Results of a Phase II Study of Maytansine in Patients
`With Breast Carcinoma and Melanoma 1•2
`
`Fernando Cabanillas,* Gerald P. Bodey, Michael A. Burgess, and Emil J
`Freireich 3
`
`Brief Reports
`and
`Preliminary
`Communications
`
`SUMMARY
`
`During the phase I study of maytansine at our
`institution, some activity was observed against breast
`carcinoma and melanoma. A phase II study was thus
`initiated to more thoroughly investigate the activity of
`this drug against these two tumors. In 33 evaluable
`patients with melanoma, no complete or partial re(cid:173)
`sponses were observed. Twenty-one evaluable pa(cid:173)
`tients with breast cancer were entered and only one
`response (partial) was seen. The toxicity was similar
`to that observed in the phase I study and consisted
`mainly of diarrhea, paresthesias, phlebitis, and flu(cid:173)
`like symptoms. Myelosuppression was infrequent and
`was short-lived when it occurred.
`
`[Cancer Treat Rep 63:507-509, 1979]
`
`Maytansine is a recently discovered ansa macro(cid:173)
`lide drug derived from the plant Maytenus ovatus
`(1). It is active against the experimental P388
`leukemia and B16 melanoma animal tumors. 4 In a
`phase I trial using a 3-day schedule conducted at
`our institute, activity was detected against breast
`carcinoma and melanoma (2). The dose we recom(cid:173)
`mended for further clinical trials was 0.5 mg/m 2
`dally x 3 every 2-3 weeks according to toxicity. The
`dose-limiting toxic effect of this agent was diarrhea;
`myelosuppression was mild or nonexistent. Conse(cid:173)
`quently, it was thought that this would be an
`
`1Received Oct 31, 1978; revised Jan 11, 1979; accepted Jan 29,
`1979.
`'Supported in part by contract N01-CM-57042 from the Divi(cid:173)
`sion of Cancer Treatment, National Cancer Institute, National
`Institutes of Health, Department of Health, Education, and
`Welfare.
`3Department of Developmental Therapeutics, The University
`of Texas System Cancer Center, M. D. Anderson Hospital and
`Tumor Institute, Houston.
`'Helman L, Henney J, and Slavik M. Clinical brochure: may(cid:173)
`tansine. Prepared by the National Cancer Institute, Bethesda,
`Md, 1976.
`*Rep1·int 1·equests to: Fernando Cabanillas, MD, Department
`of Developmental Therapeutics, The University of Texas System
`Cancer Center, M. D. Anderson Hospital and Tumor Institute,
`6723 Bertner Ave, Houston, Tex 77030.
`
`interesting compound for further clinical trials. We
`initiated a phase II trial in patients with breast
`carcinoma and melanoma in June 1977.
`
`MATERIALS AND METHODS
`
`Patients with advanced histologically documented
`carcinoma of the breast or melanoma who had
`failed to respond to regimens of higher priority were
`candidates for this study. Physical examination was.
`performed and measurable lesions were recorded.
`The following pretreatment studies were performed:
`cbc, platelet count, urinalysis, SMA-12 profile before
`every cycle of treatment, bone marrow aspiration
`and biopsy, and chest roentgenogram. The cbc was
`repeated at least once a week during therapy. Other
`tests to evaluate the presence of measurable tumor
`were obtained as indicated and were repeated after
`two cycles of therapy in order to evaluate response
`to treatment.
`This study was designed as a randomized compar(cid:173)
`ison of two different schedules: (a) 0.5 mg/m 2 in 500
`ml of dextrose in water over 30 minutes daily x 3
`repeated in 2-3 weeks with subsequent increments
`of 0.1 mg/m 2 according to toxicity, or (b) 0.75 mg/m 2
`in 500 ml of dextrose in water over 30 minutes iv
`weekly with 0.25-mg/m 2 increments every week ac(cid:173)
`cording to toxicity.
