ACS SYMPOSIUM SERIES 583
`
`Taxane Anticancer Agents
`Basic Science and Current Status
`
`Gunda I. Georg, EDITOR
`University of Kansas
`
`Thomas T. Chen, EDITOR
`University of Tennessee
`
`Iwao Ojima, EDITOR
`State University of New York at Stony Brook
`
`Dolatrai M. Vyas, EDITOR
`Bristol-Myers Squibb PRI
`
`Developed from symposia sponsored
`by the Divisions of Chemical Health and Safety,
`Medicinal Chemistry, and Organic Chemistry
`at the 207th National Meeting
`of the American Chemical Society,
`San Diego, California,
`March 13-17, 1994
`
`American Chemical Society, Washington, DC 1995
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`AVENTIS EXHIBIT 2024
`Mylan v. Aventis
`IPR2016-00627
`
`

`
`Chapter 3
`
`Current Status of Clinical Trials
`with Paclitaxel and Docetaxel
`
`F. A. Holmes1,4, A. P. Kudelka1, J. J. Kavanagh1, M. H. Huber2,
`J. A. Ajani3, and V. Valero1
`
`Departments of 1Breast and Gynecologic Medical Oncology,
`2Thoracic/Head and Neck Medical Oncology,
`and 3Gastrointestinal Medical Oncology,
`The University of Texas M. D. Anderson Cancer Center,
`Houston, TX 77030
`
`The unique mechanism of action of the antimicrotubule agent paclitaxel
`suggested that it would be a potent antineoplastic agent. However, even
`after multiple preclinical problems with paclitaxel were surmounted,
`a number of unique clinical problems still required resolution. Despite
`the existence of over 50 active antineoplastic agents, drug resistance
`and patient tolerance limit the number of effective agents in specific
`tumor types. Paclitaxel has shown antitumor activity in multiple clinical
`trials in cancers of the ovary, breast, head and neck, lung, and
`gastrointestinal tract. In many of these trials, paclitaxel was active despite
`evidence of the tumors' resistance to other important drugs. Paclitaxel
`may be the first of a series or family of drugs: in preliminary trials,
`an analogue, docetaxel, has also shown significant antineoplastic activity.
`However, many issues regarding the optimal use of both of these drugs
`remain unresolved.
`
`Overview: Paclitaxel and Chemotherapy
`
`Schiff and Horowitz's description of paclitaxel's unique mechanism of action was the
`catalyst for paclitaxel's clinical development (1). The difficulties in translating paclitaxel
`from the forest to the pharmacy were described in chapters 1 and 2. Additional difficul(cid:173)
`ties awaited clinicians who began using paclitaxel in clinical trials. This chapter will
`address those clinical problems and the results of therapeutic trials in patients with
`tumors of the ovaries, breast, head and neck, lungs, or gastrointestinal tract. The trials
`in ovarian and breast cancers confirmed that paclitaxel has major clinical activity,
`leading to its approval by the Food and Drug Administration (FDA) for commercial
`use. The clinical results of trials of a semisynthetic taxane (taxoid), docetaxel (Taxotere),
`will be also discussed here.
`
`4Current address: Department of Breast and Gynecologic Medical Oncology, Box 56,
`The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard,
`Houston, TX 77030-4009
`
`0097-6156/95/0583-0031$09.26/0
`© 1995 American Chemical Society
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`32
`
`TAXANE ANTICANCER AGENTS
`
`Clinical Problems Unique to Paclitaxel. The early clinical trials of paclitaxel faced
`three specific problems.
`
`Hypersensitivity Reactions. The purpose of the first series of trials of a new
`drug in humans (phase I trials) is to determine the maximum tolerated dose by treating
`consecutive cohorts of patients with escalating doses of the drug. The maximum tolerated
`dose is defined by the occurrence of toxic effects, called dose-limiting toxicity, whose
`severity or permanence limit further dose escalation.
`In addition to the expected
`problems of myelosuppression (low white blood cell counts) and neuropathy encountered
`in the initial phase I trials of paclitaxel, an acute allergic reaction occurred that was
`fatal in one patient (2). This reaction was similar to the severe reactions experienced
`by some patients who receive iodinated intravenous contrast medium for radiographic
`procedures. The unpredictability of this reaction terminated clinical trials until it was
`discovered that it was rapid infusion of the diluent, Cremophor EL, that caused the
`reaction. To prevent this, the National Cancer Institute (NCI) recommended infusing
`paclitaxel over 24 hours and premedication with corticosteroids and antihistamines
`(3). These strategies have reduced the incidence of serious hypersensitivity reactions
`to 1% or less. Since both the premedication regimen and the slow infusion duration
`were developed simultaneously, it was initially unclear which was more important.
