- Email: [email protected]
Randomized, multicenter, phase II study of gemcitabine plus cisplatin versus gemcitabine plus carboplatin in patients with advanced non-small cell lung cancer
Lung Cancer (2003) 41, 81 /89
www.elsevier.com/locate/lungcan
Randomized, multicenter, phase II study of gemcitabine plus cisplatin versus gemcitabine plus carboplatin in patients with advanced non-small cell lung cancer Paola Mazzantia,*, Cristian Massacesia, Marco B.L. Rocchib, Rodolfo Mattiolic, Paolo Lippec, Raffaele Trivisonned, Franco Buzzie, Giorgio De Signoribusf, Guido Tuverig, Giorgio Rossih, Liberato Di Lulloi, Fabio Sturbaj, Donatella Moraled, Sergio Catanzanie, Alberta Pilonea, Maurizio Bonsignoria, Tullio Battellik a
Department of Medical Oncology, Azienda Ospedaliera Umberto I, via Conca, 60122 Torrette di Ancona, Italy b Biomathematics Institute, University of Urbino, Loc. Crocicchia, 61029 Urbino, Italy c Medical Oncology Unit of Fano, Ospedale S. Croce, via Veneto 2, 61032 Fano, Italy d Medical Oncology Unit of Ascoli Piceno, Ospedale Mazzoni, viale delle Rimembranze, 63100 Ascoli Piceno, Italy e Medical Oncology, Ospedale S. Maria, via Tristano di Joannuccio 1, 05100 Terni, Italy f Medical Oncology Unit, Ospedale Madonna del Soccorso, via Pellico, 63039 San Benedetto, Italy g Medical Oncology, Azienda Ospedaliera Trieste, via Pieta` 19, 34100 Trieste, Italy h Medical Oncology Unit, Ospedale Murri, via Gramsci, 60035 Jesi, Italy i Medical Oncology Unit, ASL Pentria 2, 86170 Isernia, Italy j Medical Oncology Unit A.O., Contrada S. Domenico, 62012 Civitanova, Italy k Adria Medica Study Group, Italy Received 12 December 2002; received in revised form 26 February 2003; accepted 28 February 2003
KEYWORDS Carboplatin; Cisplatin; Gemcitabine; Non-small cell lung cancer; Phase II; Randomized
Summary Background: We conducted a phase II randomized study to assess the efficacy, with response as the primary endpoint, and the toxicity of gemcitabine/ cisplatin (GP) and gemcitabine/carboplatin (GC) in patients with advanced non-small cell lung cancer (NSCLC). Methods: Patients were randomized to GP (gemcitabine 1200 mg/m2, days 1 and 8 plus cisplatin 80 mg/m2 day 2) or GC (gemcitabine 1200 mg/ m2, days 1 and 8 plus carboplatin AUC /5 day 2). Cycles were repeated every 3 weeks. Results: Sixty-two patients were randomized to GP and 58 to GC. A total of 533 cycles were delivered (264 GP, 269 GC), with a median of four cycles/patient. The objective response rate was 41.9% (95% C.I., 29.6 /54.2%) for GP and 31.0% (95% C.I.,
*Corresponding author. Tel.: /39-071-5964-270; fax: /39071-5964-269. E-mail address: [email protected] (P. Mazzanti). 0169-5002/03/$ - see front matter – 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0169-5002(03)00140-5
82
P. Mazzanti et al.
18.2 /42.8%) for GC (P /0.29). No significant differences between arms were observed in median survival (10.4 months GP, 10.8 months GC) and median time to progression (5.4 months GP, 5.1 months GC). Both regimens were very well tolerated with no statistical differences between arms in grade 3/4 toxicities. When all toxicity grades were combined, emesis, neuropathy and renal toxicity occurred more frequently on the GP arm (P B/0.005). Conclusions: GC arm did not provide a significant difference in response rate compared with GP arm, with better overall tolerability. Carboplatin could be a valid alternative to cisplatin in the palliative setting. – 2003 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction Platinum compounds remain the cornerstone of chemotherapy for advanced NSCLC. A number of randomized trials and meta-analyses have reported that cisplatin-based chemotherapy produces modest but significantly improved survival with better symptom control and quality of life compared to that of supportive care alone [1 /3]. Gemcitabine (2?,2?-difluorodeoxycytidine) is a nucleoside antimetabolite that has confirmed activity against several solid tumors, including NSCLC [4]. The combination of cisplatin and gemcitabine has demonstrated synergism in preclinical studies, which is likely related to the interaction of both drugs on the inhibition of DNA repair [5]. Many phase II studies have demonstrated excellent clinical activity of the combination, although most have been characterized by significant hematologic toxicity [6 /10]. This toxicity, namely thrombocytopenia, has often been attributed to the 4-week schedule [6,8,11]. More recently, the combination of cisplatin and gemcitabine fared well when compared with cisplatin-based chemotherapy using older chemotherapy (such as vindesine, mitomcycin, etoposide, ifosfamide) and, in particular, those using third-generation chemotherapy (such as gemcitabine, vinorelbine, taxanes, irinotecan) in large phase III trials [12 / 15]. Carboplatin, an analog of cisplatin, administered alone or in combination therapy, has lower nephroand neurotoxicity compared with cisplatin, has been proven as effective as cisplatin in NSCLC [16] and unlike cisplatin, offers the possibility of ambulatory administration. Carboplatin has been used extensively in combination with paclitaxel, especially in the US, producing a similar efficacy outcome but a different toxicity profile compared
