Update on Gemcitabine/Carboplatin in Patients With AdvancedNon–Small Cell Lung Cancer

Peter Harper

Platinum-based chemotherapy regimens comprise astandard treatment approach for patients with ad-vanced and metastatic non–small cell lung cancer(NSCLC). Based on results from randomized studiesand meta-analyses, it has been established that suchtherapy significantly improves survival and maintainsor improves quality of life relative to best supportivecare. Combinations of cisplatin or carboplatin withgemcitabine, a newer-generation nucleoside antime-tabolite with single-agent activity of 20% to 26% inadvanced NSCLC, have shown antitumor activity andare well tolerated. In many studies in the advanced-disease setting, carboplatin has replaced cisplatin be-cause of its improved nonhematologic toxicity profileand greater ease of administration. Encouraging re-sults in the phase II setting have led to the design andimplementation of several phase III studies of gemcit-abine/carboplatin in the treatment of patients with ad-vanced NSCLC. Results of three phase III trials involv-ing more than 900 patients not previously treated withchemotherapy are discussed herein. These studiescompared gemcitabine/carboplatin versus gemcitabinealone, gemcitabine/carboplatin versus gemcitabine/cis-platin, and gemcitabine/carboplatin versus mitomycin/ifosfamide/cisplatin (MIP), a regimen commonly usedin Europe. Results show that gemcitabine/carboplatinefficacy was equivalent or superior to that achievedwith single-agent gemcitabine or other platinum-basedtreatments. The regimen was well tolerated overall,and available data from one study show a significantimprovement in quality of life. Thus, gemcitabine/car-boplatin appears to be a viable option in the first-linetreatment of advanced NSCLC. The results of onestudy reviewed suggest that gemcitabine/carboplatincan be considered for the treatment of patients over 70years old.Semin Oncol 30 (suppl 10):2-12. © 2003 Elsevier Inc. Allrights reserved.

LUNG CANCER is the leading cause of can-cer-related mortality among both men and

women in the United States, and accounted forapproximately 28% of all cancer deaths in 2002.1Approximately 170,000 new cases were estimated

for 2002, the majority of which (�85%) wereexpected to be of non–small cell histology.2 Morethan one half of non–small cell lung cancer(NSCLC) patients initially present with advancedor metastatic disease not amenable to surgery.These patients require chemotherapy either aloneor, if eligible, in combination with radiotherapy.

Lung cancer is primarily a disease of the elderly.More than one half of new diagnoses and morethan two thirds of annual deaths occur in peopleover 65 years old,3,4 in whom concomitant diseasesor poor performance status (PS) may complicatethe ability to tolerate chemotherapy. The avail-ability of new active drugs has spurred investiga-tions of novel combination chemotherapy regi-mens, with the goals of improving outcomes whileminimizing toxicities and improving quality of life(QOL) in patients with advanced and metastaticNSCLC. Data on the gemcitabine/carboplatin reg-imen that has been widely used in the treatment ofadvanced NSCLC are reviewed herein. Recentresults of three phase III studies in patients withadvanced and metastatic NSCLC are also pre-sented.

PLATINUM-BASED COMBINATIONS FORADVANCED NON–SMALL CELL LUNG

CANCER

Platinum-based therapy has become a standardtreatment approach for patients with advancedand metastatic NSCLC. Significant improvementin 1-year survival and reduction in risk of deathwere shown in meta-analyses and randomized tri-als of platinum therapy versus best supportive carein NSCLC.5-7 Symptom palliation and QOL werealso improved in patients receiving platinum-based therapy. Recent results of the Big Lung Trialcomparing cisplatin-based therapy and best sup-portive care in more than 700 patients with ad-vanced NSCLC also showed that treatment ben-efits were independent of disease stage (III or IV)and extended to patients with a World HealthOrganization PS of 2.6

Several new agents with single-agent activity(objective response rate �15%) against NSCLChave become available, including oxaliplatin,

From the Guy’s & St. Thomas Hospital, London, United King-dom.

Address reprint requests to Peter Harper, MB, FRCP, Guy’s &St. Thomas’ Hospital, Medical Oncology, 3rd floor, Thomas GuyHouse, London SEI 9RT, UK.

© 2003 Elsevier Inc. All rights reserved.0093-7754/03/3004-1002$30.00/0doi:10.1016/S0093-7754(03)00279-3

2 Seminars in Oncology, Vol 30, No 4, Suppl 10 (August), 2003: pp 2-12

pemetrexed, vinorelbine, paclitaxel, docetaxel, iri-notecan, and the nucleoside antimetabolite gem-citabine.8 In studies of platinum combinationsthat included a new agent, results were better thanthose achieved with platinum alone or combinedwith earlier-generation agents such as vindesine oretoposide.9-11 Among these new agents, gemcita-bine has generated considerable interest inNSCLC based on single-agent response rates of20% to 26% and median survivals of 8 to 9 monthsin phase II trials involving more than 400 pa-tients.12-14 Tolerability was good, with mild, re-versible myelosuppression as the primary toxicity.

Gemcitabine plus best supportive care providedsignificant and sustained improvements in QOLand disease-related symptoms over those achievedwith best supportive care alone in a large random-ized trial in advanced NSCLC.15 Furthermore, sin-gle-agent gemcitabine was associated with re-sponse and survival rates similar to those ofcisplatin/etoposide doublet therapy in two smallrandomized studies in stages III and IVNSCLC.16,17 Elderly patients with advanced dis-ease also benefited from single-agent gemcitabinein a retrospective analysis of phase II studies.18

Tolerability and activity in patients � 65 years old(24% response rate; 9.1 month median survival;36% 1-year survival) were similar to those in theyounger cohort (� 65 years old).

