Fax: (212) 717-3278
A phase 2 study of pemetrexed plus gemcitabine every 2 weeks for patients with recurrent or metastatic head and neck squamous cell cancer
Article first published online: 4 OCT 2010
Copyright © 2010 American Cancer Society
Volume 117, Issue 4, pages 795–801, 15 February 2011
How to Cite
Fury, M. G., Haque, S., Stambuk, H., Shen, R., Carlson, D. and Pfister, D. (2011), A phase 2 study of pemetrexed plus gemcitabine every 2 weeks for patients with recurrent or metastatic head and neck squamous cell cancer. Cancer, 117: 795–801. doi: 10.1002/cncr.25464
- Issue published online: 3 FEB 2011
- Article first published online: 4 OCT 2010
- Manuscript Accepted: 28 APR 2010
- Manuscript Revised: 16 APR 2010
- Manuscript Received: 23 DEC 2009
- head and neck;
- phase 2;
Preclinical studies suggest that additive or synergistic effects are achieved with the combination of pemetrexed plus gemcitabine. A phase 1 study of pemetrexed plus gemcitabine given every 2 weeks demonstrated encouraging preliminary efficacy against head and neck squamous cell cancer (HNSCC).
This was an open-label, single-institution, single-arm, phase 2 study for patients who had received no more than 2 cytotoxic regimens for recurrent and/or metastatic HNSCC. All patients received pemetrexed 500 mg/m2 intravenously plus gemcitabine 1250 mg/m2 intravenously every 2 weeks with vitamin B12 and folate support. The primary endpoint was the objective response rate according to Response Evaluation Criteria in Solid Tumors (RECIST); secondary endpoints were to estimate overall survival and to evaluate safety and tolerability.
Twenty-five patients received therapy. All patients had received prior radiotherapy, and half had received prior cytotoxic chemotherapy for recurrent and/or metastatic disease. Neutropenia (grade ≥3) occurred in 24% of patients. Four patients (16%) had a partial response (PR) according to RECIST, and 5 additional patients (20%) had objective tumor reductions of >20 but <30% did not meet RECIST criteria for a PR. The median overall survival for all treated patients was 8.8 months.
Treatment with pemetrexed plus gemcitabine every 2 weeks with vitamin support generally was well tolerated. The results of this study provided further evidence that pemetrexed may have significant palliative activity against advanced HNSCC. Cancer 2011. © 2010 American Cancer Society.
For patients with metastatic or locally or regionally recurrent head and neck squamous cell carcinoma (HNSCC) that is not suitable for surgery or further radiation, the median survival is approximately 6 to 9 months for patients who are treated with standard cytotoxic chemotherapy.1, 2 Objective responses are achieved with single agents, such as platinums, taxanes, and antifolates. Methotrexate may be considered a historic standard based on excellent tolerability and on survival outcomes comparable to those observed with more toxic regimens.3 To our knowledge, no randomized trial has demonstrated a significant improvement in overall survival for combination therapy with cytotoxic agents compared with monotherapy using a single cytotoxic agent, although response rates and toxicities may be increased with the former approach.1 A phase 3 trial demonstrated similar response rates and median overall survival for patients who received cisplatin plus 5-fluorouracil (27% and 8.7 months, respectively) or cisplatin plus paclitaxel (26% and 8.1 months, respectively).4 Even with the addition of cetuximab, a monoclonal antibody directed against the epidermal growth factor receptor, to the platinum plus 5-flourouracil doublet, the median overall survival was only 10.1 months in a phase 3 trial.5 These results highlight the need for novel palliative regimens in this patient population.
Pemetrexed is a multitargeted antifolate that inhibits multiple enzymes involved in both pyrimidine and purine synthesis, including thymidylate synthase (TS), dihydrofolate reductase (DHFR), and glycinamide ribonucleotide formyl transferase (GARFT). Inhibition of TS and DHFR by pemetrexed overlaps with the mechanisms of other agents with activity against HNSCC, such as 5-fluorouracil and methotrexate. A phase 2 study evaluated pemetrexed (500 mg/m2 intravenously every 21 days) in 35 patients who had locally advanced or metastatic HNSCC that was not amenable to curative surgery or radiotherapy.6 Fourteen patients (40%) in that study had received prior chemotherapy, and 27 patients (77%) had received prior radiotherapy. Patients did not receive supplementation with vitamin B12 or folate. The study mandated an interval of ≥6 months between study entry and prior chemotherapy or radiotherapy. The objective response rate was 26.5% (95% confidence interval [CI], 12.9%-44.4%), and the median overall survival was 7.3 months. Twenty-four patients (68.6%) experienced grade 3/4 neutropenia, and febrile neutropenia occurred in 4 patients (11.4%).