`Further studies have shown that each is effective independently.
`
`Cardiac Toxic Effects. When clinical trials were resumed, all patients were
`treated in an intensive care unit with cardiac monitoring. Nearly 30% of patients were
`observed to have an abnormal but generally benign slowing of the cardiac rhythm
`(sinus bradycardia) while receiving paclitaxel. In a few cases, however, this rhythm
`was so slow that a pacemaker was required to continue treatment. A few patients
`with severe but undiagnosed coronary artery occlusions died of myocardial infarctions
`(heart attacks) or had life-threatening rhythm abnormalities (4). Cardiac monitoring
`was required in all clinical trials and extensive data were collected. Analysis of 3400
`patients revealed that the incidence of life-threatening events was less than 0.5%.
`Review of trials conducted before these heart problems were observed and of historical
`data from other drug development studies revealed that multiple benign rhythm
`abnormalities are common in patients receiving chemotherapy. It was recommended
`that patients who had known disease of or took drugs affecting the conduction system
`be given paclitaxel only with cardiac monitoring.
`
`Drug Supply. After phase I trials have determined an effective and safe dose,
`most active new drugs are tested simultaneously in several trials in specific tumor
`types (phase II) and at different research centers, with exploration of different infusion
`durations and retreatment intervals (administration schedules). The early scarcity of
`paclitaxel meant that the results of each trial had to be carefully evaluated before planning
`the subsequent trial, that the numbers of patients treated would be limited, that the
`duration of treatment would be curtailed, that only limited types of tumors would be
`tested, and that the time frame for the start of these trials was delayed. To determine
`the level of activity with narrow confidence levels, the usual numbers of patients treated
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`3. H O L M ES ET A L. Clinical Trials with Paclitaxel and Docetaxel 33
`
`in a trial of a drug with such high activity would be 35-50 patients. In the early trials
`in breast cancer, for example, only 25 patients could be treated. Similarly, in a drug
`with such a high level of activity, most patients would be given at least six treatments
`before their tumor was judged unresponsive. In the lung cancer studies, patients whose
`tumors did not evince a 50% or greater shrinkage in the perpendicular diameters of
`bidimensionally measurable lesions (i.e., an objective response), were removed from
`study even though clinical information, such as shortness of breath or pain control,
`suggested they were having clinical benefit. The initial phase II trials were limited
`to only three tumor types; renal (kidney), melanoma, and ovarian. Of these, only ovarian
`cancer is fairly common. Finally, although trials in breast cancer were planned in
`1985, sufficient supplies of the drug did not become available until 1990. As noted
`above, improved extraction methods and formulation of a semisynthetic drug have
`alleviated the supply problem.
`
`Clinical Problems of Chemotherapy. Inherent limitations in the current practice of
`clinical oncology make paclitaxel a needed addition to the therapeutic armamentarium.
`
`Drug Resistance. Although there are nearly 50 different antineoplastic drugs
`in use, only a dozen or fewer are effective in the treatment of each specific tumor
`type because of intrinsic or primary resistance. The initial and subsequent regimens
`of chemotherapy allow development of secondary or acquired resistance by selecting
`cells that survive. Ultimately, by a variety of mechanisms, a multiply drug-resistant
`tumor evolves, and further chemotherapy induces only toxic effects without tumor
`kill (5). Tumors are classed by their degree of chemosensitivity. Breast and ovarian
`cancers are moderately sensitive. However, when the tumor becomes resistant to
`doxorubicin (breast) or cisplatin (ovarian), few other drugs are effective.
`
`Patient Tolerance. As the tumor grows, the patient becomes increasingly debilitat(cid:173)
`ed by the accumulation of secondary effects from the tumor (cachexia, pain) or previous
`treatments (bone marrow or heart muscle failure from irradiation or prior chemotherapy
`or both) as well as primary effects, which depend on the site of involvement (shortness
`of breath, liver failure, bone fractures). Thus, with the exception of those with very
`indolent tumors, most patients have the physical reserve to endure only a limited number
`of aggressive chemotherapy regimens.