Combinations of gemcitabine with cisplatin orcarboplatin have proven to be among the mostactive palliative treatments for advanced NSCLC.Preclinical data showed synergistic activity be-tween gemcitabine and platinum agents inNSCLC cell lines,19,20 which may be because ofthe ability of gemcitabine to inhibit the repair ofplatinum-induced DNA damage.20,21 Many of theclinical trials of gemcitabine/platinum used cispla-tin, based in part on a Southwest Oncology Groupanalysis showing cisplatin as an independent pre-dictor of improved outcome in extensive-stageNSCLC. (Among the 904 patients from recenttrials included in the Cox model, nearly all hadreceived cisplatin-based therapy.)22 Response ratesin phase II trials ranged from 30% to 54%, withmedian survival times of approximately 8 to 15months and 1-year survival rates from 34% to61%.10,23-29 Myelosuppression was the major tox-icity, while cisplatin-associated nonhematologicgastrointestinal, renal, and neurologic effects werealso problematic. Attempts to reduce the toxicity

associated with the gemcitabine/cisplatin regimenhave included replacing cisplatin with carboplatin,a platinum analog possessing similar efficacy andhematologic effects, but lacking the nonhemato-logic toxicities commonly experienced with cispla-tin therapy.30-32 Furthermore, carboplatin can beadministered on an outpatient basis with no needfor prehydration to avoid renal toxicity.

CLINICAL TRIAL OFGEMCITABINE/CARBOPLATIN IN

ADVANCED NON–SMALL CELL LUNGCANCER

Table 1 shows phase II and III clinical trials ofthe gemcitabine/carboplatin regimen in patientswith stages IIIB and IV NSCLC.33-47 In phase IIstudies, carboplatin was administered at an areaunder the concentration-time curve (AUC) of 4to 6 on day 1 or 2 or 8 with gemcitabine doses of1,000 to 1,250 mg/m2 given in most cases eitheron days 1, 8, and 15 of each 28-day cycle, or ondays 1 and 8 of each 21-day cycle. (In the phase IIstudy reported by Iaffaioli et al,41 gemcitabine wasadministered on days 1 and 8 of a 28-day cycle.)Objective response rates ranged from 25% to 59%,with median survivals of 10 to 16 months. In earlytrials of the 28-day regimen, severe hematologiceffects, particularly neutropenia and thrombocyto-penia, led to premature termination of two stud-ies.34,35 A modified administration schedule for thegemcitabine/carboplatin regimen was studied inattempts to reduce hematologic toxicities whilemaintaining response and survival rates.

In a phase II trial, Carrato et al48 compared21-day and 28-day schedules of the gemcitabine/carboplatin regimen in 75 advanced NSCLC pa-tients. Patients received gemcitabine 1,000 mg/m2

on days 1, 8, and 15 with carboplatin AUC � 5 onday 1 every 28 days (n � 32), or a modifiedregimen omitting the day-15 gemcitabine dose andrepeating cycles every 21 days (n � 43). Inci-dences of grades 3 and 4 neutropenia were 52%and 38%, respectively, with the 28-day regimen,and decreased to 39% and 23% with the modifiedregimen. Significant reductions in rates of grades 3and 4 thrombocytopenia were noted, from 45%and 61%, respectively, in the 28-day cohort to24% (P � .04) and 17% (P � .002), respectively,in the 21-day group. Response rate (46% and 37%,respectively) and median survival time (38 weeksfor both schedules) were maintained in patients

PETER HARPER 3

receiving the 21-day schedule. Furthermore,higher dose intensities of both gemcitabine andcarboplatin were achieved with every-3-week ver-sus every-4-week schedules (gemcitabine, 1,133 v1,002 mg/wk; carboplatin, 162 v 124 mg/wk).34

It should be noted that, despite high rates ofgrade 3/4 thrombocytopenia and neutropenia, fewserious clinical consequences were seen in studiesof gemcitabine/carboplatin. Hematologic nadirs

were short lasting and the majority of affectedpatients recovered without having serious bleedingepisodes or febrile neutropenia.33,34,36,37,41 Tolera-bility of the gemcitabine/carboplatin regimen isalso notable for a lack of nonhematologic effects,such as nausea/vomiting, and of potentially non-reversible effects, such as nephrotoxicity and neu-rotoxicity, which are associated with other doubletregimens used to treat NSCLC.49,50

Table 1. Summary of Phase II and III Trials of Gemcitabine/Carboplatin in Stage IIIB and IV Non–Small Cell Lung Cancer

Study Phase

No. of

patients Regimen

OR

(%)

MST

(mo)

Neutropenia Thrombocytopenia

Gr 3 Gr 4 Gr 3 Gr 4

Every 28-day regimen

Heching33 II 34 G: 1,000 mg/m2, d 1,8,15 48 12 — — — —

C: AUC � 5

Carrato34 II* 33 G: 1,000 mg/m2, d 1, 8, 15 47 11 39% 30% 45% 52%

C: AUC � 5

Ng35 II* 7 G: 1,000 mg/m2, d 1,8,15 — — — — 1 3

C: AUC � 5

Masotti36 II 50 G: 1,000 mg/m2, d 1,8,15 50 — — 32% — 56%

C: AUC � 5

Jovtis37 II 58 G: 1,000 mg/m2, d 1,8,15 43 11 12% of cycles 4% of cycles 11% of cycles 7% of cycles

