928
Gemcitabine in Pancreatic Carcinoma
The Pemetrexed/Gemcitabine Combination in
Pancreatic Cancer
Hedy Lee Kindler, M.D.
Director of Gastrointestinal Oncology, Section of
Hematology/Oncology, University of Chicago, Chi-
cago, Illinois.
Dr. Kindler is a consultant for Lilly Oncology and
has received research funding from Lilly Oncology.
Address for reprints: Hedy Lee Kindler, M.D., Sec-
tion of Hematology/Oncology, University of Chi-
cago, 5841 S, Maryland Avenue, MC2115, Chi-
cago, IL 60637; Fax: (773) 702-0963; E-mail:
hkindler@medicine.bsd.uchicago.edu
Manuscript received March 14, 2002; accepted
April 3, 2002
© 2002 American Cancer Society
Supplement to Cancer
Gemcitabine has modest antitumor activity in advanced pancreatic cancer. New
agents are clearly needed. Pemetrexed (ALIMTA®) is a novel antifolate that inhibits
thymidylate synthase, dihydrofolate reductase, and glycinamide ribonucleotide
formyltransferase. Pemetrexed has shown in vitro activity against pancreatic can-
cer cell lines. In a Phase I trial of pemetrexed, two patients with pancreatic cancer
achieved partial responses on the once every 21 days schedule. A Phase II trial of
pemetrexed in patients with advanced pancreatic cancer showed an objective
response rate of 6%, a one year survival of 28%, and a mild toxicity profile. The
combination of pemetrexed and gemcitabine is synergistic in vitro and was broadly
active in a Phase I trial. The pemetrexed/gemcitabine combination was evaluated
in a Phase II trial in 42 patients with advanced pancreatic cancer. The promising
activity observed in this study has led to an ongoing international, randomized,
Phase III trial in 520 patients comparing the pemetrexed/gemcitabine combination
with gemcitabine alone. Cancer 2002;95:928 –32.
© 2002 American Cancer Society.
DOI 10.1002/cncr.10755
ver 29,000 Americans developed cancer of the pancreas in 2001.1
O Only 1– 4% of all patients with pancreatic cancer will survive five
years, making it the fourth leading cause of cancer death in the United
States.1 Long term survival is limited to those individuals who have
resectable tumors. Since symptoms usually develop late, more than
80% of patients present with locally advanced or metastatic disease;
their survival is usually measured in months.2
Few chemotherapeutic agents produce objective responses in
patients with pancreatic cancer. Between 1991 and 1994, 25 new
investigational agents were evaluated in Phase II clinical trials. The
median response rate was 0% (range, 0 –14%), and the median sur-
vival observed was three months.3 Clearly, novel active agents are
needed.
Single agent gemcitabine is currently the chemotherapy standard
of care for patients with advanced pancreatic cancer. However, re-
sponse rates are modest. In the initial Phase II trial of 44 patients
reported by Casper et al., 11% of patients had an objective response,
but a subset of patients experienced a clinical benefit response that
consisted of an improvement in pain as reflected by analgesic con-
sumption, pain intensity, and improvement in performance status
(PS).4 A subsequent randomized Phase III trial (n ⫽ 126) by Burris et
al. compared gemcitabine to weekly 5-fluorouracil (5-FU). Patients
who received gemcitabine experienced clinical benefits and achieved
a modest survival advantage over those patients treated with 5-FU.5
Treatment with gemcitabine yielded an objective tumor response rate
of 5.4%, a clinical benefit response of 23.8%, and a one year survival

FIGURE 1
of 18%, compared with an objective tumor response
rate of 0%, a clinical benefit response rate of 4.8%, and
a one year survival of 2% for 5-FU.