Gemcitabine is a pyrimidine antimetabolite that is anabolized into a diphosphate form, which inhibits ribonucleotide reductase, and a triphosphate form, which is incorporated into DNA, resulting in chain termination. In recurrent or metastatic HNSCC, gemcitabine monotherapy has modest activity. In a phase 2 study, 47 patients with advanced and/or recurrent HNSCC received gemcitabine 800 mg/m2 intravenously on Days 1, 8, and 15 of a 28-day cycle (the gemcitabine dose was increased to 1250 mg/m2 on the same schedule for an additional 12 patients). The overall response rate was 13% (7 of 54 evaluable patients; 95% CI, 4%-22%).7 In another phase 2 trial, 26 patients with recurrent or metastatic HNSCC received intravenous gemcitabine 1250 mg/m2 weekly for 3 weeks followed by a 1-week rest. There were no objective responses (95% CI, 0%-13%).8
Because both pemetrexed and gemcitabine deplete intracellular deoxynucleotide triphosphate pools through the inhibition of multiple enzymes, it was proposed that the combination of these drugs might yield synergistic activity against solid tumors.9 In cell culture assays, synergistic activity has been described with combined administration of pemetrexed and gemcitabine.10-12 In studies of H460 nonsmall cell lung cancer xenografts, simultaneous administration of both drugs yielded synergy.13 Pemetrexed (150 mg/kg intraperitoneal injection for 5 consecutive days per week) and gemcitabine (60 mg/kg intraperitoneal injection every third day) were evaluated as monotherapy and as combination therapy. Tumor growth delay (TGD) was 4 days with pemetrexed monotherapy and 8 days with gemcitabine monotherapy. When both drugs were combined at these same doses and schedules, TGD was 20 days.13
A phase 1 study performed at our institution of pemetrexed plus gemcitabine established a maximum tolerated dose (MTD) for patients with advanced solid tumors of pemetrexed 500 mg/m2 intravenously plus gemcitabine 1500 mg/m2 intravenously given every 14 days in the context of supplementation with vitamin B12 and folate. The toxicities in the phase 1 study were mostly hematologic, as expected. Among 7 patients with HNSCC who were enrolled in the study, there were 2 objective responses.14
These observations suggest that pemetrexed may have significant activity against HNSCC and that gemcitabine may have a role in enhancing the activity of pemetrexed. Therefore, we conducted a phase 2 study to further evaluate this doublet regimen.
MATERIALS AND METHODS
This was an open-label, single-institution trial and was approved by the Institutional Review Board of Memorial Sloan-Kettering Cancer Center. Adult patients aged ≥18 years who had a Karnofsky performance status (KPS) ≥70% were eligible. All patients had recurrent or metastatic epidermoid/squamous carcinoma of the oral cavity, pharynx, larynx, paranasal sinus, or HNSCC of unknown primary. Pathologic confirmation at this institution was required. Patients were not included if they had received >2 prior cytotoxic chemotherapy regimens in the recurrent or metastatic disease setting. The presence of measurable disease was required. Patients also must not have received radiation therapy within 4 weeks of starting gemcitabine and pemetrexed. Required laboratory parameters were: an absolute neutrophil count >1.5 × 109/L; hemoglobin >9 g/dL; platelet count >100 × 109/L; serum bilirubin ≤1.5 times the upper limit of normal; alkaline phosphatase (AP), aspartate transaminase (AST), and alanine transaminase (ALT) levels ≤3 times the upper normal limits (AP, AST, and ALT up to 5 times the upper limit of normal were allowed for patients who had hepatic metastases); and a calculated creatinine clearance ≥45 mL per minute based on the standard Cockcroft and Gault formula.
