases reported to
`the Autologous Blood and Marrow Transplant Registry (ABMTR)from
`1989 to 1995 indicate that breast cancer became the most frequent indi-
`cation for autologous transplantation, and the use of transplantation in
`breast cancer cases increased sixfold during that time interval. Annually,
`from 1992 on, more breast cancer patients have received transplants than
`have patients with all types of lymphoma. Casesof transplantation report-
`ed to the ABMTRprobably represent about 50% ofthe actual transplan-
`tations performed.’ In 1994, 1,513 transplants were reported to the
`ABMTR,suggesting that as many as 3,000 procedures were performed
`that year.
`Even with the conservative estimate of the cost of an individual trans-
`plant at $50,000, the annual total cost of all of the transplants performed
`is now in excess of $150,000,000. Many third-party payers have refused
`to pay for this procedure, based on their assessmentthat there is incon-
`clusive evidence of any benefit to the patient. Multiple legal battles have
`ensued that are frequently decided in favor ofthe patient. Many physi-
`
`142
`
`Curr Probl Cancer, May/June 1998
`
`

`

`cians believe that this procedure is of value and are willing to testify on
`behalf of the patients. Henderson surveyed physicians in the United
`States and found that 80% felt bone marrow transplantation should be
`offered to women with metastatic breast cancer, despite the indefinite
`proofof benefit. Canellos felt that clinical trials have not yet answered the
`question of the value of transplantation in breast cancer, and he pointed
`out that only large randomized trials would be sufficient to do so.4
`Antmansaid in her survey that there has been a trend toward performing
`transplants in breast cancerpatients with earlier stage, drug-sensitive dis-
`ease, and in small uncontrolled studies. Only 11% of patients with
`advancedlocal disease and < 1% of patients with metastatic disease,all
`who had received transplants, were entered into a randomized controlled
`trial. Although this shift to performing transplants on patients whosedis-
`ease has a better prognosis is a welcome change,it has the confounding
`problem of making early data analysis overly optimistic when contempo-
`rary, high-dose nonrandomized studies are compared with standard-dose
`chemotherapy studies in a less highly selected and often worse prognosis
`group.
`Although breast cancer has been the most extensively studied, there
`have also been studies of high-dose therapyin othersolid tumors that typ-
`ically display some degree of sensitivity to conventional doses of drugs.
`Thus this article will review the trials of high-dose chemotherapy and
`autologous transplantation in breast cancer, ovarian cancer, small-cell
`lung cancer, and germ cell testis cancer. Transplants performed in other
`tumors, such assoft tissue sarcoma,will not be reviewedhere, since there
`is a limited amountof data available on those subjects. Data about pedi-
`atric solid tumors will not be included either, as this review will be limit-
`ed to the discussion of the adult patient.
`
`Breast Cancer
`
`In orderto fairly andcritically evaluate the role of transplantation in breast
`cancer, one can look only at prospective, randomized, controlled trials—
`the gold standard for decision-making in oncology—or one may also
`include phaseII trials, in which there is an adequate and comparable con-
`trol group with similar parameters of biologic behavior and in which stan-
`dard therapy was administered. Thesetrials can be looked at whether one
`is assessing response rate, progression-free survival, or overall survival,
`since in both metastatic and adjuvant settings known factors influence
`these outcome measures. Some factors may not be known, and random-
`ized trials are less likely to inadvertently include an excessof either lower
`or higher risk patients in any arm of the study, thus skewingtheresults.
`
`Curr Probl Cancer, May/June 1998
`
`143
`
`