The symptomatic benefit achieved with gemcitab-
ine and its mild toxicity profile have prompted numer-
ous investigators to combine other agents with gem-
citabine in an effort to enhance its activity.2
Pemetrexed (ALIMTA®)
Pemetrexed (ALIMTA®; N-[4-[2-(2-amino-3,4-dihy-
dro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl]-
benzoyl]-L-glutamic acid; Eli Lilly and Company, In-
dianapolis, IN) is a novel pyrrolopyrimidine-based
antifolate.6 Pemetrexed enters cells via the reduced
folate carrier and undergoes intracellular polygluta-
mation, resulting in prolonged intracellular retention.7
Pemetrexed and its polyglutamated metabolites in-
hibit several key folate-requiring enzymes, including
thymidylate synthase (TS), dihydrofolate reductase,
glycinamide ribonucleotide formyltransferase, and
aminoimidazole carboxamide ribonucleotide formyl-
transferase.7,8 Three of these pathways are illustrated
in Figure 1. The cytotoxicity of pemetrexed is thought
to be primarily mediated through TS inhibition.8 How-
ever, inhibition of the other folate enzymes may also
be important, since pemetrexed is active against hu-
man cancer cell lines that are resistant to raltitrexed
and 5-FU due to TS amplification.9 Pemetrexed shows
significant activity against a broad spectrum of tumors
Pemetrexed/Gemcitabine in Pancreatic Cancer/Kindler 929
in a variety of in vitro tumor models, including pan-
creatic cancer cell lines (generally considered che-
moresistant) in the human tumor cloning assay.10
Phase I Trials of Single Agent Pemetrexed
Three Phase I studies of pemetrexed have been con-
ducted using three different schedules of administra-
tion: once every 21 days; weekly for 4 weeks every 42
days; and daily for 5 days every 21 days. Table 1
summarizes these studies.
McDonald et al. administered a total of 116 cycles
daily for 5 days every 21 days to 38 patients with
advanced malignancies refractory or not amenable to
conventional therapy. Ten dose levels were evaluated,
ranging from 0.2 to 5.2 mg/m2/d. Dose limiting toxic-
ities (DLTs) included neutropenia and reversible ele-
vations of hepatic enzymes. The maximum tolerated
dose (MTD) was 4.0 mg/m2/d. Minor responses were
achieved in two patients, one each with nonsmall cell
lung cancer and colorectal cancer.11
Fifty eight courses of pemetrexed were adminis-
tered to the 25 patients who received the weekly for 4
weeks every 42 days schedule, as reported by Rinaldi
et al. Doses ranged from 10 to 40 mg/m2/wk. Neutro-
penia was dose-limiting at 40 mg/m2/week. Minor
responses were noted in two patients with colorectal
cancer.12
The same group of investigators also treated 37
patients on the once every 21 days schedule. A total of
930 CANCER Supplement August 15, 2002 / Volume 95 / Number 4

TABLE 1
Phase I Trials of Pemetrexed

Author Schedule No. of patients MTD DLT Partial responses Minor responses

McDonald11 daily ⫻ 5 every 21 days 38 4 mg/m2/d neutropenia, elevated liver None One NSCLC one
enzymes colorectal cancer
Rinaldi12 weekly ⫻ 4 every 42 days 25 40 mg/m2/wk neutropenia None two colorectal cancer
Rinaldi13 once every 21 days 37 600 mg/m2 neutropenia, thrombocytopenia, two pancreatic cancer two six colorectal cancer
fatigue colorectal cancer

MTD: Maximum tolerated dose; DLT: dose-limiting toxicity; NSCLC: nonsmall-cell lung cancer.

132 courses of pemetrexed were administered at nine TABLE 2

dose levels, ranging from 50 –700 mg/m2. The MTD Phase II Trial of Single Agent Pemetrexed in Pancreatic Cancer14
was 600 mg/m2, and DLTs were neutropenia, throm-
Feature Measure
bocytopenia, and cumulative fatigue. Other toxicities
included rash, mucositis, and elevations of hepatic No. of Patients 42
transaminases. Partial responses were observed in two Dose/Schedule 600 mg/m2 every 21 days
patients with pancreatic cancer and in two colorectal Vitamin supplementation None
cancer patients. Six minor responses were noted in Complete responses 1
Partial responses 1
patients with colorectal cancer. Three of the four pa- Overall response rate 5.7%
tients who achieved partial responses in this trial had Stable disease 40%
failed prior treatment with a TS inhibitor.13 One year survival 28%
The once every 21 days schedule, which permitted Median survival 6.5 months
the highest dose intensity with the most acceptable
toxicity profile, was selected for Phase II evaluation.