Exclusion criteria were serious concomitant disorders (for example, active infection); brain metastases; previous exposure to pemetrexed or gemcitabine; pregnancy or lactation; refusal or inability to take folic acid, vitamin B12, and dexamethasone; and the use of nonsteroidal anti-inflammatory agents (NSAIDS) with short-elimination half-lives (for example, ibuprofen) for 2 days before, the day of, and 2 days after receiving pemetrexed (a total of 5 days held). If patients were taking an NSAID or salicylate with a long half-life (for example, piroxicam or nabumetone), then they were required to hold the medication for 5 days before, the day of, and 2 days after receiving pemetrexed (a total of 8 days held). All patients resided within geographic proximity to the treatment center for adequate follow-up during treatment at the investigators' discretion. All patients provided written informed consent. Men and women of reproductive potential were required to use a medically acceptable form of birth control while on study.
In this study, pemetrexed was given at the MTD (500 mg/m2 intravenously every 2 weeks), as established in the previous phase 1 study at this hospital.14 Many patients with advanced head and neck cancer have received prior chemoradiation therapy, and the potential for severe myelotoxicity was a concern during the planning and development of this study. Because the proposed role for gemcitabine was as a chemosensitizer for pemetrexed, the starting dose for gemcitabine, 1250 mg/m2 intravenously every 2 weeks, was lower than the MTD established in the phase 1 study.14 This choice of dose allowed us to explore the potential for gemcitabine to enhance the activity of pemetrexed without unacceptably increasing the risk of severe myelotoxicity in this population of pretreated patients with advanced disease.
Pretreatment assessment of all patients included complete medical history and physical examination. Baseline laboratory studies within 14 days of treatment included a complete blood count with white blood cell differential and platelet counts; a comprehensive metabolic profile (including electrolytes, bicarbonate, blood urea nitrogen, creatinine, glucose, AP, AST, ALT, total bilirubin, total protein, albumin, and glucose); prothrombin time; and activated partial thromboplastin time. Radiologic imaging studies of sites of measurable disease (preferably computed tomography [CT] scans or magnetic resonance imaging [MRI] studies) were obtained within 4 weeks before therapy. An electrocardiogram was obtained within 8 weeks before therapy. Pemetrexed and gemcitabine were provided by Eli Lilly and Company (Indianapolis, Ind).
Supplementation with folic acid and vitamin B12 began at least 1 week before chemotherapy. Vitamin supplementation consisted of oral folate 250 to 1000 μg daily and a single intramuscular injection of vitamin B12 1000 μg administered every 9 weeks. Patients continued vitamin supplementation (folate daily, vitamin B12 injections every 9 weeks) until 3 weeks after the last dose of pemetrexed.
Pemetrexed was given immediately before gemcitabine infusion. Patients received pemetrexed (500 mg/m2 intravenously over approximately 10 minutes) followed by gemcitabine (1250 mg/m2 intravenously over approximately 30 minutes) on Days 1 and 15 of each 28-day cycle. For rash prophylaxis, patients received oral dexamethasone (4 mg twice daily) on the day before, the day of, and the day after each pemetrexed/gemcitabine treatment.
Evaluation of Response
Radiologic response criteria were based on previously published RECIST guidelines.15 Radiologic imaging studies for response assessment of indicator lesions were obtained every 2 cycles of therapy (typically, approximately every 8 weeks) and/or after the last final study treatment. The best overall response was defined as the best response recorded from the start of treatment until disease progression/recurrence. Stable disease was measured from the start of the treatment until the criteria for disease progression was met.
The dose-reduction scheme is shown Table 1. The first dose reduction involved gemcitabine rather than pemetrexed. This dose reduction scheme was selected because phase 2 data suggest that, in HNSCC, the activity of pemetrexed is greater than that of gemcitabine.6-8
|Dose Level||Pemetrexed Dose, mg/m2||Gemcitabine Dose, mg/m2|
|Starting dose level: 0||500||1250|
After a dose reduction for toxicity, no dose re-escalation was allowed. Any patient with 2 prior dose reductions who experienced a toxicity that would cause a third dose reduction had to be withdrawn from study therapy. Treatment delay for up to 28 days from the last treatment day was permitted to allow a patient sufficient time to recover from any study drug-related toxicity. A patient who still could not be administered study drug after 28 days from the time of last treatment was withdrawn from the study.