with other regimens [14,15]. Indeed, recently, Rosell et al. [17] making a direct comparison between paclitaxel/cisplatin and paclitaxel/carboplatin, obtained a significantly longer survival in the cisplatin arm. Few studies have been conducted, however, using the combination of carboplatin and gemcitabine. Phase I/II trials have explored both 3- and 4-week schedules, but these studies have often produced contradictory results in terms of recommended doses and toxicity profiles [18 /21]. No differences between the 3and 4-week schedules in activity were observed, although less thrombocytopenia was detected with the 3-week regimen [22,23]. Having said that, we can consider gemcitabine/ carboplatin (GC) an attractive alternative to gemcitabine/cisplatin (GP) in the treatment of NSCLC. Therefore, we decided to evaluate the efficacy, with response as the primary endpoint and the toxicity of GP versus GC, administered as 3-week schedules to patients with advanced NSCLC, in a multicenter, randomized trial.
2. Methods 2.1. Patient selection Patients had to meet the following inclusion criteria to be enrolled: histologically or cytologically confirmed stage IIIB or IV NSCLC (according to the American Joint Committee on Cancer staging system, 1992); at least one measurable lesion; age 18 /75 years; no prior chemotherapy or immunotherapy; performance status (PS) of 0 /2 on the Eastern Cooperative Oncology Group (ECOG) scale; life expectancy /12 weeks; adequate bone marrow reserve (white blood cell (WBC) count ]/4000/ mm3, neutrophil count ]/2000/mm3, platelet count ]/100,000/mm3 and hemoglobin ]/10 g/
Immunohistochemical expression of integrins and ECM proteins in NSCLC
dl); renal and hepatic function within the normal range (bilirubin B/1.5 mg/dl, transaminase levels less than three times the upper limit of normal and serum creatinine 5/1.4 mg/dl); and normal cardiac function. Patients who had received previous radiotherapy were included if their assessable disease was outside of the radiation field. The main exclusion criteria were: symptomatic central nervous system metastases, second primary malignancy and serious systemic disorders. Written informed consent was obtained from each patient before enrollment. The study protocol was in accordance with the ethical standards of Human Investigation Committee of Ancona, the coordinating center.
2.2. Study design This was a phase II, prospective, randomized, multicenter trial. Eligible patients were randomized to one of two arms, GP or GC, using a concealed list of random numbers. The randomization algorithm, based on the Pocock and Simon method [24], included ECOG PS (0 /1 vs. 2) and disease stage (IIIB vs. IV) as stratification factors. The trial was planned as a randomized phase II study to obtain some useful information for further development in a controlled randomized phase III setting. Thus, the findings obtained from the treatment-arm comparisons of this phase II study should be considered exploratory. The primary endpoint was response rate. Assuming a response rate of 50% in the GP arm [6,10,25] and an expected response rate of 25% in the GC arm, a sample size of 63 patients in each arm was required to detect a difference of 25%. This sample size was determined using a statistical power (1-b) of 80% and a significance level (a ) of 5%. Two sided testing was applied. Secondary endpoints included duration of response, toxicity, time to progression and overall and 1-year survival.
2.3. Treatment plan On the GP arm, patients received gemcitabine administered at 1200 mg/m2 over 30 min in 500 ml of normal saline solution on days 1 and 8 and cisplatin administered at 80 mg/m2 over 45 min in 500 ml of normal saline solution on day 2. On the GC arm, patients received the same gemcitabine regimen, with carboplatin at an area under the curve (AUC) [26] of 5 mg/ml per minute given over 60 min in 500 ml of normal saline solution on day 2. Drugs were administered over a 21-day cycles. Prophylactic antiemetics for cisplatin and carbo-
83
platin were i.v. ondansetron 8 mg or i.v. granisetron 3 mg and i.v. dexamethasone 8 mg. Drugs were administered on an inpatient or outpatient basis according to institutional and/or patient preference. Patients completed treatment upon receiving six cycles unless one of the following occurred: disease progression, unacceptable toxicity, withdrawal of consent by the patient, physician’s discretion, external beam radiotherapy in patients with stage IIIB disease or surgical evaluation in downstaged patients. Post-study therapy was permitted at the discretion of the investigator. The day-8 gemcitabine dose was omitted either if the WBC count was B/2500/mm3 or the neutrophil count was B/1500/mm3 and/or the platelet count was B/100,000/mm3. Patients with febrile neutropenia, grade 4 thrombocytopenia or thrombocytopenia with bleeding had doses of all drugs reduced by 25% for subsequent cycles. If the WBC count was B/3500/mm3 and/or the platelet count was B/100,000/mm3 at the beginning of each cycle, a 1-week delay was required. In the event of a peak creatinine value of /5.0 /N or World Health Organization (WHO) grade 3/4 peripheral neurotoxicity, treatment was discontinued. The use of granulocyte-colony stimulating factors was not allowed during treatment, except for patients with febrile neutropenia, WHO grade 4 leukopenia or grade 3/4 neutropenia, according to investigator decision. Supportive care, including administration of blood transfusions, erythropoietin, antibiotics, antiemetics and analgesics was provided if judged appropriate by investigators.