C: AUC � 5

Carmichael38 I-II 13 G: 1,000 mg/m2, d 1,8,15 31 11 7 2 1 2

C: AUC � 5.2

Parente39 III 34 G: 1,000 mg/m2, d 1,8,15 56 11 3.5% 9% — —

C: AUC � 6

Danson40 III 108 G: 1,000 mg/m2, d 1,8,15 32 — — — — —

C: AUC � 5

Iaffaioli41 I-II 38 G: 1,100 mg/m2 d 1,8 50 16 3 1 1 0

C: AUC � 5

Every 21-day regimen

Carrato34 II 56 G: 1,000 mg/m2 d 1,8 48 13 53% 27% 36% 21%

C: AUC � 5

Sederholm42 II 10 G: 1,200 mg/m2 d 1,8 44 — 6% of infusions 0 4% of infusions 0

C: AUC � 5

Edelman43† II 37 G: 1,000 mg/m2 d 1,8 25 10 — 21% — 19%

C: AUC � 5.5

Sederholm42 II 25 G: 1,250 mg/m2 d 1,8 45 — 0 4% of infusions 0 7% of infusions

C: AUC � 6

Domine44 II 48 G: 1,250 mg/m2 d 1,8 59 14 29% (gr 3�4) — 19% (gr 3�4) —

C: AUC � 6

Mazzanti45 III 53 G: 1,200 mg/m2 d 1,8 26 10 6% — 9% —

C: AUC � 5

Murad46 II 23 G: 1,000 mg/m2 d 1,8 39 11 19% (gr 3�4) — 16% (gr 3�4) —

C: AUC � 5

Maestu47 II 79

(�65 yrs old)

G: 1,250 mg/m2 d 1,8 39.2 9.9 — — — —

C: AUC � 4

Abbreviations: C, carboplatin; d, days of gemcitabine administration; G, gemcitabine; OR, objective response rate; MST, median survival time.*Study terminated due to toxicity.†Three courses carboplatin/gemcitabine followed by three courses paclitaxel.Modified and reprinted from Thomas et al55 with permission.

4 GEMCITABINE/CARBOPLATIN IN NSCLC

RECENT REPORTS OF PHASE IIIGEMCITABINE/CARBOPLATIN TRIALS IN

ADVANCED NON–SMALL CELL LUNGCANCER

Several phase III studies designed to help definethe role of gemcitabine/carboplatin in patientswith advanced NSCLC are ongoing or were re-cently completed. Some of the trials include de-tailed QOL analyses, which have been lacking todate and are particularly important in palliativetreatment of advanced disease. Three phase IIItrials reported at the 2002 annual meeting of theAmerican Society of Clinical Oncology (Orlando,FL) included randomized comparisons of gemcit-abine/carboplatin versus gemcitabine alone,51 ver-sus the gemcitabine/cisplatin doublet,52 and versustriple therapy with mitomycin/ifosfamide/cisplatin(MIP) (Table 2).53 In each trial, gemcitabine/carboplatin was associated with equivalent or su-perior effects relative to the comparator regimens,further confirming the utility of this combinationin patients with advanced and metastatic NSCLC.

Gemcitabine/Carboplatin Versus Gemcitabine

Sederholm51 reported results from the first largerandomized study of the gemcitabine/carboplatin

combination in advanced NSCLC, which wasconducted by The Swedish Lung Cancer StudyGroup. A total of 334 patients who had not pre-viously received chemotherapy were recruitedfrom 17 centers in Sweden. Patients were ran-domly assigned to receive gemcitabine (1,250 mg/m2, days 1 and 8 intravenously [IV]) or gemcitab-ine (1,250 mg/m2, days 1 and 8 IV) pluscarboplatin (AUC � 5, day 1, IV), with cyclesrepeated every 21 days for a maximum of six cy-cles. Dose adjustments were based on hematologictoxicities.

At baseline, treatment groups were fairly wellbalanced with respect to patient age (median age,67 years; 63% of patients � 70 years), gender(56% male), tumor histology (49% adenocarci-noma, 23% squamous), disease stage (41% stageIIIB, 59% stage IV), and PS (86% World HealthOrganization PS 0–1, 14% PS 2). A substantialproportion of patients (� 37%) were � 70 years ofage. A mean of 4.3 (median, 5) versus 4.4 (me-dian, 6) treatment courses were administered inthe gemcitabine versus gemcitabine/carboplatingroups. Dose intensities achieved were 1.0 (range,0.5 to 1.0) for gemcitabine alone and 0.9 (range,0.5 to 1.0) for both gemcitabine and carboplatin in

Table 2. Selected Phase III Trials of Gemcitabine/Carboplatin in Patients With Advanced Non–Small Cell Lung Cancer

StudyNo. ofPatients Regimen (Every 21 Days) OR (%)

MST(mo)

1-Yr Survival(%) Notes

Sederholm51 334 G alone: 1,250 mg/m2, d 1,8 11.5% 9 32% Treatment benefits extended to elderly studypopulation (37% of patients � 70 years old).Increase in grade 3/4 thrombocytopenia inGC patients, but without significantdifferences in bleeding episodes

G: 1,250 mg/m2, d 1,8 29.6% 10 41%C: AUC � 5, dl P � .0001 P � .0157

Zatloukal52 176 G: 1,200 mg/m2, d 1,8 47%* 8.1 — Grade 3/4 nausea/vomiting more common withGCs; grade 3/4 thrombocytopenia morecommon with GC.

C: AUC � 5, d 1G: 1,200 mg/m2, d 1,8 48† 8.1 —Cis: 80 mg/m2, d 1

Rudd53 422 G: 1,200 mg/m2, d 1,8 41% 10.2 38% Significantly better QOL, fewer hospitalizations,and less nonhematologic toxicity in GCgroup. Thrombocytopenia significantly morecommon with GC, with no seriousconsequences in the majority of affectedpatients.