`
`Ovarian Cancer
`
`leading cause of death among
`Background.
`the
`is
`Ovarian cancer
`gynecologic malignancies in the USA, surpassing the mortality from cervical and
`endometrial cancer combined. Approximately one woman in 70 will develop ovarian
`cancer. In American women it is the sixth most common cancer and the fourth most
`common cause of death (6). In 1994, 24,000 cases and 13,600 deaths are attributed
`to ovarian cancer (7). The peak incidence is in the seventh decade; it is uncommon
`below 50 years of age (8). The incidence is high in North America and Northern Europe,
`and low in Japan (9).
`The cause of ovarian cancer is unknown, but it is associated with consumption
`of animal fat, and is more common in patients with a history of breast cancer (6).
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`34
`
`TAXANE ANTICANCER AGENTS
`
`Only about 5% of cases are hereditary. Childbearing and use of birth control pills
`reduce the risk of developing ovarian cancer by 30-60%, but use of replacement estrogen
`has no effect on the incidence (10).
`The two most important features of the disease which determine outcome are
`the extent of the disease (stage) and the aggressiveness of the tumor as determined
`by microscopic evaluation (histologic grade). Unfortunately, 75-85% of patients are
`diagnosed with advanced disease which has metastasized from the ovaries in the pelvic
`cavity to the abdominal cavity, because symptoms are often absent until the disease
`involves other organs in the abdominal cavity. These patients are rarely cured (11-14),
`but treatment with chemotherapy may reduce symptoms and prolong life.
`Standard therapy consists of cyclophosphamide with either cisplatin or its newer
`analogue, carboplatin. This causes tumor regression in 60-80% of patients of which
`30-50% are complete responses (CR). The median duration of survival is 18 to 24
`months, and 5- and 10-year survival for ovarian cancer metastatic to the abdominal
`cavity and elsewhere is 5-20%, and 0-10%, respectively. Carboplatin has less neurologic,
`kidney, and auditory toxic effects than cisplatin, causes less nausea and vomiting, and
`provides a better quality of life than cisplatin. However, it is more expensive and
`causes more depression of the white blood cell and platelet counts (myelosuppression
`and thrombocytopenia) (15). Hormonal therapies are transiently effective in 10-20%
`of patients.
`Patients whose tumors previously responded to chemotherapy and who have
`a treatment-free interval of at least six months are defined as potentially "platinum-"
`or Hplatin-sensitive,M as 3 0-50% of these patients will have tumor regression if retreated
`with either carboplatin or cisplatin (16,17). Patients with tumors that are platin-resistant,
`defined as worsening disease during treatment, persistent disease after four to six
`treatments, or recurrent disease within 6 months after completing therapy (18,19),
`have a median survival of 12 months or less, and no currently available drugs have
`been shown to prolong these patients' life span. It is in this group of patients that
`new drug treatments are urgently needed and in whom they are first tested.
`
`Paclitaxel Trials in Ovarian Cancer
`
`The initial clinical trials of paclitaxel in humans treated patients with multiple tumor
`types that had failed all standard therapies. The intent of these trials was to determine
`the safest and most effective dose (maximum tolerated dose, MTD) and infusion duration
`(schedule). Evidence of tumor regression is uncommon in such trials. However,
`unexpectedly, tumor regression was seen in patients with platin-resistant ovarian cancer
`(20).
`
`Single-agent trials. These above results were the basis for a series of trials
`in patients with ovarian cancer designed to evaluate antineoplastic activity (phase II
`trials) of paclitaxel. A total of 111 patients were treated with doses of 100-250 mg/m2
`infused over 24 hours every three weeks (21-24). Overall, 20-37% of patients had tumor
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`3. H O L M ES ET AL. Clinical Trials with Paclitaxel and Docetaxel 35
`
`regression, and in seven patients it was complete. Analysis by platin-sensitivity revealed
`responses in 40-50% of patients with potentially platin-sensitive rumors, and in 24-30%
`of patients with platin-resistant tumors, with at least two of the patients achieving a
`CR. The median duration of response was 6 months (range, 2-30). The overall median
`survival was 11 months. It was 17 months for patients with potentially platin-sensitive
`tumors and 9 months for those with platin-resistant tumors (24). The major toxic effect
`in these studies was granulocytopenia (lowered granulocyte count, a subset of the
`white blood cells responsible for preventing bacterial infection) which necessitated
`dose-reduction to prevent infection. These studies showed the potential range of
`antineoplastic activity but did not define whether the response rate was a function of
`dose.