41% of patients, no infections greater than Grade 1
Phase II Trial of Pemetrexed in Pancreatic Cancer developed. Other Grade 3/4 hematologic toxicities in-
The two partial responses observed in the Phase I trial cluded thrombocytopenia and anemia in 17% and
of pemetrexed on the every 21 day schedule prompted 19% of patients, respectively. Nonhematologic Grade
Miller et al. to evaluate the activity of pemetrexed in a 3/4 toxicities included rash in 10% of patients, nausea
Phase II trial involving 42 previously untreated pa- in 17%, mucositis in 7%, and fatigue in 19%. Only one
tients with advanced pancreatic cancer.14 Patients re- patient developed alopecia. Grade 3/4 alanine amino-
ceived 600 mg/m2 of pemetrexed via a 10 minute transferase and aspartate aminotransferase elevations
intravenous (IV) infusion once every 21 days. Most were observed in 7% and 15% of patients, respectively.
patients (79%) had metastases; 21% had locally ad-
vanced unresectable disease. The median age was 60.3 The Pemetrexed/Gemcitabine Combination
years (range, 37 to 77). A total of 57% of the patients Pemetrexed and gemcitabine are synergistic in human
were male. Eastern Cooperative Oncology Group PS tumor xenograft models.15 Using in vitro clonogenic
was 0 in 32% of patients, 1 in 51%, and 2 in 17%. All assays, Adjei et al. reported cytotoxic synergy when
patients were evaluable for toxicity; 35 patients re- gemcitabine was administered prior to pemetrexed in
ceived at least two doses of chemotherapy and were human HCT-8 colon cancer cell lines.16 Other inves-
evaluable for response. Clinical benefit response was tigators have shown cytotoxic synergy for the opposite
not assessed. This study is summarized in Table 2. sequence, pemetrexed followed 24 hours later by gem-
The overall response rate in evaluable patients citabine, in HT29 colon cancer cell lines and xeno-
was 5.7%, with one complete response (lasting 16.2 grafts.17 These data suggest that patterns of interac-
months) and one partial response (lasting 6.9 tion between gemcitabine and pemetrexed may be
months). Stable disease was achieved in 40% of evalu- cell line specific.
able patients (n ⫽ 17), including five who maintained The synergy observed with gemcitabine and pem-
stable disease for more than six months. The 1 year etrexed in vitro led Adjei et al. to perform a Phase I
survival was 28%, median survival 6.5 months, and trial of the gemcitabine/pemetrexed combination in
median progression free survival 4 months (95% con- patients with advanced solid tumors.18 Gemcitabine
fidence interval [CI], 1.6 – 4.7 months). was administered at doses of 1000 or 1250 mg/m2 IV
Although Grade 3/4 neutropenia was observed in over 30 minutes on Days 1 and 8 of a 21 day cycle.

TABLE 3
Phase I Trial of Pemetrexed Plus Gemcitabine in Patients with Advanced Solid Tumors18
Feature Group I
Patients 34
Cycles 164
Vitamin supplementation None
Maximum tolerated dose Gemcitabine 1000 mg/m2 Days 1 and 8
2
Pemetrexed 500 mg/m Day 1
Dose limiting toxicity Neutropenia
Partial responses 12: 3 colorectal cancer, 3 nonsmall cell lung cancer, 2 cholangiocarcinoma, 2 ovarian cancer,
1 mesothelioma, 1 breast cancer
Pemetrexed was delivered 90 minutes later in escalat-
ing dose levels of 200, 300, 400, 500, and 600 mg/m2 on
Day 1 of a 21 day cycle (Group I). Since neutropenia
that developed on Day 8 led to reduction or omission
of the gemcitabine dose in more than 57% of treat-
ment courses, a second group of patients (Group II)
received pemetrexed on Day 8. The results of this trial
are summarized in Table 3.
In Group I, 35 patients received 164 cycles of
pemetrexed/gemcitabine. Twenty one patients in
Group II received 85 cycles of chemotherapy. Neutro-
penia was dose limiting. Other toxicities included
rash, fatigue, and increased transaminases. The rec-
ommended Phase II doses were 1250 mg/m2 of gem-
citabine on Days 1 and 8 and 500 mg/m2 of pem-
etrexed given on Day 8. This combination showed
activity in a broad range of tumors. There were 12
partial responses in the 35 evaluable patients. Phar-
macokinetic analysis determined that prior adminis-
tration of gemcitabine did not alter the disposition of
pemetrexed.