`

`A recent report by Rahman? makes this pointin patients with metasta-
`tic breast cancer. In that retrospective analysis, which spannednineyears,
`1,581 patients were enrolled in 18 studies in which doxorubicin-contain-
`ing chemotherapy was administered in standard doses. Applying com-
`moneligibility criteria for entry into contemporary transplantation stud-
`ies (age < 60, performancestatus < 2, response to recent chemotherapy,
`and normal end organ function), the authors of the study found 645
`patients who were eligible and 936 whowereineligible for one or more
`reasons. The following results were found for the eligible and ineligible
`groups, respectively: complete response, 27% and 7%; median progres-
`sion-free survival, 16 months and 8 months (p < 0.001); median overall
`survival, 30 months and 17 months (p < 0.001); 5-year survival, 21% and
`6%; and 10-year survival, 7% and 2%. The authors concluded that
`“encouraging results of single-arm trials of high-dose chemotherapy
`could partially be due to selection of patients with better prognosis and
`further stresses the importance of completing ongoing randomized trials.”
`In a similar adjuvant study® of patients with high-risk disease (> 10 pos-
`itive nodes), 265 patients were identified from a single institution over a
`20-year period. Of these 265, 171 received standard adjuvant chemo-
`therapy. These patients were grouped into those eligible for transplanta-
`tion (n = 128) or those ineligible (n = 43), using criteria similar to those
`in Rahman’s study of metastatic disease. Additionally, a contemporary
`cohort of patients (n = 39) that met transplantation eligibility was select-
`ed, and that group was treated with high-dose therapy. Outcomeresults
`for the two cohorts of eligible and ineligible historical patients, respec-
`tively, were: 5-year disease-free survival, 37% and 16% (p < 0.05); and
`5-year overall survival, 55% and 23% (p < 0.01). Wheneligible histori-
`cal patients were compared with contemporary treated patients, no dif-
`ferences were observedin either disease-free survival or overall survival.
`The authors concludedthat “meeting high-dose chemotherapy inclusion
`criteria is an independent indicator of favorable prognosis in high-risk
`breast cancer patients and the selection of patients by these criteria may
`explain, at least in part, the promising short-term results of nonrandom-
`ized adjuvant high-dose chemotherapytrials in high-risk breast cancer.”
`With this introduction and this caveat in place, the next sections will
`review high-dose trials in metastatic and high-risk breast cancer.
`
`Metastatic Disease
`
`Trials that used high-dose chemotherapy and bone marrow orstem cell
`transplantation were derived from a search of the Medline files at the
`National Library of Medicine. Phase II studies from the past 10 years that
`
`144
`
`Curr Probl Cancer, May/June 1998
`
`

`

`included at least 15 treated patients were reviewed. Only thetrials that
`reported progression-free survival (or median response duration), overall
`median survival
`(or 2-year survival), or both, were included, since
`responserate alone giveslittle indication of the value to the patient. The
`Food and Drug Administration requires demonstration of survival advan-
`tage or, in rare cases, improvement of quality of life before a new phar-
`maceutical
`is approved for use.
`It seems appropriate to use similar
`standards when reviewing the high-dose studies.
`Of 41 trials evaluated, 16 met the abovecriteria for inclusion into Table
`1, which depicts trials of high-dose chemotherapy for metastatic disease.
`In some ofthe trials, high-risk adjuvant and metastatic patients were
`reported together and outcomes were not described for each group indi-
`vidually. In those situations, outcome measures may have beenestimates,
`or the groups may have been reported separately. All but a single trial
`were performed in the United States, where the major growth in trans-
`plantation technology has occurred. Save for one study,'® the trials were
`performed in single institutions. Although these studies were frequently
`performed on patients who may not have responded to pre-transplant
`chemotherapy, the patients are not separated, since this practice repre-
`sents the reality of overall treatment of breast cancer.
`In Table J, the last two columns show the percentage of patients who
`survived for 2 years following the transplant and the transplant-related
`mortality rate (typically, death within 100 days of the transplant). In
`many of the studies these numbers are similar, suggesting that the thera-
`peutic index of the transplant procedures was not good. Since the decade
`in which these studies were performed, however,
`there has been an
`improvement in the supportive care of patients, which may reduce the
`transplant mortality rate. Nevertheless, these statistics speak for them-
`selves, and any patient contemplating transplantation for metastatic
`breast cancer should be aware of the chance of early and treatment-relat-
`ed death.
`In the 16 trials listed, a total of 602 patients received transplants (an
`average of 38 patients pertrial), and response rates varied from 41% to
`90%. These response rates were calculated in several ways. Sometrials,
`which included standard induction chemotherapy followed by autologous
`bone and marrow transplant (ABMT), only reported patients who under-
`went ABMT in the denominator; others used an intent-to-treat analysis
`that includedall patients, regardless of their progression to ABMT.There
`was no discernible difference in the responserates of the studies of cases
`in which transplantation took place after induction chemotherapy versus
`those in which transplantation occurredinitially.
`
`Curr Probl Cancer, May/June 1998
`
`145
`
`