The primary endpoint was to determine the objective response rate of pemetrexed plus gemcitabine administered every 14 days for patients with recurrent or metastatic HNSCC. In this population of patients, the historic response rate for conventional, single-agent chemotherapy is approximately 10%.3 An experimental regimen is considered worthy of further study if the observed response rate is ≥30%. To demonstrate an improvement compared with the historic control response rate, a sample size of 25 patients was needed to demonstrate an improvement to a response rate of 30% with alpha and beta error rates of 0.1. The study followed a single-stage design based on the binomial distribution. If ≥5 of 25 patients responded, then we would recommend further study of the regimen. If 25 patients were treated, then the response rate could be estimated with a 95% CI of ≤19.6%. For secondary objectives, overall survival for each group was estimated according to the Kaplan-Meier method. Descriptive statistics were used for data regarding safety and tolerability.
Patient Characteristics and Exposure to the Study Regimen
Twenty-six patients were enrolled between October 2006 and February 2008. Patient characteristics are summarized in Table 2. Most patients were men (77%), and the median KPS was 80%. The most common anatomic subset was oropharynx (38%), followed by oral cavity (31%), and larynx (19%). Fifty percent of patients (13 of 26) had received a previous cytotoxic chemotherapy regimen for recurrent and/or metastatic disease (median number of regimens, 0.5; range, 0-1). All patients had received previous radiation therapy. One patient with oral cavity cancer was registered but was removed from study because of disease progression before the receipt of any chemotherapy on the study; this patient was replaced and was not included in the analyses of safety and efficacy.
|No. of men/women||20/6|
|Median age (range), y||60 (22-83)|
|Median Karnofsky performance status (range)||80 (70-90)|
|Previous chemotherapy for R/M disease, no. of patients|
|No prior cytotoxic regimen for R/M disease||13|
|Received 1 previous cytotoxic regimen for R/M disease||13|
|Received any previous prior radiotherapy||26|
|Received concurrent chemotherapy or cetuximab||20|
|Received radiotherapy alone||6|
|Subsite of primary tumor, no. of patients|
Drug Exposure and Adverse Events
Twenty-five patients received a median of 6 treatments with gemcitabine plus pemetrexed (range, 1-24 treatments) on study. Table 3 presents cumulative safety data for all 25 treated patients. Adverse events are included in the table if they were common (ie, if they occurred in ≥30% of patients) or if at least 1 patient experienced a grade ≥3 event regardless of attribution. There were no grade 5 events.
|Adverse Event||No. of Patients (%)|
|All Events||Grade 3||Grade 4|
|Hemoglobin, low||25 (100)||6 (24)||0 (0)|
|Lymphopenia||21 (84)||16 (64)||5 (20)|
|Fatigue||19 (76)||3 (12)||1 (4)|
|ALT elevation||18 (72)||3 (12)||0 (0)|
|AST elevation||15 (60)||0 (0)||0 (0)|
|Hyponatremia||15 (60)||2 (8)||1 (4)|
|Hyperglycemia||13 (52)||1 (4)||0 (0)|
|Leukopenia, total WBC||12 (48)||2 (8)||4 (16)|
|Alkaline phosphatase, elevation||11 (44)||0 (0||0 (0)|
|Constipation||11 (44)||0 (0||0 (0)|
|Thrombocytopenia||10 (40)||1 (4)||1 (4)|
|Hyperkalemia||10 (40)||0 (0||0 (0)|
|INR elevation||8 (32)||2 (8)||0 (0)|
|Mucositis||7 (28)||1 (4)||0 (0)|
|Neutropenia||7 (28)||4 (16)||2 (8)|
|PTT elevation||7 (28)||1 (4)||0 (0)|
|Hypokalemia||6 (24)||0 (0||2 (8)|
|Hypophosphatemia||5 (20)||3 (12)||0 (0)|
|Venous thrombosis/embolism||3 (12)||0 (0||2 (8)|
|Dyspnea||2 (8)||2 (8)||0 (0)|
|Febrile neutropenia||2 (8)||2 (8)||0 (0)|
|Hypotension||2 (8)||1 (4)||0 (0)|
|Dehydration||1 (4)||1 (4)||0 (0)|
|Pneumonia||1 (4)||1 (4)||0 (0)|
|Upper respiratory symptoms||1 (4)||1 (4)||0 (0)|
Generally, the treatment was well tolerated. However, 2 patients were removed from study because of toxicity during the first cycle. A man aged 74 years with metastatic oropharynx cancer who was eligible for the study in the context of mild baseline renal insufficiency (creatinine clearance, 48 mL per minute on Cycle 1/Day 1) experienced grade 4 neutropenia, grade 4 thrombocytopenia, grade 3 mucositis, and grade 3 pneumonia that required hospitalization after his first treatment with gemcitabine and pemetrexed. Because >28 days elapsed before the resolution of these adverse events, the patient was removed from study, as required by the protocol. A CT scan of the chest after his first treatment revealed evidence of slight disease progression (4% increase; stable disease according to RECIST measurements). A man aged 83 years with metastatic oropharyngeal cancer elected to withdraw from the study after he experienced grade 3 fatigue during Cycle 1 despite a dose reduction for the second treatment. A CT scan of the chest after Cycle 1 revealed a minor response in lung metastasis (24% reduction; stable disease according to RECIST measurements).