`
`High-Dose Trials. To determine whether antineoplastic activity was higher
`at higher doses, the National Cancer Institute (NCI) and M. D. Anderson each tested
`paclitaxel in a single-arm study at 250 mg/m2 over 24 hours in patients with platin-
`resistant ovarian cancer. Granulocyte-colony stimulating factor (G-CSF), abiosynthetic
`form of the endogenous chemical that stimulates production and maturation of
`granulocytes, was given to prevent severe granulocytopenia (25-28). More than 50%
`tumor regression (a partial response) was seen in 48% of patients in each study. The
`duration of response was 6 months, and the median survival was 12 months.
`
`Trials in Patients with Multiple Prior Therapies. Although paclitaxel was still
`not approved by the FDA and thus not widely available, this striking evidence of
`antineoplastic activity suggested that many patients could potentially benefit from
`paclitaxel. Therefore, the NCI supplied paclitaxel to approved cancer centers for
`compassionate use (Treatment Referral Center, TRC, mechanism) for patients whose
`tumor had progressed despite three or more prior chemotherapy regimens. The dose
`was lower, 135 mg/m2 over 24 hours, because of the extensive previous treatment.
`The overall response rate was 22%; the median survival was 9 months (29).
`
`Dose- and Schedule-Comparison Trial. In a European-Canadian (NCI-Canada)
`study of paclitaxel in patients who failed one or two platin regimens, 60% had platin-
`resistant disease, and all had measurable or évaluable disease (30). The study was
`performed to answer two questions. First, is the response rate dependent on dose?
`Two dose levels were tested, 175 mg/m2 and 135 mg/m2. Second, is the 3-hour infusion
`schedule as safe and effective as the standard, but more cumbersome, 24-hour infusion?
`Thus, there were four treatment arms, because the high- and the low-dose arms were
`each given by 3- or 24-hour infusion. Responses were more frequent with the larger
`dose (20% versus 15%, respectively) and with the longer infusion schedule (19% versus
`16%, respectively). However; neither of these differences achieved statistical significance.
`The response rate for platin-resistant patients was 13%, but the survival was not reported
`for this subgroup (30).
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`36
`
`TAXANE ANTICANCER AGENTS
`
`TABLEL
`
`Studies of Paclitaxel in Advanced and Refractory Ovarian Cancer
`
`Institution
`(Reference)
`
`No. of
`Patients
`
`Dose, mg/m2 Overall CR % Median
`Response (No.) Survival,
`mo.
`
`Single Agent
`JHOC (21)
`40
`GOG (23)
`41
`Einstein (22)
`30
`NCI-TRC (29)
`619
`European-
`195
`Canadian (30)
`187
`High Dose
`(With G-CSF^
`NCI (24,25)
`44
`MD Anderson
`48
`(27, 28)
`
`135 (110-170)
`170 ( I)
`180-250
`135
`135
`175
`
`250
`250
`
`30%
`37%
`20%
`22%
`15%
`20%
`
`48%
`48%
`
`2.5(1)
`12(5)
`3(1)
`3
`1(2)
`2(4)
`
`14
`4
`
`8.2
`15.9
`6.5
`9
`11.0
`11.5
`
`11.5
`12
`
`Conclusions. Single-agent paclitaxel infused over 24 hours produces antineoplastic
`responses of 10-22% and 48-50%, respectively, for doses of 13 5 mg/m2 and 250 mg/m2,
`suggesting a benefit to higher doses. However, two additional points are necessary
`to interpret these data correctly. First, only a concurrent, randomized trial comparing
`the 135 mg/m2 and 250 mg/m2 doses can prove this hypothesis. Second, even though
`more patients responded at the higher dose, the median survival was comparable in
`both trials. The apparent lack of benefit in terms of increased survival for the high-
`dose group, with its attendant increase in incidence of toxic effects and cost (owing
`to both the paclitaxel and the G-CSF), suggests to us that the higher doses may be
`most useful in alleviating the severe cancer-induced symptoms of some patients with
`advanced platin-resistant ovarian carcinoma. The dose and schedule approved by the
`FDA is 135 mg/m2 over 24 hours given every 21 days.