C: AUC � 5, d 1M: mitomycin 6 mg/m2, d 1 41% 6.9 28%I: ifosfamide 3 g/m2, d 1P: cisplatin 50 mg/m2, d 1

Abbreviations: C, carboplatin; Cis, cisplatin; d, days of administration; G, gemcitabine; OR, objective response rate; MST, median survival time;QOL, quality of life.

*14/29 evaluable patients.†16/34 evaluable patients.

PETER HARPER 5

the combination arm. Roughly one third of pa-tients in both groups received second-line treat-ments, including reinduction with the same ther-apy, docetaxel, platinum, or other agents. Nopatients in the gemcitabine/carboplatin arm re-ceived second-line platinum. Otherwise, second-line treatments were fairly evenly distributed inthe two randomized groups.

Results showed significantly higher responserate, time to disease progression, and survival im-provements in patients who received gemcitabine/carboplatin versus single-agent gemcitabine. Thisbenefit extended even to patients older than 70years of age with good PS (37% � 70 years old).Among 291 patients evaluable for response, 29.6%of 135 patients in the gemcitabine/carboplatingroup versus 11.5% of 156 in the gemcitabinegroup experienced objective responses (P �.0001). No patient in the gemcitabine-alone groupexperienced a complete response (CR), 18 hadpartial responses (PRs), 127 had stable disease, anddisease progressed in 11 patients. Among the gem-citabine/carboplatin patients, two had CRs, 38had PRs, 84 had stable disease, and disease pro-gressed in 11 patients. Respectively, intent-to-treat analyses showed significantly longer mediantime to disease progression of 6 versus 4 months(P � .0009) in the combination versus single-agent group, 1-year survival rates of 41% (n �

137) versus 32% (n � 152) (P � .0157), and2-year survival rates of 16% versus 5% (P � .001)(Fig 1). Overall median survival times of 9 and 10months in the gemcitabine and combinationgroup, respectively, were encouraging given thelarge number of elderly patients in the study.

Both regimens were well tolerated. Nonhema-tologic side effects occurred with similar frequencyin the two groups, and consisted primarily of mild-to-moderate gastrointestinal or pulmonary effects.Grade 3/4 nonhematologic effects were reported in26.1% of gemcitabine/carboplatin patients and28.4% of gemcitabine patients; the most commonnonhematologic event was pulmonary toxicity,which occurred in 21 and 19 patients in eachgroup, respectively. As expected, hematologic tox-icity was more frequent in the combination arm(Fig 2). Grade 3 thrombocytopenia was reported in24.3% and grade 4 in 24.3% of the gemcitabine/carboplatin patients. This effect generally occurredaround day 15 and was of short duration. However,only 6% of these patients received platelet trans-fusions and no major bleeding episodes were re-ported. Analyses of QOL data collected in thistrial were not completed at the time of this report.

Thus, gemcitabine/carboplatin was associatedwith a better response rate than single-agent gem-citabine. While the combination caused more

Fig 1. Overall survival in 289 patients (intent-to-treat analysis) in phase III comparison of gemcitabine plus carboplatin (Gem �

Carbo) versus gemcitabine alone (Gem) in advanced NSCLC.

6 GEMCITABINE/CARBOPLATIN IN NSCLC

grade 3/4 myelosuppression, this effect resolved inmost patients without any major clinical sequelae.

Gemcitabine/Carboplatin VersusGemcitabine/Cisplatin

Another phase III study reported at the 2002American Society of Clinical Oncology meetingdirectly compared carboplatin versus cisplatin

when combined with gemcitabine in patients withNSCLC.52 In this trial, conducted by the CzechLung Cancer Cooperative Group, 176 patientswith advanced-stage (40% stage IIIB/60% stageIV), untreated NSCLC were randomly assigned toreceive gemcitabine 1,200 mg/m2 IV on days 1 and8 plus cisplatin 80 mg/m2 IV 4 hours after gemcit-abine on day 1; or the same gemcitabine treatmentplus carboplatin (AUC � 5, IV), also 4 hoursfollowing the day-1 gemcitabine. Cycles were re-peated every 21 days. At baseline, median patientage was 62 years (range, 39 to 75 years), 135participants were men and 41 were women, me-dian Karnofsky PS was 90, and histologic subtypeswere adenocarcinoma in 30%, squamous cell in51%, large cell in 7%, and other in 13%. Patientsreceived a median of four treatment courses(range, 1 to 6). Delivered dose intensities werehigh for the three agents: medians, 93% ofplanned for gemcitabine, 96% cisplatin, and 98%carboplatin.

Both regimens were active and well tolerated.Toxicity data were available for 158 patients,and grades 3 and 4 effects are shown in Table 3.Nausea/vomiting was the only grade 3/4 non-hematologic effect that occurred significantlymore frequently in the gemcitabine/cisplatin arm.Thrombocytopenia incidence was significantlyhigher in the gemcitabine/carboplatin arm, whilethe number of bleeding episodes was similar, andlow, for both groups. One patient in each armdeveloped febrile neutropenia, both of whom diedof this effect. No grade 3/4 renal, hepatic, pulmo-nary, or cutaneous toxicities were noted. Prelimi-nary objective response rate was 48% among 63evaluable patients, 48% (2 CRs, 12 PRs) in gem-citabine/cisplatin (n � 29), and 47% (2 CRs, 14PRs) in gemcitabine/carboplatin (n � 34) pa-tients. With median follow-up times of 5.5 monthsand 6.8 months in the cisplatin and carboplatinarms, median times to disease progression in 147patients are 5.6 months and 6.1 months, respec-tively, and median survival is 8.1 months for bothgroups (n � 157).