`

`
`
`
`
`quejdsuesIBOA-ZueIpaWasuodseyasuodsayJezOLaaneiedald
`
`
`
`
`
`
`
`uelpaw
`
`
`
`
`
`
`
`Jaques)Jsed1goesea;UlAdeJayjOWEYDasog-UBIHJo(pazIWOpPUeJUON)SIE)"TSTAWL
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`AVWEIJOWNIBAIAINS(sou)jeajains(sow)uoeinga7eysjueneduawisoyuolny}sUl
`
`
`
`KET%6EOT(SY9)6MLLZeANog/dado/AD94nd
`
`%L%TSOz<(Sud)6T%G86@O/HL/ADorfaqieseueg0*6<=LT9/A0
`
`
`
`
`gslBUOWA|0“BTaAeT%08veH1/ADgSUlydo}
`
`%OE%676Zt%LPEzANOE/AD2yoseaIuD0-ST<é%060zW¥d/89/HL/A9
`
`qeqieyeueg
`
`
`
`
`
`%PT%8ESTS%OLzzHL/AD210829149
`
`
`
`%L%STA.%EEvy/9O/HL/ADerusny
`
`
`
`%T>%97Z::egSO/HL/ADprAJowg
`
`
`
`%e-6TL“BtLE13/99/49groidual
`
`13a/na
`
`
`
`
`
`%e%T9Te:%69vIt99/HL/ADgr9suodsay
`
`
`
`
`
`%8%ES>0t~=%GS6TN/HL/AD;quOSUIY9INHPedy
`
`
`
`%ST%OSeT6<%S8LzHL/ADgrOSe9IU9
`
`
`
`%ET=6z-9%AH/HL/ADgyP4SCIGON
`
`
`
`%8%SEST:%O8os13/d009/A9ozSINOT“IS
`
`
`
`%OZzecD6%OLSTVdr2vepsiewjeAoy
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`-odoye‘79‘ueljnsng‘ng‘pueisnwaujueje|Auayd‘yg‘une|doqes‘gg‘edajon)‘Hy‘siapuodsalayajdwioo‘sy9‘aulsnuues‘ANog‘uneldsis‘dado‘apmueydsaydajadg“49
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`“BaunkxoupAy‘AH‘apis
`
`146
`
`Curr Probl Cancer, May/June 1998
`
`