`
`Combination Trials. The next obvious step to optimize paclitaxel's antineoplastic activity
`was to combine it with platin and to treat patients who had not previously received
`chemotherapy. A trial comparing the standard combination of cisplatin with
`cyclophosphamide to the same dose of cisplatin with paclitaxel 135 mg/m2 was initiated.
`Only preliminary data has been reported (24). Patients on the paclitaxel experienced
`significantly more numbness, hair loss, allergic reactions, and episodes of fever during
`periods of granulocytopenia. The overall response rate was 64% for the standard arm
`and 77% for the paclitaxel-containing arm, a statistically significant difference with
`ρ = 0.02 (24). Still unreported is information on survival or comparison of the patients'
`quality of life or cost of treatment. For all of these reasons, the use of paclitaxel
`combinations in previously untreated ovarian cancer patients should be considered
`investigational and not a standard of care. A number of other combination trials with
`other agents are ongoing, but the data are too preliminary for conclusions.
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`3. H O L M ES ET A L. Clinical Trials with Paclitaxel and Docetaxel
`
`37
`
`Breast Cancer
`
`In 1994, 182,000 patients will be diagnosed with breast cancer; 1%
`Background.
`of these will be men. Of the 46,000 patients who will die of this disease, most will
`have been diagnosed previously (7). Breast cancer is the most common malignancy
`affecting women. For unexplained reasons, the incidence of breast cancer has stabilized
`recently. Exposure to pesticides at a young age has recently been implicated in breast
`cancer because of the estrogen-like effect of DDT (31). The current trend for delayed
`or deferred childbearing may cause the incidence of breast cancer to increase (32),
`but even if the rate of breast cancer does not increase, the absolute numbers of patients
`will rise with the "graying" of America.
`The most important variable determining a patient's outcome is the extent of
`disease, or stage (33). Tumors that are confined to the locoregional area comprising
`the breast and draining lymph nodes in the axilla (armpit), stages I - III, are potentially
`curable with local therapy (total mastectomy or lumpectomy and axillary lymph node
`dissection followed by irradiation of the remaining breast tissue) with or without systemic
`therapy, that is, chemotherapy or hormones or both (adjuvant or prophylactic therapy
`for micrometastases). Tumors that have metastasized beyond the locoregional area
`(stage IV) are incurable. However, even in patients with incurable tumors, palliative
`therapy with chemotherapy, hormones, and/or radiation may prolong good quality of
`life. Before the use of chemotherapy, the average life expectancy after the diagnosis
`of stage IV (metastatic) breast cancer was 9 months (34). With chemotherapy, the
`median life expectancy is 2 to 3 years. The first chemotherapy regimen is generally
`effective at reducing tumor size by at least half in 50-70% of patients, and this effect
`lasts for a median of 9 to 12 months. Subsequent treatments are effective in about
`25-40% of patients for 3 to 6 months. The most effective antineoplastic drug available
`for breast cancer is doxorubicin. Tumors that develop resistance to doxorubicin usually
`have only very brief responses to other agents. It is in these patients, who have stage
`IV disease and who have received a variety of previous treatments, that new chemotherapy
`agents are first tested. With the increased awareness of the multiple mechanisms of
`drug resistance, however, many new drugs are tested in patients who have received
`only one or fewer prior therapies in order to prevent false-negative results from tumors
`that are resistant to all drugs.
`In contrast to some other tumors, in breast cancer the histologic type, with few
`exceptions, has little impact on outcome. The degree of differentiation of the tumor,
`or grade, as determined by a variety of methods, is more important. Poorly differentiated,
`or anaplastic, tumors grow rapidly and develop early resistance to therapy. In stage
`IV disease, the extent and sites of metastases correlate with outcome. Patients with
`metastases to visceral organs generally have a more rapidly fatal outcome. A unique
`feature of breast cancer is its responsiveness to hormonal treatments in a subset of
`patients. Eventually, however, all hormone-sensitive tumors become insensitive.
`
`Paclitaxel Trials in Breast Cancer Patients. The initial investigative trials tested paclitaxel
`alone in patients who had received various numbers of prior chemotherapy regimens.
`Later trials investigated paclitaxel in combination with other standard antineoplastic
`drugs.