Gemcitabine/Carboplatin VersusMitomycin/Ifosfamide/Cisplatin

Rudd et al53 presented efficacy, safety, and QOLresults of a phase III trial comparing gemcitabine/carboplatin with MIP, a regimen commonly usedin Europe. A total of 422 patients with stages

Fig 2. Proportions of patients with grade 3 or 4 (A) anemia,(B) leukopenia, or (C) thrombocytopenia, by treatment arm, inphase III study of gemcitabine/carboplatin (Gem � Carbo)versus gemcitabine alone (Gem) in advanced NSCLC. Six per-cent of patients in the Gem � Carbo group received proph.platelet transfusions but no major bleeding episodes werenoted.

PETER HARPER 7

IIIB/IV NSCLC who had not previously receivedradiotherapy or chemotherapy were randomly as-signed to receive gemcitabine 1,200 mg/m2 IV ondays 1 and 8 plus carboplatin (AUC � 5) IV day1 (n � 212); or mitomycin 6 mg/m2 plus ifos-famide 3 g/m2 plus cisplatin 50 mg/m2, all admin-istered IV on day 1 (n � 210). Both regimens wererepeated every 3 weeks. Baseline patient charac-teristics were balanced in the two treatmentgroups. Median patient age was 62 years (range, 34to 81 years), 70% were male, approximately 41%and 35% had squamous cell and adenocarcinomahistologies, respectively, disease stage was IIIB in47% and IV in 52%. Eastern Cooperative Oncol-ogy Group PS was 0–1 in 87% of patients.

At the time of report, approximately 90% ofpatients had completed therapy, and 64% and61% of the gemcitabine/carboplatin and MIP pa-tients, respectively, had received the intended fourtreatment courses. Hematologic toxicities were theprimary reason for treatment delays, which werenecessary in 12% of gemcitabine/carboplatin and7% of MIP courses (P � .009), and for dosemodifications performed in 19% and 14% (P �.01) of courses in the respective treatment groups.

No significant differences in anemia (72% v 64%),leukopenia (50% v 50%), or grade 2 thrombocy-topenia (4% v 6%) were noted for gemcitabine/carboplatin versus MIP, respectively. Grade 3/4thrombocytopenia occurred more often in gemcit-abine/carboplatin patients (25% v 7%; P � .0002),but did not lead to increased hospital admissionsor number of toxic deaths. Patients receiving gem-citabine/carboplatin required fewer in-patient hos-pital admissions for chemotherapy administration(14% v 89%; P � .0001), and experienced signif-icantly less grade 3/4 nonhematologic toxicities,including nausea (5% v 14%; P � .0001), vomit-ing (3% v 10%; P � .0001), constipation (2% v7%; P � .002), and alopecia (1% v 9%; P �.0001).

Response rates were identical in the gemcita-bine/carboplatin (41%) and MIP (41%) groups,whereas 1-year survival rates and median survivalswere 38% versus 28% and 10.2 versus 6.9 monthsfor gemcitabine/carboplatin and MIP, respectively,at 11.5 and 11.6 months follow-up.

This study included QOL analyses using theEuropean Organization for Research and Treat-ment of Cancer core questionnaire Quality of LifeC30 and L17 lung cancer module. These wereconducted at baseline, at each course of chemo-therapy, and at the first and second follow-upvisits. Gemcitabine/carboplatin was associatedwith significantly better QOL than MIP based onchanges from baseline to 6 weeks in several gen-eral symptom scores, functional scales, and globalQOL scores (Fig 3). Changes in lung cancer symp-tom scores for parameters such as cough, shortnessof breath, hoarse voice, sore throat, trouble swal-lowing, tingling in hands/feet, numbness in toes/fingers, weak arms/legs, and fever were similar inthe gemcitabine/carboplatin and MIP patients,whereas gemcitabine/carboplatin patients reportedbetter scores for hair loss and being upset abouthair loss (P � .0001 for both parameters). Resultsof this trial showed significantly improved survival,less toxicity, and better QOL with the gemcita-bine/carboplatin regimen than with MIP in ad-vanced NSCLC.

DISCUSSION

The gemcitabine/carboplatin combination hasbeen evaluated extensively in patients with ad-vanced and metastatic NSCLC. The regimen hasproduced high response rates and is well tolerated

Table 3. World Health Organization Grade 3/4Toxicities in Phase III Trial of

Gemcitabine/Carboplatin (GCarb) VersusGemcitabine/Cisplatin (GCis) in Patients With Stages

IIIB/IV Non–Small Cell Lung Cancer

GCis(n � 77)% Patients

GCarb(n � 81)% Patients

PValue*

Hematologic, Grades 3/4Neutropenia 18.9 31.2 .122Leukopenia 9.5 14.3 .5068Thrombocytopenia 13.5 36.4 .0023Bleeding 0 2.6 .8901Anemia 14.9 16.9 .9073

Nonhematologic, Grades 3/4Nausea/vomiting 16.2 3.8 .0224Diarrhea 2.7 0 � .05Infection 2.7 1.3 � .05Neurotoxicity 0 3.8 � .05Alopecia 10.8 7.7 � .05Cardiac toxicity 4.1 0 � .05Gastrointestinal toxicity-

other 0 2.7 � .05

*Chi-squared test.

8 GEMCITABINE/CARBOPLATIN IN NSCLC

in the majority of patients. The incidence of se-vere myelosuppression, which was high in trialsusing gemcitabine dosed on days 1, 8, and 15 every28 days, has been substantially reduced withoutloss of efficacy by eliminating the day-15 gemcit-abine dose and cycling therapy every 21 days.Furthermore, few patients with severe hematologiceffects require platelet transfusions or suffer bleed-ing complications or neutropenic fever.