`

`In the largest study which used only stage IV, chemotherapy-naive
`patients, the response rate was mid-range (69%) and the median survival
`was the longest reported (31 months). Only 23 patients in that study
`(20%) had received adjuvant chemotherapy, so most represented a popu-
`lation without the negative prognostic factor of recurrence following
`adjuvant therapy. Among patients with measurable disease, there was a
`response rate of 58% to induction with doxorubicin, methotrexate and
`fluorouracil, but this was notstatistically significantly different from the
`50% response rate in patients in which adjuvant therapy failed. Ninety-
`three patients (82%) received ABMT, while 21 did not, for reasons
`including treatment-related death, inadequate marrow collection, interval
`progression of disease, patient refusal, and major protocol violation. The
`final response rate of 69% after ABMTgenerated an additional response
`rate of only 11%, suggesting that the high-dose therapy was marginally
`effective in the induction of further antitumor activity. Only 6 patients
`were converted to a better response category by high-dose chemotherapy,
`and long-term disease control was rare, with only 18% of the patients
`remaining progression-free at this time.
`To place ABMT into perspective, it is necessary to evaluate the out-
`comes of the manypatients treated with standard-dose therapy. It is also
`important to reiterate the fact that end results in most of the trials of
`ABMT were based on denominators that included only patients who
`eventually received transplants rather than on the entire cohort. Only the
`studies from Hopkins* and Response Technology" reported endresults as
`intent-to-treat, which is the more conventional way to report results and
`has been used in standard-dose studies overthe years.
`Table 2 summarizes the outcomes of standard-dose chemotherapy in
`metastatic breast cancer studies that have been published in the last 10
`years and before the era of taxanes, which have only recently been incor-
`porated into ABMT preparative regimens. Again, 16 trials were identified
`in which response and survival rates were reported, allowing for compar-
`ison, in a nonrandomized way, with those trials that employed high-dose
`therapy. One study*evaluated two different doses of epirubicin, neither
`requiring stem cell support. The overall response rates in the studies of
`standard-dose therapy are lower than those of the high-dose studies, but
`the median survival rates are quite similar. Even in the standard-dose
`studies performed in single institutions, the response rates and median
`survivals are marginally better than in the large, cooperative group, multi-
`institutional trials. With the exception of the Response Technology study,
`all high-dose studies were performedin single institutions where there
`wasa potential for greater selection bias. To further emphasizethe effect
`
`Curr Probl Cancer, May/June 1998
`
`147
`
`

`

`
`
`
`
`-G/uisiqnioxop/apnueydsoydojoAo“yD
`
`
`
`
`
`/auoqUEXOYW/A}eXajOYyeW“WAI*|}OeINOJONY-G
`
`96S
`
`vT
`
`QTSACT
`
`ve
`
`60SABOT
`
`OESATVE
`
`VETSAGET
`
`TSTSAGET
`
`BTSAZST
`
`ZSTSAGOT
`
`ETSAELT
`
`%S9
`
`vO
`
`*TGSA%09
`
`YO%ST
`
`HELSA%OD
`
`%S9SA%TL
`
`%EBCSA%TS
`
`MBESAMCE
`
`WESSAHEP
`
`BPSAKOE
`
`%SESA%TE
`
`LTTSAP'SCL
`
`VESSATLE
`
`O'STSALLT
`
`
`
`TUTESASTT
`
`LESAHES
`
`TtSA%69
`
`%OSSA%CS
`
`LYSAKOE
`
`LTSAJTSAGT
`%TGSA%LGSA%OS
`
`ad
`
`BL
`
`v6
`
`90T
`
`vst
`
`9EE
`
`6re
`
`80c
`
`160
`
`VSe
`
`Tet
`
`OES
`
`Srl
`
`}o.NSEqINIS|
`
`UB/paW
`(sow)jeAlAINS
`
`
`ajyeyesuodsey
`[210]
`
`uSsWIsaYyJUSUIZEOIL
`Apnygjoeset
`
` squsped
`
`uoWSU]
`
`pevT
`
`ZEL
`
`Scl
`
`dago/xod/a1A/XLW
`
`WINSASIND
`
`aldninw
`
`4V0SA4IWO
`
`4dVOSAAWO
`
`SadF4WO
`
`WIND
`
`HAVLSAHAWG
`
`d4dINOSAdso
`
`dvossJNO
`
`dJINDSAL
`
`dAJINO/HLWASAHLWASAAVO
`
`JVSAINO
`
`93454034GH
`
`a9SA4VO
`
`(p)oaaSA(T)934
`
`
`
`
`
`egAVsieauneueIpu}
`
`zzuosiepuyGIN
`
`
`
`pelapsueWjeéoy
`
`
`
`gz/eqe4eueg
`
`a¢zDO03
`
`oz)OO0F
`
`1t2DLDON
`
`ezD004
`
`6zDLOON
`
`oeDION
`
`
`
`pePuejeaZMaN/elleajsny
`
`eefOqUBOINW
`
`peslessmig
`
`zeB8DIWO
`
`ee4Oualy
`
`gePURUl
`
`
`
`jaoueyysed1goIJeISeIEWUlAderayjoWaYDesog-psepuelsJoSjell|"%FTAVL
`
`
`
`
`
`‘994OH‘auosiupaid/aunsuoun/}oeunoJONY-G/ayexoJ]OYOUipueJa0ueD
`
`
`
`fapiueydsoydojoAo‘d4wi9‘auostupaid/j!9e!
`
`
`
`
`
`/ujoiqnioxop/|oYo|NpowelgIp“HAT
`
`‘gO7¥9‘uayxowey4y
`
`
`
`*|}oeunosony/ulsiqnajde/apiweydsoydojoAo‘499
`
`‘dnosyABojoouQeanesadoogwseysey“DO0F‘ulgIgnJoxop/OeuNOuONYy-G/ayeXe}OueU
`
`
`
`
`
`
`
`
`1oBsnNOJONY-G/auosjuexOWW
`
`
`
`
`
`
`/ayexasoujow/apimeydsoydojaXo‘wo:uneldsio‘dddd‘uIDIqnJOXOP‘NOC‘AUlISBQUIA‘GTA‘@}BXaNOUIN“XL
`
`
`Noony-G/apiweydsoydojoAo‘dD‘QUDIAISOWAXONY,/BUIISIOUIA/Ul
`
`
`
`
`
`
`Aujaip‘sag‘dnosyuawyeas,JA9URDBIUJOJIEDWEYVON‘DLOIN
`
`
`
`
`/apiweydsoydo|ako‘39‘epeuedJOany}sulJe9UeD|BUONEN‘SION‘aU
`
`
`
`
`
`‘apiueydsoudojoAo/ujaiqnaida/|!9esnos0nYG“OFF‘apnueydsoydooAo/uoiqniide/|I9eINOJONEGasop-ysiy
`
`apiueydsoydold4o‘dAJID
`
`
`
`
`
`
`‘guosaysaluAxony/edayoiyy/Ula|qnJoxop/AUNISUOUIA"HLAtgdnoudelwayns)anoy
`
`
`‘aUSUdU|A/ULIGNJOXOp/JOUSINPOWOgIP‘AY-|/9esnosoNy
`
`/apiueydsoydojoAo“pyWO-DulAWOJIWW
`
`
`
`os|upaid/jloeunouony-g/eyexas}OYJaW
`
`
`
`
`
`
`aignioxop/edayoiy)"HAVL
`
`
`
`‘guQJaSaWAKONY/BUNSHOUIA
`
`148
`
`Curr Probl Cancer, May/June 1998
`
`