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`38
`
`TAXANE ANTICANCER AGENTS
`
`Single-Agent Trials in Patients with Limited Prior Chemotherapy. The initial
`trial of the antineoplastic activity of paclitaxel in breast cancer was at M. D. Anderson
`in patients with stage IV breast cancer who had received one prior regimen of
`chemotherapy (35). Paclitaxel was given at a dose of 250 mg/m2 over 24 hours to
`25 patients of whom 66% had visceral disease. In 56% of these patients, tumors shrank
`50% or more, and this objective response lasted for a median of 9 months (range, 5-27).
`In three patients, tumor regressed completely. The treatments were generally well
`tolerated; the median number of treatments per patient was 13 courses (range, 2-21).
`The toxic effects were those expected: low blood counts in 88% of patients (but
`complications of that in only 6% of the 297 courses), total hair loss, and muscle aches
`which were severe in only a minority. Long-term effects included a sensory neuropathy
`manifested primarily by impairment of fine motor functions such as buttoning, sewing,
`fastening jewelry, etc. However, no patient experienced an allergic reaction. This
`unexpected result was confirmed by review of all x-rays and patient charts at the NCI.
`A trial at the Memorial Sloan-Kettering Cancer Center in New York confirmed
`these results in a similar group of patients except that only 16 had received prior
`chemotherapy (36). The dose and duration of infusion were similar, except that G-CSF
`was administered to prevent the severe drop in white blood cell counts, with the hope
`of decreasing the incidence of infection. Of 26 patients, 62% had a decrease in tumor
`size of 50% or more. The use of G-CSF reduced the incidence of infectious
`complications by half.
`Finally, when more paclitaxel became available, a large trial commenced in
`Europe and Canada (Canadian-European Taxol Study Group, CETSG) which focused
`on women who had received only one prior chemotherapy treatment for metastatic
`disease regardless of whether they had received preventive therapy (adjuvant) after
`mastectomy (37). The objectives of the trial were two: to determine the efficacy in
`breast cancer of the 3-hour schedule, which had been developed for ovarian cancer,
`and to compare doses of 175 mg/m2 and 135 mg/m2. A total of 471 women were
`treated. Nearly a third had received adjuvant treatment, another third had received
`treatment for metastatic disease, and the remaining third had received both adjuvant
`and metastatic treatment. The response rates, 29% for the high dose and 22% for the
`low dose, were not significantly different. Although these response rates were not
`as high as in the previous two trials, the doses used were lower and the infusion schedule
`was shorter. Both of these changes diminished the effective given dose. One measure
`of this is the determination of the incidence of granulocyte counts below 500 cells/mm3
`(grade 4 granulocytopenia). In the original M. D. Anderson study, 100% of patients
`(88% of courses) experienced grade 4 granulocytopenia In the Memorial Sloan-Kettering
`study, which used G-CSF to prevent or diminish this effect, 66% of courses had grade
`4 granulocytopenia. In the European- Canadian study, only 27% and 21% of courses,
`respectively, had grade 4 granulocytopenia. The data from this trial were the basis
`for approval of paclitaxel as "salvage" therapy for metastatic breast cancer which
`worsened despite doxorubicin treatment in December 1993. The approved dose and
`schedule is 175 mg/m2 by 3-hour infusion given every 21 days.
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`3. H O L M ES ET A L. Clinical Trials with Paclitaxel and Docetaxel
`
`39
`
`Single-Agent Trials in Patients with Multiple Prior Therapies. Both M. D.
`Anderson and Memorial Sloan-Kettering tested the 24-hour infusion schedule in patients
`who had received two or more prior treatment regimens. The Memorial Sloan-Kettering
`trial used G-CSF which allowed treatment at higher doses. The planned dose level
`in the Memorial Sloan-Kettering study was 200 mg/m2(38). In patients who had received
`two or three or more prior regimens, respectively, response rates were 31% and 20%.
`At M. D. Anderson, doses of 175 mg/m2 and 150 mg/m2, respectively, without G-CSF,
`were used for patients with two or three or more prior regimens (39). Responses were
`seen in 23% and 20% of patients, respectively. Approximately 35 patients were treated
`in each of the four trials. Nearly 15% of patients on each trial had received high doses
`of chemotherapy on a regimen that required transplant of bone marrow or stem cells.
`
`Using data from cell cultures, which showed that prolonged infusion of drugs
`retards the development of drug resistance, investigators at NCI tested paclitaxel by
`96-hour infusion in patients who had received two or more prior chemotherapy regimens
`and were resistant to doxorubicin, a group of patients with a dismal prognosis (40).