Results from current phase III trials are consis-tent in showing good activity and tolerability ofthe gemcitabine/carboplatin combination in ad-vanced NSCLC, and establish this regimen as anappealing approach for treating this disease. Re-

sponse rates in the phase III trials reported herein,some of which are preliminary, ranged from 30%to 47% for patients receiving gemcitabine/carbo-platin doublet therapy.51-53 Median survival timewith gemcitabine/carboplatin was similar to thatachieved with gemcitabine/cisplatin,52 but im-proved relative to single-agent gemcitabine orMIP (see Table 2).51,53 The Swedish Lung CancerStudy Group study showed at least a doubling ofobjective response (CR � PR) rate in patientsreceiving gemcitabine/carboplatin compared withsingle-agent gemcitabine (29.6 v 11.5%; P �.0001) and significant prolongation of time to dis-ease progression (6 v 4 months; P � .0009).51 In

Fig 3. Quality-of-life (QOL) assessments based on European Organization for Research and Treatment of Cancer core question-naire QLQ C30 and L17 lung cancer module in phase III trial of gemcitabine/carboplatin (GC) versus MIP in patients with stages IIIB/IVNSCLC. A positive change indicates that the function improves. (A) Changes in general symptom scores between baseline and 6weeks (* P � .02; ** P < .008; *** P < .0001). (B) Changes in functional scales and global QOL scores between baseline and 6 weeks(* P < .006; ** P < .0001). Data from Rudd et al.53

PETER HARPER 9

addition, more than 37% of patients in this trialwere older than 70 years of age, and the highresponse rates and good tolerability observed over-all support gemcitabine/carboplatin as a feasibletreatment alternative in appropriately selected el-derly patients with advanced NSCLC.

The comparison of gemcitabine/carboplatinversus MIP conducted by Rudd et al53 corroboratesother study results showing no significant responseand survival advantages of a triple-drug regimenover standard doublet therapy in patients withadvanced disease.8,10,54 In this trial, gemcitabine/carboplatin had improved median survival and1-year survival rate and was associated with lesstoxicity than MIP. Quality of life improvementswere also more profound in gemcitabine/carbopla-tin than in MIP patients, as shown by significantdifferences in changes from baseline to 6 weeks inseveral measured parameters.

CONCLUSION

The data reported herein support that gemcit-abine/carboplatin is an effective, well-toleratedregimen for outpatient treatment of patients withadvanced and metastatic NSCLC. Moreover, gem-citabine in combination with carboplatin can beconsidered a viable first-line approach for selectpatients.

REFERENCES

American Cancer Society: Cancer Facts and Figures 2002.Atlanta, GA, American Cancer Society, 2002, p 11

2. Page N, Read W, Tierney M, et al: The epidemiology ofsmall cell lung cancer. Proc Am Soc Clin Oncol 21:305a, 2002(abstr 1216)

3. Gridelli C, Perrone F, Monfardini S: Lung cancer in theelderly. Eur J Cancer 33:2313-2314, 1997

4. Havlik RJ, Yancik R, Long S, et al: The NationalCancer Institute on Aging and the National Cancer InstituteSEER Collaborative study on comorbidity and early diagnosisof cancer in the elderly. Proc Am Soc Clin Oncol 20:322a,2001 (abstr 1285)

5. Non–Small Cell Lung Cancer Collaborative Group: Che-motherapy in non–small cell lung cancer: A meta-analysisusing updated data on individual patients from 52 randomizedclinical trials. BMJ 311:899-909, 1995

6. Stephens R, Fairlamb D, Gower N, et al: The Big LungTrial (BLT): Determining the value of cisplatin-based chemo-therapy for all patients with NSCLC. Preliminary results in thesupportive care setting. Proc Am Soc Clin Oncol 21:291a,2002 (abstr 1161)

7. Cullen MH, Billingham LJ, Woodroffe CM, et al: Mit-omycin, ifosfamide, and cisplatin in unresectable non–smallcell lung cancer: A randomized trial. Ann Oncol 17:3188-3194, 1999

8. Crino L, Cappuzzo F: Present and future treatment ofadvanced non–small cell lung cancer. Semin Oncol 29:9-16,2002 (suppl 9)

9. Cardenal F, Lopez-Cabrerizo M, Anton A, et al: Ran-domized phase III study of gemcitabine-cisplatin versus etopo-side-cisplatin in the treatment of locally advanced or meta-static non–small cell lung cancer. J Clin Oncol 17:12-18, 1999

10. Crino L, Scagliotti GV, Ricci S, et al: Gemcitabine andcisplatin versus mitomycin, ifosfamide, and cisplatin in ad-vanced non–small cell lung cancer: A randomized phase IIIstudy of the Italian Lung Cancer Project. J Clin Oncol 17:3522-3530, 1999

11. Sandler AB, Nemunaitis J, Denham C, et al: Phase IIItrial of gemcitabine plus cisplatin versus cisplatin alone inpatients with locally advanced or metastatic non–small celllung cancer. J Clin Oncol 18:122-130, 2000

12. Abratt RP, Bezwoda WR, Falkson G, et al: Efficacy andsafety profile of gemcitabine in non–small cell lung cancer: Aphase II study. J Clin Oncol 12:1535-1540, 1994

13. Anderson H, Lund B, Bach F, et al: Single-agent activityof weekly gemcitabine in advanced non–small cell lung cancer:A multicenter extended phase II study. Eur J Cancer 32A:243-248, 1996