`

`that selection bias may have, one need only examinethe results of a large
`Eastern Cooperative Onocology Group study in which 10-year follow-up
`is available. In thattrial,*’ the overall response rate was 68%, the com-
`plete response rate was 28%, and the median survival was 17 months for
`the entire cohort of patients. In the group of estrogen receptor-positive
`patients treated with cyclophosphamide/doxorubicin/fluorouracil (CAF)
`and receiving oophorectomies, however, the response rate (73%) and the
`complete response rate (20%) were similar, but the median survival was
`59 months. If one eliminated those patients with bony or hepatic metasta-
`tic disease (the former difficult to assess for response and the latter usu-
`ally a negative outcome prognosticator), the response rate rose to 91%
`and the complete response rate to 51%. Thus patient selection may sig-
`nificantly affect outcomes, in both the response rate and the median sur-
`vival time. In the often quoted high-dose study from Duke University,’ all
`the patients were premenopausal (age < 50 years) and few had bone or
`liver involvements. As a result, there was an impressive 77% response
`rate and a complete response rate of 54%, but the median survival was a
`disappointing 10 months.
`One can attempt to compare nonrandomized studies with randomized
`ones by weighting the response rates and the median survivals according
`to the number of patient entries so that larger studies have a greater
`impact on outcome measures. When this approach is used regarding the
`studies outlined in Tables | and 2, there is a weighted responserate of
`68% and a median survival of 18.5 months in the high-dose studies, while
`in the standard-dosetrials the response rate is 53% (significantly lower
`statistically), but the median survival is 18.2 months. Thusit is difficult
`to conclude from this semi-mathematical approach that high-dose thera-
`py with transplantation offers much except an improved responserate.
`Although important, only if this can improve median survival or cure
`rates will it have any appeal to health care practitioners or economists.
`In a recently reported randomizedtrial,*® patients with hormone-insen-
`sitive and chemotherapy-naive metastatic breast cancer were initially
`induced into a complete remission with doxorubicin-based chemothera-
`py. Ninety-six of the 423 patients entered a complete remission and were
`then randomly putinto groupsfor transplantation that would occureither
`immediately or at the time of relapse. Although the disease-free survival
`was superior in the group that received transplantation immediately (0.9
`months versus 0.3 months), the overall survival was superior whentrans-
`plantation was delayed until relapse (1.9 years versus 3.2 years). Of
`course,
`this randomized trial begs the question of whether high-dose
`chemotherapyinitially or at relapse is superior to more standard doses of
`
`Curr Probl Cancer, May/June 1998
`
`149
`
`