2.4. Efficacy and tolerability assessments Staging assessments included a physical examination, complete blood count, electrocardiogram, chest computed tomography (CT) scan, chest X-ray, brain CT or MRI scan, bone scan and abdominal echography or CT scan, or, if appropriate, other radiological assessments (X-ray or MRI for bone lesions) and endoscopy. Routine blood tests for blood chemistry and hematological toxicity were performed before each chemotherapy administration. Response was assessed with CT scan or other staging methods after three and six cycles of treatment, using WHO objective response criteria [27]. The duration of response was measured from the date of randomization until documented progressive disease for responders only. Toxicity was graded according to WHO criteria and was assessed on days 1, 8 and 21 of each cycle by physical
84
examination, direct questioning and measurement of hematological and biochemical variables.
2.5. Statistical analysis Response rates, including 95% confidence intervals (CI), were calculated according to an intentto-treat (ITT) principle and proportions were compared using either a Pearson’s x2-test (corrected according to the Yates method) or the Fisher’s exact test, as appropriate. Survival endpoints were estimated using the Kaplan /Meier product-limit method and compared using the log-rank test. The prognostic significance of baseline, treatment and post-treatment variables on overall survival and time to disease progression was assessed using Cox’s regression model, after verification of the assumptions of proportional hazards. Incidences of toxicities were compared using Pearson’s x2-test. For all assessments, differences between treatment groups were considered significant at P B/ 0.05.
3. Results
P. Mazzanti et al.
treatment was discontinued for renal toxicity after the first cycle. According to an ITT analysis, objective responses were seen in 44 patients (26 GP and 18 GC; no complete responses), for an overall response rate of 36.7% (95% CI, 28.1 / 45.3%), with no significant difference found between the GP arm (41.9%) (95% CI, 29.6 /54.2%) and the GC arm (31.0%) (95% CI, 18.2 /42.8%) (P / 0.29). Most responses occurred within the first three cycles of treatment, with a median time to response of 10 weeks (range: 8 /25 weeks). The overall median duration of response was 43 weeks (range: 19 /141/), which was also the case for each chemotherapy arm (43 weeks). For patients with stage IIIB disease, response rates were 46.1% on the GP arm and 45.4% on the GC arm. Respective rates for patients with stage IV disease were 38.9 and 22.2%. When evaluated by metastatic sites, lung metastases were the most likely to respond to chemotherapy (31.0%), while liver metastases were the least likely (11.1%). Stable disease, which was recorded in 37 (30.8%) patients (24.2% GP and 37.9% GC), lasted /6 months for about half (18) of the patients and 36 (30.0%) patients (29.0% GP and 31.0% GC) progressed.
3.1. Patient and treatment characteristics 3.3. Survival and time to progression A total of 125 patients (62 GP and 63 GC) were enrolled in the study between January 1998 and August 2001. Five of the patients (two with stage IIIB and three with stage IV disease) were randomly assigned to the GC arm, but were ineligible to receive treatment (three with an ECOG PS of three at baseline, one pretreated with chemotherapy and one affected by a serious cardiac disease). Baseline patient and disease characteristics for the 120 patients treated with study drug (62 GP and 59 GC) are summarized in Table 1. Histological classification and differentiation information was missing in :/22 and 60% of the patients, respectively. This missing information was primarily related to patients who did not undergo surgical resection of the primary tumor and received only a cytological evaluation. Between stage IIIB patients, four had pleural effusions (two in GP and two in GC arm). Although there were some discrepancies between arms in terms of patient and disease characteristics, no differences reached statistical significance.