Vitamin Supplementation in Pemetrexed Trials
Folic acid administered with pemetrexed preserves
antitumor activity and reduces toxicity in human tu-
mor cell lines and in tumor bearing mice.19 A very
sensitive marker of folate status is plasma homocys-
teine. In multivariate analysis, a baseline homocys-
teine level ⱖ 12 ␮M is a predictor of Grade 3/4 pem-
etrexed toxicities, including diarrhea, rash,
neutropenia, thrombocytopenia, mucositis, and fa-
tigue.20 Supplementation with dietary levels of folate
and B12 significantly decreases toxicities related to
pemetrexed.19 Since November 1999, all patients on
pemetrexed trials have received oral supplementation
with 350-1000 ␮g of folic acid daily and 1000 ␮g B12
intramuscularly every nine weeks beginning at least
one week prior to chemotherapy and continuing
throughout treatment. It is thought that supplemen-
tation of folic acid at a level equivalent to an American
Pemetrexed/Gemcitabine in Pancreatic Cancer/Kindler 931
Group II
21
85
None
Gemcitabine 1250 mg/m2 Days 1 and 8
Pemetrexed 500 mg/m2 Day 8
Neutropenia
diet should not affect the efficacy of pemetrexed. B12 is
required for the 15% of patients who are B12-deficient.
Phase II Trial of Pemetrexed/Gemcitabine in
Pancreatic Cancer
Given the known, albeit modest, activity of both gem-
citabine and pemetrexed in pancreatic cancer, the in
vitro synergy of the two agents, and the broad activity
of the combination observed in the Phase I trial, Kin-
dler et al. initiated a multicenter Phase II trial of gem-
citabine plus pemetrexed in September 1999. The trial
enrolled 42 previously untreated patients with histo-
logically proven, locally advanced or metastatic pan-
creatic cancer. Patients were required to have a
Karnofsky PS of at least 60%, a life expectancy of 12
weeks, bidimensionally measurable disease, and nor-
mal hematologic, hepatic, and renal function.
Patients received gemcitabine 1250 mg/m2 by IV
infusion over 30 minutes on Days 1 and 8 of a 21 day
cycle. Pemetrexed 500 mg/m2 was administered by IV
infusion over 10 minutes on Day 8, 90 minutes after
gemcitabine. All patients received premedication with
dexamethasone to prevent rash. In November 1999,
when safety data determined that toxicity could be
prevented with folate and B12 supplementation, all
patients were administered dietary levels of these vi-
tamins. Computed tomography scans were obtained
every two cycles to assess for response. Clinical benefit
response, assessed by changes in pain, PS, or weight
change, was monitored weekly.
The results of this trial will be presented at the
2002 Meeting of the American Society of Clinical On-
cology.
Phase III Trial of Pemetrexed/Gemcitabine
Promising results obtained in the Phase II trial of the
combination of pemetrexed plus gemcitabine led to
the development of an ongoing Phase III international
trial in 520 patients. Previously untreated patients
with locally advanced or metastatic pancreatic cancer
932 CANCER Supplement August 15, 2002 / Volume 95 / Number 4
are randomized by PS, disease stage, investigational
center, and baseline homocysteine level to one of two
treatment regimens: gemcitabine 1,000 mg/m2 by IV
infusion weekly on Days 1, 8, and 15 every 28 days, or
gemcitabine 1,250 mg/m2 by IV infusion on Days 1
and 8 of a 21 day cycle plus pemetrexed 500 mg/m2 by
IV infusion on Day 8. Patients on the combination arm
also receive vitamin supplementation with folate and
B12. End points include overall survival, progression
free survival, objective response rate, quality of life,
and toxicity.
CONCLUSIONS
Pemetrexed, a new generation antifolate antimetabo-
lite, has activity in pancreatic cancer. The 6% response
rate, while modest, is similar to other single agents in
previously untreated patients with this disease, while
the 28% one year survival is promising. Pemetrexed is
synergistic with gemcitabine in vitro, and the combi-
nation of these two drugs was broadly active and well
tolerated in a Phase I study. The pemetrexed/gemcit-
abine combination has been evaluated in a recently
completed Phase II trial. An ongoing Phase III trial
that compares the combination of pemetrexed/gem-
citabine to single agent gemcitabine will help to de-
termine whether this combination improves survival
and quality of life for patients with advanced pancre-
atic cancer.
REFERENCES
1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer
statistics 2001. CA Cancer J Clin. 2001;51:15–36.
2. Heinemann V. Gemcitabine: progress in the treatment of
pancreatic cancer. Oncology. 2001;60:8 –18.
3. Rothenberg ML. New developments in chemotherapy for
patients with advanced pancreatic cancer. Oncology. 1996;
10(9 Suppl):18 –22.