`Early during the study, it was found that most
`patients with normal liver function tests could
`tolerate a dose of 0.6 mg/m 2 daily x 3 and, thus, the
`starting dose was increased. In the weekly schedule,
`the starting dose was increased to 1 mg/m 2• Patients
`with abnormal liver function tests (SGOT or alkaline
`phosphatase levels increased to twice the normal
`levels) received a 50% dose reduction, and if the
`first course was well-tolerated the following dose
`was increased to the standard starting dose. The
`dose of maytansine was modified according to the
`severity of diarrhea. When this was severe (requir(cid:173)
`ing iv fluids for treatment of dehydration or lasting
`> 5 days), the dose was reduced by 33%. If ileus
`developed, the dose was reduced by 50% after
`
`Cancer Treatment Reports Vol. 63, No. 3, March 1979
`
`507
`
`IMMUNOGEN 2003, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`resolution of the obstruction. Myelosuppression was
`not used as a criterion for dose modification because
`it occurred infrequently and was almost always
`associated with diarrhea.
`After several patients were entered in the study,
`the weekly schedule was discontinued at the re(cid:173)
`quest of the National Cancer Institute (not because
`of excessive toxicity). Thus, fewer patients were
`entered in the weekly regimen compared to the
`daily x 3 regimen.
`
`RESULTS
`
`Response of Patients With Melanoma
`
`At the close of this study, 33 evaluable patients
`with melanoma had been entered in the 3-day
`schedule and 12 had been entered in the weekly
`schedule. All patients had visceral disease with or
`without associated nodal or skin lesions. There were
`no complete or partial responses. Four patients had
`a less than partial remission, all except one in soft
`tissue disease. Three of these minimal responses
`occurred in subcutaneous melanoma nodes which
`decreased in size with no evidence of a simultaneous
`response in visceral sites. The fourth minimal re(cid:173)
`sponse was in a patient who had malignant ascites
`and pleural effusion which decreased in size with
`no evident response in a supraclavicular mass. Five
`patients were stable for 6 weeks or more and 24
`failed to respond at all. Of the four less than partial
`responses, two occurred in the 3-day schedule and
`two in the weekly schedule.
`
`Response of Patients With Breast Carcinoma
`
`Twenty-one evaluable patients with breast carci(cid:173)
`noma were entered. Eighteen were treated with the
`3-day schedule and three with the weekly schedule.
`With the exception of one patient who had exclu(cid:173)
`sively nodal disease, all others had visceral disease.
`There were no complete remissions. One patient
`had a partial remission (bone marrow disease) last(cid:173)
`ing 4 months and one had a less than partial
`remission in a supraclavicular node. Both of these
`patients received the 3-day schedule. Three patients
`had stable disease for ~ 6 weeks and 16 failed to
`respond. These patients were heavily pretreated
`and in most cases maytansine was the third or
`fourth drug used.
`
`Myelosuppressive Toxicity
`
`Since more patients with breast carcinoma had a
`history of prior radiation therapy and prior expo(cid:173)
`sure to myelosuppressive drugs (eg, mitomycin C)
`
`with cumulative toxicity (compared to patients with
`melanoma), the myelosuppressive toxicity of may(cid:173)
`tansine has been analyzed separately for patients
`with these two diagnoses. The myelosuppressive
`toxicity is shown in table 1. Myelosuppression was
`observed infrequently and was seen less frequently
`in the group that received the weekly schedule.
`
`Gastrointestinal Toxicity
`
`Gastrointestinal toxicity for patients with both
`breast carcinoma and melanoma is shown in table 2
`according to the schedule of treatment. The total
`amount of drug in milligrams per square meter that
`was delivered per month is also shown in table 2.
`The major and most frequent toxic. effect of maytan(cid:173)
`sine was diarrhea, which in some cases was severe
`enough to require hospitalization. This problem
`seemed to occur less frequently with the weekly
`schedule. The total amount of drug delivered per
`month with the weekly schedule was higher, yet the
`toxicity was less.