The median follow-up period was 10 months for all 120 patients, and 23 months for the 19 patients still alive as of April 2002. Eight patients (five in the GP arm, three in the GC arm) were lost to followup. The median survival time for all patients was 11.0 months (95% CI, 8.5 /13.4), with 1- and 2-year survival rates of 43 and 20%, respectively. Median survival times for the GP and GC arms were 10.4 months (95% CI, 8.0 /12.8) and 10.8 months (95% CI, 7.9 /13.8), respectively (P /0.39). Respective 1-year survival rates were 42 and 43% (P /0.92). There were no statistically significant differences between the two arms in overall survival (Fig. 1). As of this writing, 88.7% of the patients on the GP arm and 84.5% on the GC arm have progressed. Median time to disease progression for all treated patients was 5.4 months (95% CI, 4.0 /6.8), 6.7 months (95% CI, 5.4 /8.1) for the GP arm and 5.1 months (95% CI, 4.0 /6.2) for the GC arm (P /0.77) (Fig. 2).
3.4. Toxicity 3.2. Response Three (2.5%) patients (all on GP) were considered unassessable for response: two patients were lost to follow-up before restaging and one patient’s
In general, both treatments were well tolerated and WHO toxicities recorded at each cycle were mild (Table 2). The most frequently observed toxicities were leukopenia, anemia, thrombocyto-
Immunohistochemical expression of integrins and ECM proteins in NSCLC
85
Table 1 Patient characteristics Characteristic
Overall
No. of patients
120
GP arm
GC arm
62
58
Gender Male Female
94 (78.3%) 26 (21.7%)
45 (72.6%) 17 (27.4%)
49 (84.5%) 9 (15.5%)
Age, years Median Range
63 40 /75
60 40 /75
65 45 /75
ECOG performance status 0 1 2
25 (20.8%) 74 (61.7%) 21 (17.5%)
13 (21.0%) 36 (58.0%) 13 (21.0%)
12 (20.7%) 38 (65.5%) 8 (13.8%)
Differentiation Well/moderate Poor Unknown
10 (8.3%) 40 (33.3%) 70 (58.3%)
5 (8.1%) 16 (25.8%) 41 (66.1%)
5 (8.6%) 24 (41.4%) 29 (50.0%)
Histology Adenocarcinoma Squamous Unclassified Othera
54 (45.0%) 34 (28.3%) 27 (22.5%) 5 (4.2%)
24 (38.7%) 20 (32.3%) 15 (24.2%) 3 (4.8%)
30 (51.7%) 14 (24.1%) 12 (20.7%) 2 (3.5%)
13.5 8900 380
13.8 8700 428
13.0 9300 353
Stage IIIB IV
48 (40.0%) 72 (60.0%)
26 (41.9%) 36 (58.1%)
22 (37.9%) 36 (62.1%)
Metastatic sites (stage IV) Lung Bone Adrenals Distant nodes Liver Asymptomatic CNS Otherb Primary
29 (40.3%) 29 (40.3%) 17 (23.6%) 14 (19.4%) 9 (12.5%) 7 (9.7%) 3 (4.2%) 67 (93.0%)
15 (41.7%) 18 (50.0%) 7 (19.4%) 6 (16.7%) 4 (11.1%) 2 (5.6%) 2 (5.6%) 33 (91.7%)
14 (38.9%) 11 (30.6%) 10 (27.8%) 8 (22.2%) 5 (13.9%) 5 (13.9%) 1 (2.8%) 34 (94.4%)
No. metastatic sites (stage IV, primary not included) 1 43 (59.7%) 2 20 (27.8%) 3 or more 9 (12.5%)
19 (52.8%) 13 (36.1%) 4 (11.1%)
24 (66.7%) 7 (19.4%) 5 (13.9%)
Median pre-treatment values Hemoglobin (g/dl) White blood cells (/mm3) Lactate dehydrogenase (U/l)
GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; ECOG, Eastern Cooperative Oncology Group. a Large cell, adenosquamous. b Skin, spleen.
penia, nausea/vomiting, neuropathy and alopecia. Only one patient had febrile neutropenia (on the GP arm), no patients on either arm required platelet transfusions and there were no treatment-related deaths. As shown in Table 2, a significantly higher percentage of cycles on the GP arm were asso-
ciated with nausea/vomiting, renal toxicity and neuropathy versus the GC arm (P B/0.0005). When toxicity was evaluated by a patient the GP arm produced significantly higher incidences of nausea/ vomiting and renal toxicity compared with the GC arm (P B/0.02). When only grade 3/4 toxicities were compared (by cycle or by patient), however,
86
P. Mazzanti et al.
five (8.1%) patients on the GP arm and seven (12.1%) patients on the GC arm.
3.5. Dose administration and intensity Dose administration and intensity data are presented in Table 3. The total number of cycles administered (533) and the median number of cycles per patient were well balanced between the treatment groups. Twenty-two (35.5%) patients in the GP arm and 28 (48.3%) in the GC arm received the full six cycles of therapy. The day-8 gemcitabine dose was omitted in 55 (10.3%) cycles (31 on the GP arm and 24 on the GC arm), mainly because of hematological toxicity; the difference between arms was not statistically significant (P / 0.35). Dose intensities were similar across arms and well maintained during the first three cycles and over the course of the study.