`

`therapyat relapse, butthe details of this study are not yet published. One
`must wonderif the initial high-dose therapy group which relapsed had
`such poor survival because of recurrence with a highly drug-resistant
`clone of cells or because of fatal long-term complications from the trans-
`plant. The message that this study should send to physicians conducting
`trials is that early analysis of the indicatorsof efficacy—tresponserate and
`progression-free survival—maynottranslate into improvementin either
`quantity or quality of life, which are the ultimate requirements for any
`new therapy to be considered as a standard ofcare.
`There has been a single randomizedtrial of standard versus high-dose
`therapy in metastatic breast cancer.*’ In thattrial, 90 patients were ran-
`domly assigned to either 6 to 8 cycles of standard doses of mitoxantrone,
`vincristine, and cyclophosphamide, or two courses of high-dose mitox-
`antrone (40 mg/m’), cyclophosphamide (2400 mg/m?), and etoposide
`(2500 mg/m?) x 2 with stem cell support. Responsesrates (53% vs 96%),
`complete response rates (4% vs 51%), and median survival (45 weeks vs
`90 weeks) were superior for the high-dose arm. This study, however,
`failed to resolve the debate regarding high-dose therapy in breast cancer.
`First, the trial was small andstatistically underpowered. Small changesin
`the recurrence rate—inthis case, if the disease recurred in as few as 1 or
`2 patients—could have rendered the results statistically insignificant.
`Second, the follow-up in this study was relatively short and long-term
`survival has not been assessed. Third, the response rate of 53% and the
`complete response rate of 4% in the conventional arm were lower than
`would be expected. The studies shown in Table 2 had apparently similar
`patient populations, but outcomesthis poor were not reported. Fourth, the
`patients were generally younger than an unselected population and may
`have had other prognostic factors that make comparison with the larger
`unselected group of patients in Table 2 untenable. Finally, there is not a
`significant difference between median progression-free survival and
`median overall survival in the patients who received transplants. One
`must consider that the transplantation may have changed some of the
`already drug-resistant cellular clones to a biologically more aggressive
`state. These results must be cautiously interpreted and the outcomes of
`larger randomized trials must be reported before the adoption of this
`high-dose approach. Ongoing studies that may help answerthis question
`include the study mentioned above,*® which is asking when it is best to
`transplantratherthanif it is wise to transplantat all. A second study from
`Philadelphia, which recently closed, gave patients with measurable
`metastatic disease standard doses of CAF followed by standard
`cyclophosphamide/methotrexate/fluorouracil
`(CMF) or high-dose
`
`150
`
`Curr Probl Cancer, May/June 1998
`
`

`

`therapy with carboplatin, cyclophosphamide,and thiotepa. Results of this
`study are anxiously awaited. The National CancerInstitute of Canada is
`also mounting a trial in which patients are initially treated with an epiru-
`bicin-based combination and then those patients with re