`
`Other Toxic Effects
`
`Aside from myelosuppression and gastrointestinal
`toxicity, other side effects consisted of paresthesias
`in six patients, superficial phlebitis in four patients,
`
`TABLE 1.-Myelosuppressive toxicity of maytansine
`
`Melanoma
`No. of evaluable courses
`No. of patients with
`myelosuppression*
`
`Breast carcinoma
`No. of evaluable courses
`No. of patients with
`myelosuppression*
`
`3-day
`schedule
`
`Weekly
`schedule
`
`33
`4
`
`24
`7
`
`36
`1
`
`8
`0
`
`*Any drop in the wbc count to < 3000/mm 3 or the platelet
`count to < 100,000/mm3•
`
`TABLE 2.-Gastrointestinal toxicity of 3-day vs weekly schedule
`compared to total maytansine dose delivered (includes both
`breast carcinoma and melanoma patients)
`
`3-day
`schedule
`
`Weekly
`schedule
`
`No. of evaluable courses
`
`102
`
`49
`
`No. of patients with(cid:173)
`Nausea/vomiting
`Diarrhea
`Ileus
`Cohstipation
`
`Dose delivered permo (mg/m 2)
`
`36(35%)
`43(42%)
`3(3%)
`6(6%)
`
`3.29
`
`17(35%)
`16(33%)
`1(2%)
`1(2%)
`
`4.07
`
`508
`
`Cancer Treatment Reports
`
`IMMUNOGEN 2003, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`flu-like symptoms in four patients, fever or chills in
`three patients, skin slough in one patient, and
`mucositis in one patient. In this study, the develop(cid:173)
`ment of paresthesias was not related to previous
`vinca alkaloid administration. None of the patients
`who experienced this toxicity had been previously
`exposed to vinca alkaloids.
`
`DISCUSSION
`
`The response rate observed in this phase II study
`of maytansine has been disappointingly low. Using
`the schedules outlined above, maytansine cannot be
`considered an active drug against melanoma or
`breast carcinoma. However, further trials are justi(cid:173)
`fied in other tumors. Furthermore, it is conceivable
`that the schedule selected for this trial might not
`be the most adequate. Pharmacology studies are
`necessary to delineate the pharmacokinetics of this
`interesting compound so that a rational schedule of
`administration can be designed. Unfortunately, an
`assay sensitive enough to measure the minute
`levels of maytansine is not available.
`The weekly schedule of administration allowed
`more drug to be administered with less toxicity
`than the 3-day schedule. This, however, did not
`result in a significant increase in response rate.
`
`Should this drug prove to be active against other
`tumors, the weekly schedule might be preferable if
`the drug has a sufficiently long half-life.
`Because of its virtual lack of myelosuppression
`and overlapping toxic effects with most of the
`currently available drugs, this agent should be
`explored more extensively. Using a single-dose
`schedule of maytansine, the National Cancer Insti(cid:173)
`tute has also found very little myelosuppression (3).
`This compound possesses significant activity against
`experimental animal tumors. The type of toxicity
`observed in this and other studies suggests a poten(cid:173)
`tially important role in combination with myelosup(cid:173)
`pressive drugs. Further exploration of its activity in
`other tumors and with several different schedules
`is necessary.
`
`REFERENCES
`
`1. KUPCHAN SM, KOMODA Y, COURT W A, ET AL. Maytansine, a
`novel antileukemic ansa macrolide from Maytenus ovatus. J
`Am Chern Soc 94:1354-1356, 1972.
`2. -CABANILLAS F, RODRIGUEZ'V, HALL SW, ET AL. Phase I
`study of maytansine using a 3-day schedule. Cancer Treat
`Rep 62:425-428, 1978.
`3. CHABNER BA, LEVINE AS, JOHNSON BL, ET AL. Initial clinical
`trials of maytansine, an antitumor plant alkaloid. Cancer
`Treat Rep 62:429-433, 1978.
`
`Vol. 63, No. 3, March 1979
`
`509
`
`IMMUNOGEN 2003, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`