Fig. 1 Overall survival.
3.6. Post-study therapy After primary chemotherapy, 14 (28.6%) patients with stage IIIB disease underwent explorative thoracotomy, but only nine patients (five GP and four GC) underwent radical resection (pneumonectomy and lobectomy in six and three patients, respectively). Two other patients deemed resectable did not receive surgery because of patient refusal. Two (2.8%) patients with stage IV disease, both with primary tumor and lung metastases,
Fig. 2 Overall time to progression.
no significant differences between the two arms were found. Growth factors were administered in 9.8 and 8.2% of cycles on the GP arm and GC arms, respectively, generally for grade 4 leukopenia or neutropenia. Erythropoietin was administered to Table 2 WHO toxicity evaluated by cycle (N/533) Toxicity
Anemia Neutropenia Granulocytopenia Thrombocytopenia Nausea/vomiting Mucositis Diarrhea Hepatic Renal Neurologic Constipation Hypersensitivity Skin Othersb
GP arm (264 cycles)
P valuea
GC arm (269 cycles)
G 1/2
G 3/4
Total
%
G 1/2
G 3/4
Total
%
69 47 27 28 79 7 2 8 15 32 6 2 6 12
3 6 10 17 11 0 0 0 1 2 0 0 0 0
72 53 37 45 90 7 2 8 16 34 6 2 6 12
27.3 20.1 14.0 17.0 34.1 2.6 0.7 3.0 6.1 12.9 2.3 0.7 2.3 4.5
61 39 26 30 38 7 4 4 1 11 15 3 2 5
6 7 7 12 5 0 0 0 0 0 1 0 0 0
67 46 33 42 43 7 4 4 1 11 16 3 2 5
24.9 17.1 12.3 15.6 16.0 2.6 1.5 1.5 0.4 4.1 5.9 1.1 0.7 1.8
0.601 0.440 0.639 0.741 B/0.0001 0.814 0.698 0.363 0.0005 0.0005 0.056 0.983 0.273 0.077
WHO, World Health Organization; GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; G, grade. a Differences between arms in WHO toxicities (all grades combined) were detected using a x2-test; P B/0.05 was considered significant. b Thrombocytosis, cardiac toxicity, anorexia.
Immunohistochemical expression of integrins and ECM proteins in NSCLC
87
Table 3 Dose administration and intensity
No. of cycles delivered Total Median Range Mean delivered dose/patient/week Cycles 1 /3 mg/m2 % Of planned All cycles mg/m2 % Of planned
GP arm
GC arm
264 4 1 /6
269 5 2 /6
G
P
G
C
649.0 81
23.3 87
643.0 80
AUC 86
631.0 79
22.9 86
629.0 78
AUC 86
GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; AUC, area under curve.
underwent radical surgery after chemotherapy disease shrinkage. Twenty-six patients with stage IIIB disease (14 GP and 12 GC) were treated by curative external beam radiotherapy of the chest. Radiotherapy for symptom palliation was administered to 21 patients. A total of 45 patients (37.5%, 22 GP and 23 GC arm) received second-line chemotherapy and eight (6.7%) patients received third-line chemotherapy.
3.7. Prognostic factors The multivariate analysis according to the Cox regression model, identified an ECOG PS of 0 or 1, stage IIIB disease, higher hemoglobin levels at baseline, response to treatment and second-line
chemotherapy as significant independent prognostic indicators of better overall survival (Table 4). Moreover, when the variables, ‘radical surgery’ and ‘external beam chest radiotherapy’ (both performed after chemotherapy) were included in the model, they were associated with a higher survival (P /0.03 and P /0.001, respectively); whereas, the variable, ‘stage,’ lost its significance (P / 0.134). For time to disease progression, significant favorable prognostic factors were stage IIIB disease, higher hemoglobin levels at baseline and response to treatment. Median survival and median time to progression were not affected after adjustment for prognostic factors.
Table 4 Prognostic factors for overall survival and time to disease progression: results of the multivariate analysis Factors
Survival Wald test (x2)
Age (continuous variable) 0.312 Sex (male vs. female) 3.500 Performance status (ECOG 0 /1 vs. 2) 3.843 Hemoglobin (continuous variable) 4.607 White blood cells (continuous variable) 2.730 Histology (adenocarcinoma vs. squamous vs. unclassi4.448 fied vs. other) Stage (IIIB vs. IV) 11.413 Type of treatment (GP vs. GC) 0.0007 Weekly dose intensity for platinum compounds (contin- 1.821 uous variable) Weekly dose intensity for gemcitabine (continuous 1.294 variable) Treatment response (PR vs. SD vs. PD) 25.111 Second-line chemotherapy (yes vs. no) 10.357
Time to disease progression dof P value
Wald test (x2)
dof P value
1 1 1 1 1 3
0.577 0.061 0.049 0.032 0.098 0.217
1.431 1.815 3.875 5.544 0.004 5.197
1 1 1 1 1 3
0.231 0.177 0.586 0.018 0.944 0.157
1 1 1
0.0007 0.980 0.177
8.689 0.988 2.335
1 1 1
0.003 0.320 0.126
1
0.255
0.542
1
0.461
B/0.0001 66.499 0.001 /
2 /
2 1
B/0.0001 /
dof, Degrees of freedom; ECOG, Eastern Cooperative Oncology Group; GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; PR, partial response; SD, stable disease; PD, partial disease.