14. Gatzemeier U, Shepherd F, LeChevalier T, et al: Activ-ity of gemcitabine in patients with non–small cell lung cancer:A multicenter, extended phase II study. Eur J Cancer 32A:243-248, 1996

15. Anderson H, Hopwood P, Stephens RJ, et al: Gemcit-abine plus best supportive care (BSC) vs BSC in inoperablenon–small cell lung cancer – A randomized trial with quality oflife as the primary outcome. Br J Cancer 83:447-453, 2000

16. Perng RP, Chen YM, Ming-Liu J, et al: Gemcitabine vsthe combination of cisplatin and etoposide in patients withinoperable non–small cell lung cancer in a phase II randomizedstudy. J Clin Oncol 15:2097-2102, 1997

17. ten Bokkel Huinink WW, Bergman B, Chemaissani A,et al: Single-agent gemcitabine: An active and better toleratedalternative to standard cisplatin-based chemotherapy in locallyadvanced or metastatic non–small cell lung cancer. Lung Can-cer 26:85-94, 1999

18. Shepherd FA, Abratt RP, Anderson H, et al: Gemcit-abine in the treatment of elderly patients with advanced non–small cell lung cancer. Semin Oncol 24:50-55, 1997 (suppl 7)

19. Edelman MJ, Quam H, Mullins B: Interactions of gem-citabine, carboplatin, and paclitaxel in molecularly definednon–small cell lung cancer cell lines. Cancer Chemother Phar-macol 48:141-144, 2001

20. Peters GJ, Bergman AM, Ruiz van Haperen VWT, et al:Interaction between cisplatin and gemcitabine in vitro and invivo. Semin Oncol 22:72-79, 1995

21. Fram RJ, Cusik PS, Marinos MG: Studies on mutagen-esis and repair induced by platinum analogs. Mutat Res 173:13-18, 1986

22. Albain KS, Crowley JJ, LeBlanc M, et al: Survival de-terminants in extensive-stage non–small cell lung cancer: TheSouthwest Oncology Group experience. J Clin Oncol 9:1618-1626, 1991

23. Sandler A, Nemunaitis J, Dehnam C, et al: Phase IIIstudy of gemcitabine plus cisplatin versus cisplatin alone inpatients with locally advanced or metastatic non–small celllung cancer. J Clin Oncol 18:122-130, 2000

10 GEMCITABINE/CARBOPLATIN IN NSCLC

24. Cardenal F, Lopez-Cabrerizo MP, Anton A, et al: Ran-domized phase III study of gemcitabine-cisplatin versus etopo-side-cisplatin in the treatment of locally advanced or meta-static non–small cell lung cancer. J Clin Oncol 17:12-18, 1999

25. Steward WP, Dunlop DJ, Dabouis G, et al: Phase I/IIstudy of gemcitabine and cisplatin in non–small cell lungcancer: Preliminary results. Semin Oncol 5:43-47, 1996 (suppl10)

26. Crino L, Scagliotti G, Marangolo M, et al: Cisplatin-gemcitabine combination in advanced non–small cell lungcancer: Preliminary results. Semin Oncol 15:297-303, 1997

27. Sandler A, Crino L, Steward WP, et al: Extended sur-vival in stage III and IV non–small cell lung cancer (NSCLC)patients treated with gemcitabine plus monthly cisplatin. Pro-ceedings of the 21st European Society of Medical Oncology.Ann Oncol 7:91a, 1996 (suppl 5)

28. Abratt RP, Bezwoda WR, Goedhals L, et al: Weeklygemcitabine with monthly cisplatin: Effective chemotherapyfor advanced non–small cell lung cancer. J Clin Oncol 15:744-749, 1997

29. Anton A, Carrato A, Gonzalez-Larriba JL, et al: Phase IIactivity of gemcitabine in combination with cisplatin in ad-vanced non–small cell lung cancer. Proc Am Soc Clin Oncol15:380a, 1996 (abstr 1134)

30. Bonomi PD, Finkelstein DM, Ruckdeschel JC, et al:Combination chemotherapy versus single agents followed bycombination chemotherapy in stage IV NSCLC: A study of theEastern Cooperative Oncology Group. J Clin Oncol 7:1602-1613, 1989

31. Green MR, Kreisman H, Lickart S, et al: Carboplatin innon–small cell lung cancers: The Cancer and Leukemia GroupB experience. Carboplatin (JM-9), in Bunn PA Jr, Canetta R,Ozols R, et al (eds): Current Perspectives and Future Direc-tions. Philadelphia, PA, Saunders, 1990, pp 301-306

32. Klastersky J, Sculier JP, Lacroix A, et al: A randomizedstudy comparing cisplatin or carboplatin with etoposide inpatients with advanced non–small cell lung cancer: EuropeanOrganization of Research and Treatment of Cancer Protocol07861. J Clin Oncol 8:1556-1562, 1989

33. Heching NI, Lotan C, Kaduri L, et al: Gemcitabine andcarboplatinum in advanced non–small cell lung carcinoma:Results of a single institution phase II study. Proc Am Soc ClinOncol 18:508a, 1999 (abstr 1959)

34. Carrato A, Garcia-Gomez J, Alberola V, et al: Carbo-platin in combination with gemcitabine in advanced non–small cell lung cancer: Comparison of trials using differentschedules. Proc Am Soc Clin Oncol 18:498a, 1999 (abstr 1922)

35. Ng EW, Sandler AB, Einhorn LH, et al: A phase II studyof carboplatin plus gemcitabine in non small cell lung cancer.(NSCLC). Proc Am Soc Clin Oncol 17:468, 1998 (abstr 1801)