The most common toxicities that we believed were possibly related to study drugs were low hemoglobin, lymphopenia, and fatigue. Low hemoglobin is a common finding in patients with advanced head and neck cancer but also may be treatment-related. Neutropenia (grade ≥3) occurred in 24% of patients, and 2 patients experienced febrile neutropenia. Two patients experienced pulmonary embolism. Hyperglycemia, probably because of steroid premedications rather than study drugs, was common but was clinically significant (grade 3) in only 1 patient.
Seven patients (28% of the study population) required at least 1 dose reduction because of toxicity on this study. Two patients (8% of the study population) required a second dose reduction. The most common reason for dose reduction in this study was neutropenia/febrile neutropenia, which accounted for 3 of the 9 dose reductions in the study.
The objective response rate was 16% according to RECIST, and 4 of 25 patients achieved a partial response. There were no complete responses. For 18 patients, the best response according to RECIST was stable disease. For 7 patients, the duration of stable disease was ≥120 days (range, 120-397 days). Two patients had disease progression at the first radiographic response assessment. One patient was lost to follow-up after Cycle 1 and did not undergo radiology studies for response assessment; however, that patient was included in the current analysis according to the intention-to-treat principle.
The median overall survival was 8.8 months (range, from 1.9 months to ≥30.9 months), and the median follow-up duration was 22 months (Fig. 1). Figure 2 presents a waterfall plot that illustrates the best response of all evaluable patients on study. In addition to the 4 patients who had a partial response (≥30% reduction in the sum of the greatest tumor dimensions), 5 additional patients had reductions >20% in the sum the greatest tumor dimensions that did not meet RECIST parameters for a partial response as defined in the protocol. In total, 9 of 25 patients (36%) on this study experienced tumor reductions >20% as their best response.
The current phase 2 study establishes that the combination of pemetrexed plus gemcitabine administered every 2 weeks has activity against advanced HNSCC. The most common grade ≥3 toxicities were anemia, lymphopenia, and neutropenia. Twenty-eight percent of patients required dose reductions for treatment-related toxicities. The partial response rate according to RECIST was 16%, and the median overall survival was 8.8 months. According to the statistical plan of the protocol, this regimen is not considered worthy of further study, because only 4 objective responses were attained. There was a wide range in the length of overall survival, suggesting heterogeneous characteristics within the study population.
Table 4 summarizes phase 2 efficacy data from the current study of gemcitabine plus pemetrexed, the pemetrexed monotherapy study by Pivot et al,6 and an ongoing study by Feinstein et al regarding pemetrexed plus bevacizumab for recurrent/metastatic disease.16 Because of inherent variability in baseline characteristics among patients who are treated on separate phase 2 studies, a comparative analysis of the efficacies of the regimens used in these studies is not possible. Half of the patients in the current study had received previous cytotoxic chemotherapy for recurrent/metastatic disease; whereas, in the studies by Pivot et al and Feinstein et al, patients were excluded if they had received any previous chemotherapy for recurrent or metastatic disease. It is noteworthy that most patients in the current study experienced at least minor tumor regression (Fig. 2).