88
4. Discussion Cisplatin-based chemotherapy is a reasonable treatment option for patients with advanced NSCLC. Old cisplatin-based regimens have given way to those using newer third generation chemotherapy (such as gemcitabine, paclitaxel, vinorelbine) that have demonstrated higher activity and better tolerability and palliation capacity [28]. None of these newer regimens, however, have clearly demonstrated superior efficacy in phase III trials [14,15]. Therefore, tolerability has become an important consideration when making treatment decisions. In our trial in advanced NSCLC, there were no statistical differences between GC and GP in terms of efficacy. We obtained similar response rates in the two arms in stage IIIB patients and a slight, but not statistically significant, advantage for stage IV patients in the GP arm (38.9 vs. 22.2%, P /0.20). The similarity of the two regimens was underscored by the results of the Cox’s regression analysis, which did not identify the type of treatment as a prognostic factor for overall survival (P /0.98). The efficacy results obtained with GC are consistent with those reported in previous phase I/II trials applying the same regimen [18]. Of interest, the overall response rate of 31%, the median survival time of 10.8 months and the 1-year survival rate of 43% obtained with the GC regimen in this study compared well to those reported for other active third-generation regimens [14 /16]. The importance of clinical factors as prognostic indicators is generally well known and should not be overlooked when making treatment choices for patients with NSCLC. Through the multivariate analysis, we found that a better survival outcome was predicted for patients with stage IIIB disease as opposed to stage IV disease; however, the outcome for patients with stage IIIB disease was strictly dependent on the patient’s opportunity to receive a radical resection of down-staged pulmonary disease or external-beam chest radiotherapy after chemotherapy. Our analysis further indicated that second-line chemotherapy was an independent prognostic factor for improved survival in patients with advanced NSCLC. For time to disease progression, stage IIIB disease, an ECOG PS of 0 or 1, high baseline hemoglobin and response to treatment were identified as having favorable independent prognostic potential. We observed grade 3/4 thrombocytopenia in 16.1% of patients and 6.4% of cycles on the GP arm; however, no patients on this arm required platelet transfusions. These results compare well with Spanish [10] and Italian [13] trials employing
P. Mazzanti et al.
the same 3-week schedule. The incidence of grade 3/4 thrombocytopenia on our GC arm (13.8% of patients and 4.5% of cycles) did not differ significantly from that of the GP arm. The use of growth factors was very limited. Better overall tolerability was seen in the GC arm for nausea/vomiting, renal and neurological toxicity (all grades combined). Compared with previous studies applying the same regimens, our study had a lower overall incidence of toxic events. In addition, hematological toxicity had a minimal effect on the day-8 delivered dose intensity of gemcitabine, only affecting :/10% of the doses missed for both arms combined. One explanation for this could be our choice of regimen (both the 3-week schedule and the administration of platinum-compounds on day 2). An assessment of the data from previous trials revealed that an early administration of cisplatin is associated with a low incidence of neutropenia [29], a finding that is supported in the current study, together with a low incidence of thrombocytopenia.
5. Conclusion In this multicenter randomized phase II trial, GP and GC did not provide any significant difference in response rate in patients with advanced NSCLC, with GC exhibiting better overall tolerability than GP. The potential therapeutic advantages of GC in terms of overall tolerability as well as ease of administration, make this novel combination a valid alternative in the palliative treatment especially for the ‘unfit’ patients.
Acknowledgements We are grateful to Mary Alice Miller, PhD, Rachel Shemer, Scientific Communications Consultants and Noelle Gasco, Scientific Communications Associate, of Eli Lilly and Co. for general preparation of this manuscript. We also thank Dr Filippo de Braud, Director of New Drugs Unit and Dr Tommaso De Pas, Vice-Director of Medical Oncology, of European Institute of Oncology for critically reviewing the manuscript.