4. Casper ES, Green MR, Kelsen DP, et al. Phase II trial of
gemcitabine (2⬘2⬘ difluorodeoxycytidine) in patients with
adenocarcinoma of the pancreas. Invest New Drugs. 1994;
12:29 –34.
5. Burris HA 3rd, Moore MJ, Andersen J, et al. Improvements in
survival and clinical benefit with gemcitabine as first-line
therapy for patients with advanced pancreas cancer: a ran-
domized trial. J Clin Oncol. 1997;15:2403–2413.
6. Taylor EC, Kuhnt D, Shih C, et al. A dideazatetrahydrofolate
analogue lacking a chiral center at C-6, N-[4-[2-(2-amino-
3,4-dihydro-4-oxo-7H-pyrrolo[2,3-d]pyrimidin-5-yl)ethyl-
]benzoyl]-L-glutamic acid, is an inhibitor of thymidylate
synthase. J Med Chem. 1992;35:4450 – 4454.
7. Mendelsohn LG, Shih C, Chen VJ, Habeck LL, Gates SB,
Shackelford KA. Enzyme inhibition, polyglutamation, and
the effect of LY231514 (MTA) on purine biosynthesis. Semin
Oncol. 1999;26(2 Suppl 6):42– 47.
8. Shih C, Chen VJ, Gossett LS, et al. LY231514, a pyrrolo[2,3-
d]pyrimidine-based antifolate that inhibits multiple folate-
requiring enzymes. Cancer Res. 1997;57:1116 –1123.
9. Schultz RM, Patel VF, Worzalla JF, Shih C. Role of thymidy-
late synthase in the antitumor activity of the multitargeted
antifolate, LY231514. Anticancer Res. 1999;19:437– 443.
10. Britten CD, Izbicka E, Hilsenbeck S, et al. Activity of the
multitargeted antifolate LY231514 in the human tumor clon-
ing assay. Cancer Chemother Pharmacol. 1999;44:105–110.
11. McDonald AC, Vasey PA, Adams L, et al. A Phase I and
pharmacokinetic study of LY231514, the multitargeted anti-
folate. Clin Cancer Res. 1998;4:605– 610.
12. Rinaldi DA, Burris HA, Dorr FA, et al. Initial Phase I evalu-
ation of the novel thymidylate synthase inhibitor, LY231514,
using the modified continual reassessment method for dose
escalation. J Clin Oncol. 1995;13:2842–2850.
13. Rinaldi DA, Kuhn JG, Burris HA, et al. A Phase I evaluation of
multitargeted antifolate (MTA, LY231514) administered ev-
ery 21 days, utilizing the modified continual reassessment
method for dose escalation. Cancer Chemother Pharmacol.
1999;44:372–380.
14. Miller KD, Picus J, Blanke C, et al. Phase II study of the
multitargeted antifolate LY231514 (ALIMTA, MTA, pem-
etrexed disodium) in patients with advanced pancreatic
cancer. Ann Oncol. 2000;11:101–103.
15. Teicher BA, Chen V, Shih C, et al. Treatment regimens
including the multitargeted antifolate LY231514 in human
tumor xenografts. Clin Cancer Res. 2000;6:1016 –1023.
16. Adjei AA, Erlichman C, Thornton D, et al. Synergistic cyto-
toxicity of MTA (LY231514) and gemcitabine in vitro and in
vivo. 10th NCI-EORTC Symposium on New Drugs in Cancer
Therapy. Ann Oncol. 1998;9:168.
17. Tonkinson JL, Worzalla JF, Teng CH, Mendelsohn LG. Cell
cycle modulation by a multitargeted antifolate, LY231514,
increases the cytotoxicity and antitumor activity of gemcit-
abine in HT29 colon carcinoma. Cancer Res. 1999;59:3671–
3676.
18. Adjei AA, Erlichman C, Sloan JA, et al. Phase I and pharma-
cologic study of sequences of gemcitabine and the multitar-
geted antifolate agent in patients with advanced solid tu-
mors. J Clin Oncol. 2000;18:1748 –1757.
19. Worzalla JF, Shih C, Schultz RM. Role of folic acid in mod-
ulating the toxicity and efficacy of the multitargeted antifo-
late, LY231514. Anticancer Res. 1998;18:3235–3239.
20. Niyikiza C, Walling J, Thornton D, Seitz D, Allen R. LY231514
(MTA): relationship of vitamin metabolite profile to toxicity
[abstract]. Proc Am Soc Clin Oncol. 1998;17:558a.