|Study||Regimen||No. of Patients||Median No. of Prior Regimens for R/M Disease||Response Rate, %|
|Pivot 20016||Pemetrexed monotherapy||35||0||26.5|
|Feinstein 200816,a||Pemetrexed plus bevacizumab||22a||0||36a|
|Current study||Pemetrexed plus gemcitabine||25||0.5||16|
Unlike the study of pemetrexed in advanced HNSCC by Pivot et al,6 the current study required vitamin B12 and folate supplementation to reduce the risk of pemetrexed-related toxicities. In a study of malignant mesothelioma, vitamin B12 and folate supplementation reduced the incidence of neutropenia, thrombocytopenia, diarrhea, mucositis, and drug-related death.17, 18 A subsequent randomized trial in nonsmall cell lung cancer confirmed that vitamin supplementation reduces pemetrexed-related adverse events without compromising efficacy.18, 19 In the current study, the incidence of grade ≥3 neutropenia appeared to be lower than that observed in the study by Pivot et al, although the chemotherapy regimen in the current study was more intensive. These safety results indicate that vitamin supplementation should be included in any future studies of pemetrexed in HNSCC.
Patients with recurrent or metastatic HNSCC typically have received cisplatin-based treatment as part of their initial combined-modality therapy. Cisplatin-based doublets remain the widely accepted standard of care for recurrent or metastatic head and neck cancer.4 In this patient population, which generally is treated with palliative intent, additional exposure to cisplatin can be problematic because of the cumulative nature of toxicities with cisplatin, including potential nephrotoxicity, neurotoxicity, and ototoxicity. The safety profile of pemetrexed provides rationale for further study of other pemetrexed-based doublets as palliative therapy for patients with advanced HNSCC.
The efficacy of pemetrexed against HNSCC appears to highlight a distinction between this disease and squamous cell lung cancer. A phase 3 study compared pemetrexed plus cisplatin versus gemcitabine plus cisplatin for patients with advanced nonsmall cell lung cancer.20 Among patients who had nonsquamous lung cancer (adenocarcinoma or large cell carcinoma), survival was superior in the pemetrexed/cisplatin group (P = .005). Among patients who had squamous lung cancer, survival was better in the gemcitabine/cisplatin arm (P = .05). The authors suggested that pemetrexed-based chemotherapy may be a preferred option for patients with nonsquamous lung cancer. Despite this concern regarding the use of pemetrexed for squamous cell lung cancer, the current study and other studies in advanced HNSCC6, 16 indicate that pemetrexed has significant activity against squamous cell carcinomas of the upper aerodigestive tract.
The current results do not exclude the possibility that other schedules of gemcitabine and pemetrexed may be superior to pemetrexed alone. In a randomized phase 2 trial for patients with advanced nonsmall cell lung cancer, the optimal schedule was pemetrexed followed by gemcitabine on Day 1 and gemcitabine on Day 8 of a 21-day cycle.21 Future studies might determine whether the efficacy of the combination of gemcitabine and pemetrexed in advanced HNSCC can be enhanced by alternate schedules, perhaps with weekly gemcitabine.
In summary, the regimen of pemetrexed plus gemcitabine on an every-2-week cycle administered with vitamin B12 and folate supplementation generally was well tolerated and had modest palliative activity against recurrent or metastatic HNSCC. The most clinically significant toxicity was myelosuppression. The results of this study provide further evidence that pemetrexed has significant palliative activity against advanced HNSCC.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 10Synergistic cytotoxicity of MTA (LY231514) and gemcitabine in vitro and in vivo [abstract]. Paper presented at: 10th NCI-EORTC Symposium on New Drugs in Cancer Therapy; June 16-19, 1998; Amsterdam, the Netherlands. Abstract 644., ,
- 11Phase I and pharmacologic study of sequences of gemcitabine and the multitargeted antifolate agent in patients with advanced solid tumors. J Clin Oncol. 2000; 18: 11748-11757., , , et al.
- 15New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000; 92: 205-216., , , et al.
- 16Pemetrexed (P) and bevacizumab (B) in patients (pts) with recurrent or metastatic head and neck squamous cell carcinoma (HNSCC): updated results of a phase II trial—ASCO [abstract]. J Clin Oncol. 2008; 26( May 20 suppl). Abstract 6069., , , et al.
- 17Vitamin B12 and folate reduce toxicity of Alimta (pemetrexed disodium, LY231514, MTA), a novel antifolate/antimetabolite [abstract]. Proc Am Soc Clin Oncol. 2001; 20: 76a. Abstract 300., , , et al.