References [1] Rapp E, Pater JL, Willan A, et al. Chemotherapy can prolong survival in patients with advanced non-small-cell lung cancer: report of a Canadian multicenter randomized trial. J Clin Oncol 1988;6:633 /41. [2] Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis
Immunohistochemical expression of integrins and ECM proteins in NSCLC
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
using updated data on individual patients from 52 randomized clinical trials. Br Med J 2000;311:899 /909. Bunn PA, Jr, Kelly K. New chemotherapeutic agents prolong survival and improve quality of life in non-small cell lung cancer: a review of the literature and future directions. Clin Cancer Res 1998;4:1087 /2100. Sandler A, Ettinger DS. Gemcitabine: single-agent and combination therapy in non-small cell lung cancer. Oncologist 1999;4:241 /51. Bergman AM, Ruiz van Haperen VW, Veerman G, et al. Synergistic interaction between cisplatin and gemcitabine in vitro. Clin Cancer Res 1996;2:521 /30. Crino ` L, Scagliotti G, Marangolo M, et al. Cisplatin / gemcitabine combination in non-small-cell lung cancer. J Clin Oncol 1997;15:297 /303. Abratt RP, Bezwoda WR, Goedhals L, et al. Weekly gemcitabine with monthly cisplatin: effective chemotherapy for advanced non-small-cell lung cancer. J Clin Oncol 1997;15:744 /9. Sandler AB, Ansari R, McClean J, et al. A Hoosier Oncology Group phase II study of gemcitabine plus cisplatin in nonsmall cell lung cancer. Cancer Therapeut 1998;1:158 /63. Anton A, Diaz-Fernandez N, Gonzalez Larriba JL, et al. Phase II trial assessing the combination of gemcitabine and cisplatin in advanced non-small cell lung cancer (NSCLC). Lung Cancer 1998;22:139 /48. Castellano D, Lianes P, Paz-Ares L, et al. A phase II study of a novel gemcitabine plus cisplatin regimen administered every three weeks for advanced non-small-cell lung cancer. Ann Oncol 1998;9:457 /9. Soto Parra HJ, Cavina R, Antonelli G, et al. Superiority of three-week vs four-week schedule of cisplatin (CDDP) and gemcitabine (GEM): results of a randomized phase II study. Proc Am Soc Clin Oncol 2000;19:546a (abstract 2152). Cardenal F, Lopez-Cabrerizo MP, Anton A, et al. Randomized phase III study of gemcitabine-cisplatin versus etoposide-cisplatin in the treatment of locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 1999;17:12 /8. Crino ` L, Scagliotti GV, Ricci S, et al. Gemcitabine and cisplatin versus mitomycin, ifosfamide, and cisplatin in advanced non-small-cell lung cancer: a randomized phase III study of the Italian Lung Cancer Project. J Clin Oncol 1999;17:3522 /30. Scagliotti GV, De Marinis F, Rinaldi M, et al. Phase III randomized trial comparing three platinum-based doublets in advanced non-small cell lung cancer. J Clin Oncol 2002;20(21):4285 /91. Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. New Engl J Med 2002;346:92 /8.
.
89
[16] Klastersky J, Sculier JP, Lacroix H, et al. A randomized study comparing cisplatin or carboplatin with etoposide in patients with advanced non-small-cell lung cancer: European Organization for Research and Treatment of Cancer Protocol 07861. J Clin Oncol 1990;8:1556 /62. [17] Rosell R, Gatzeimer U, Betticher DC, et al. Phase III randomized trial comparing paclitaxel/carboplatin with paclitaxel/cisplatin in patients with advanced non-smallcell lung cancer: a cooperative multinational trial. Ann Oncol 2002;13:1539 /49. [18] Langer CJ, Gandara DR, Calvert P, et al. Gemcitabine and carboplatin in combination: an update of phase I and phase II studies in non-small cell lung cancer. Semin Oncol 1999;26(Suppl 4):12 /8. [19] Iaffaioli RV, Tortoriello A, Facchini G, et al. Phase I /II study of gemcitabine and carboplatin in stage IIIB /IV non-smallcell lung cancer. J Clin Oncol 1999;17:921 /6. [20] Ng EW, Sandler AB, Robinson L, Einhorn LH. A phase II study of carboplatin plus gemcitabine in advanced non-small-cell lung cancer (NSCLC): a Hoosier oncology group study. Am J Clin Oncol 1999;22:550 /3. [21] Gajra A, Mehdi SA, Kirshner J, et al. Phase I study of gemcitabine and carboplatin in advanced non-small cell lung carcinoma. Lung Cancer 2001;32:189 /96. [22] Carrato A, Garcia-Gomez J, Alberola V, et al. Carboplatin (CARBO) in combination with gemcitabine (GEM) in advanced non-small cell lung cancer (NSCLC). Comparison of two consecutive phase II trials using different schedules. Proc Am Soc Clin Oncol 1999;18:498a (Abstract 1922). [23] Masotti A, Zannini G, Gentile A, Morandini G. Activity of gemcitabine and carboplatin in advanced non-small cell lung cancer: a phase II trial. Lung Cancer 2002;36:99 /103. [24] Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 1975;31:103 /15. [25] Mazzanti P, Delprete S, Battelli N, et al. Combination of gemcitabine and cisplatin in advanced non-small cell lung cancer (NSCLC): a phase II study. Proc Am Soc Clin Oncol 1999;18:517a (Abstract 1993). [26] Calvert AH, Newell DR, Gumbrell LA, et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989;7:1748 /56. [27] Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981;47:207 /14. [28] Gebbia V, Galetta D, Riccardi F, et al. Vinorelbine plus cisplatin versus cisplatin plus vindesine and mitomycin C in stage IIIB /IV non-small cell lung carcinoma: a prospective randomized study. Lung Cancer 2002;37:179 /87. [29] Shepherd FA, Abratt R, Crino ` L, et al. The influence of gemcitabine and cisplatin schedule on response and survival in advanced non-small cell lung cancer. Lung Cancer 2000;30:117 /25.