`Objective responses were seen in 48% of patients.
`
`TABLE II. Phase Π Trials of Paclitaxel in Metastatic Breast Cancer
`
`Institution
`(Reference)
`MD Anderson (35)
`Memorial (36)
`
`No. of
`Patients
`25
`26
`
`CETSG (37)
`
`471
`
`No. Prior
`Chemorx.
`1
`0 - 32% pts
`1 - 68% pts
`1 - 69%
`2 - 31%
`
`Memorial (38)
`
`Memorial (38)
`
`22
`
`24
`
`2
`
`>3
`
`Dose/infusion
`schedule (hr)
`200-250/24
`200-250/24
`+ G-CSF
`randomized
`135/3
`175/3
`200/24
`+ G-CSF
`200/24
`+ G-CSF
`MD Anderson (39)
`33
`20%
`175/24
`2
`MD Anderson (39)
`> 3
`35
`18%
`150/24
`NCI-96 (40)
`33
`48%
`140/96
`2
`Abbreviations: No., number; Chemorx., chemotherapy treatments
`
`Response
`
`56%
`62%
`
`22%
`29%
`36%
`
`21%
`
`Combination Trials with Doxorubicin. Because doxorubicin is the most effective
`drug for breast cancer, it was the logical choice for a combination trial. The scarcity
`of paclitaxel limited trials to two centers, M. D. Anderson and the NCI. The centers
`used different schedules, but both used G-CSF to diminish effects on the white blood
`cell counts. Based on the cell culture data with prolonged drug infusions, the NCI
`gave each drug by 72 hours concurrently (41). M. D. Anderson used a sequential
`
` Georg et al.; Taxane Anticancer Agents
`
`ACS Symposium Series; American Chemical Society: Washington, DC, 1994.
`
`

`
`40
`
`TAXANE ANTICANCER AGENTS
`
`approach, giving paclitaxel first over 24 hours followed by doxorubicin over 48 hours
`(39). The purpose of these phase I trials was to determine the maximum tolerated
`dose. Since the trials were conducted simultaneously, data on the incidence and type
`of side effects associated with various dose levels were shared by the investigative
`teams. The sequential combination with paclitaxel preceding doxorubicin (paclitaxel-
`doxorubicin) produced much more severe mucous membrane toxicity (mucositis) which
`precluded intake of solid food at paclitaxel doses which were 30% lower than those
`in the simultaneous combination. This suggested that paclitaxel potentiated the effects
`of doxorubicin. For this reason, the reverse sequence (doxorubicin-paclitaxel) was
`then studied at M. D. Anderson (42). With this sequence, the maximum tolerated
`doses of paclitaxel and doxorubicin were very similar to those reached in the NCI
`concurrent infusion schedule. To further define the nature of this drug interaction,
`pharmacokinetic studies were performed in a third group of patients to evaluate
`doxorubicin levels in patients who had received either sequential regimen. In the first
`cohort of patients doxorubicin preceded paclitaxel for course 1 and was reversed in
`course 2. In the second cohort of patients the sequences were reversed for courses
`1 and 2. The results showed that when paclitaxel preceded doxorubicin, the peak
`doxorubicin concentration at the end of infusion as well as the area under the
`concentration χ time curve (AUC) was 30% higher. Conversely, doxorubicin clearance
`was decreased by 70% when paclitaxel preceded it. The pharmacodynamic effects
`were similarly striking: the median granulocyte count, in the sequence paclitaxel-
`doxorubicin, was 0.2 granulocytes/mm3 versus 1.3 in the reverse sequence. Also, the
`incidence and severity of mucositis was greater when paclitaxel preceded doxorubicin.
`The conclusion was that if paclitaxel by 24-hour infusion is to be given in sequence
`with doxorubicin, doxorubicin should be given first.
`Although phase I studies are not designed to determine response rates with
`a narrow confidence interval, response data are evaluated. Independent of schedule,
`the combination of doxorubicin with paclitaxel by 24-hour or longer infusion resulted
`in 70% objective responses. Only 10% of these responses were complete, a condition
`necessary for development of a curative regimen. This is similar to standard aggressive
`combinations of doxorubicin with 5-fluorouracil and cyclophosphamide (MCAFH or
`MFACM). Additionally, the N