36. Masotti A, Morandini G: Phase II trial of gemcitabineand carboplatin in advanced non–small cell lung cancers. ProcAm Soc Clin Oncol 18:517a, 1999 (abstr 1992)

37. Jovtis S, Brocato N, Temperley G, et al: Final results ofa phase II trial of first-line therapy with gemcitabine andcarboplatin in patients with advanced non–small cell lungcancer. Lung Cancer 29:46, 2000 (suppl 1) (abstr 150)

38. Carmichael J, Allerheiligen S, Walling J: A phase Istudy of gemcitabine and carboplatin in non–small cell lungcancer. Semin Oncol 23:55-59, 1996 (suppl 10)

39. Parente B, Barroso A, Conde S, et al: A randomizedphase III study of gemcitabine and carboplatinum versus vi-norelbine and carboplatinum in advanced non–small cell lungcancer. Lung Cancer 29:69, 2000 (suppl 1) (abstr 197)

40. Danson S, Clemons M, Middleton M, et al: A ran-domized study of gemcitabine with carboplatin (GC) versusmitomycin C, ifosfamide, and cisplatin (MIC) as first linechemotherapy in advanced non–small cell lung cancer(NSCLC). Proc Am Soc Clin Oncol 20:322a, 2001 (abstr1285)

41. Iaffaioli RV, Tortoriello A, Facchini G, et al: Phase I-IIstudy of gemcitabine and carboplatin in stage III-IV non–smallcell lung cancer. J Clin Oncol 17:921-926, 1999

42. Sederholm C: A phase II study of gemcitabine pluscarboplatin in chemonaive patients with advanced non–smallcell lung cancer. Proc Am Soc Clin Oncol 18:490a, 1999 (abstr1889)

43. Edelman MJ, Gandara DR, Lau D, et al: Sequentialchemotherapy in non–small cell lung cancer (NSCLC): Car-boplatin and gemcitabine followed by paclitaxel. Proc Am SocClin Oncol 19:525a, 2000 (abstr 2063)

44. Domine M, Casado V, Estevez LG, et al: Carboplatinand gemcitabine in advanced non–small cell lung cancer: Aneffective and well tolerated regimen. Lung Cancer 29:85a, 2000(suppl 1) (abstr 281)

45. Mazzanti P, Massacesi C, Mattioli R, et al: Gemcitabine-cisplatin (GP) vs. gemcitabine-carboplatin (GC) in advancednon–small cell lung cancer (NSCLC): A multicenter phase IIrandomized trial. Eur J Cancer 37:550, 2001 (suppl 6) (abstr172)

46. Murad A, Guimares RC, Aranao BC, et al: Phase II trialof the use of carboplatin and gemcitabine (G) in the treatmentof advanced non–small cell lung cancer. Proc Am Soc ClinOncol 21:228b, 2002 (abstr 2729)

47. Maestu IC, Torregrosa D, Llorca C, et al: Efficacy ofcarboplatin and gemcitabine in the treatment of elderly pa-tients with advanced non small cell lung cancer (ANSCLC).Proc Am Soc Clin Oncol 20:401a, 2001 (abstr 1599)

48. Carrato A, Alberola V, Massuti B, et al: Combinationof gemcitabine and carboplatin as first-line treatment innon–small cell lung cancer. Ann Oncol 9:89, 1998 (abstr431P)

49. LeChevalier T, Brisgand D, Douillard JY, et al: Random-ized study of vinorelbine and cisplatin versus vindesine andcisplatin versus vinorelbine alone in advanced non–small celllung cancer: Results of a European multicenter trial including312 patients. J Clin Oncol 12:360-367, 1994

50. Kelly K, Crowley J, Bunn PA, et al: Randomized phaseIII trial of paclitaxel plus carboplatin versus vinorelbine pluscisplatin in the treatment of patients with advanced non–smallcell lung cancer: A Southwest Oncology Group trial. J ClinOncol 19:3210-3218, 2001

51. Sederholm C: Gemcitabine compared with gemcitabineplus carboplatin in advanced NSCLC: A phase III study by theSwedish Lung Cancer Study Group. Proc Am Soc Clin Oncol21:291a, 2002 (abstr 1162)

52. Zatloukal P, Novakova L, Petruzelka L, et al: Gemcit-abine plus cisplatin (GCis) versus gemcitabine plus carboplatin(GCarb) in patients with stage IIIB and IV non–small lung

PETER HARPER 11

cancer (NSCLC): Final results of Czech Lung Cancer Coop-erative Group phase III randomized trial. Proc Am Soc ClinOncol 21:307a, 2002 (abstr 1225)

53. Rudd R, Gower N, James L, et al: Phase III randomizedcomparison of gemcitabine and carboplatin with mitomycin,ifosfamide and cisplatin in advanced NSCLC. Proc Am SocClin Oncol 21:292a, 2002 (abstr 1164)

54. Alberola V, Camps C, Provencia M, et al: Cisplatin-gemcitabine (CG) versus cisplatin-gemcitabine/vinorelbine

(CGV) versus sequential doublets of gemcitabine/vinorelbinefollowed by ifosfamide/vinorelbine (GV/IV) in advanced non–small cell lung cancer (NSCLC): Results of a Spanish LungCancer Group phase III trial (GEPC/98-02). Proc Am Soc ClinOncol 20:308a, 2001 (abstr 1229)

55. Thomas P, Robinet G, Ferri-Dessens RM, et al: Phase Itrial of gemcitabine and carboplatin in metastatic non–smallcell lung cancer: A Groupe Francais de Pneumo-Cancerologiestudy. Lung Cancer 36:191-198, 2002

12 GEMCITABINE/CARBOPLATIN IN NSCLC