www.elsevier.com/locate/lungcan
Randomized, multicenter, phase II study of gemcitabine plus cisplatin versus gemcitabine plus carboplatin in patients with advanced non-small cell lung cancer Paola Mazzantia,*, Cristian Massacesia, Marco B.L. Rocchib, Rodolfo Mattiolic, Paolo Lippec, Raffaele Trivisonned, Franco Buzzie, Giorgio De Signoribusf, Guido Tuverig, Giorgio Rossih, Liberato Di Lulloi, Fabio Sturbaj, Donatella Moraled, Sergio Catanzanie, Alberta Pilonea, Maurizio Bonsignoria, Tullio Battellik a
Department of Medical Oncology, Azienda Ospedaliera Umberto I, via Conca, 60122 Torrette di Ancona, Italy b Biomathematics Institute, University of Urbino, Loc. Crocicchia, 61029 Urbino, Italy c Medical Oncology Unit of Fano, Ospedale S. Croce, via Veneto 2, 61032 Fano, Italy d Medical Oncology Unit of Ascoli Piceno, Ospedale Mazzoni, viale delle Rimembranze, 63100 Ascoli Piceno, Italy e Medical Oncology, Ospedale S. Maria, via Tristano di Joannuccio 1, 05100 Terni, Italy f Medical Oncology Unit, Ospedale Madonna del Soccorso, via Pellico, 63039 San Benedetto, Italy g Medical Oncology, Azienda Ospedaliera Trieste, via Pieta` 19, 34100 Trieste, Italy h Medical Oncology Unit, Ospedale Murri, via Gramsci, 60035 Jesi, Italy i Medical Oncology Unit, ASL Pentria 2, 86170 Isernia, Italy j Medical Oncology Unit A.O., Contrada S. Domenico, 62012 Civitanova, Italy k Adria Medica Study Group, Italy Received 12 December 2002; received in revised form 26 February 2003; accepted 28 February 2003
KEYWORDS Carboplatin; Cisplatin; Gemcitabine; Non-small cell lung cancer; Phase II; Randomized
Summary Background: We conducted a phase II randomized study to assess the efficacy, with response as the primary endpoint, and the toxicity of gemcitabine/ cisplatin (GP) and gemcitabine/carboplatin (GC) in patients with advanced non-small cell lung cancer (NSCLC). Methods: Patients were randomized to GP (gemcitabine 1200 mg/m2, days 1 and 8 plus cisplatin 80 mg/m2 day 2) or GC (gemcitabine 1200 mg/ m2, days 1 and 8 plus carboplatin AUC /5 day 2). Cycles were repeated every 3 weeks. Results: Sixty-two patients were randomized to GP and 58 to GC. A total of 533 cycles were delivered (264 GP, 269 GC), with a median of four cycles/patient. The objective response rate was 41.9% (95% C.I., 29.6 /54.2%) for GP and 31.0% (95% C.I.,
*Corresponding author. Tel.: /39-071-5964-270; fax: /39071-5964-269. E-mail address: [email protected] (P. Mazzanti). 0169-5002/03/$ - see front matter – 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0169-5002(03)00140-5
82
P. Mazzanti et al.
18.2 /42.8%) for GC (P /0.29). No significant differences between arms were observed in median survival (10.4 months GP, 10.8 months GC) and median time to progression (5.4 months GP, 5.1 months GC). Both regimens were very well tolerated with no statistical differences between arms in grade 3/4 toxicities. When all toxicity grades were combined, emesis, neuropathy and renal toxicity occurred more frequently on the GP arm (P B/0.005). Conclusions: GC arm did not provide a significant difference in response rate compared with GP arm, with better overall tolerability. Carboplatin could be a valid alternative to cisplatin in the palliative setting. – 2003 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction Platinum compounds remain the cornerstone of chemotherapy for advanced NSCLC. A number of randomized trials and meta-analyses have reported that cisplatin-based chemotherapy produces modest but significantly improved survival with better symptom control and quality of life compared to that of supportive care alone [1 /3]. Gemcitabine (2?,2?-difluorodeoxycytidine) is a nucleoside antimetabolite that has confirmed activity against several solid tumors, including NSCLC [4]. The combination of cisplatin and gemcitabine has demonstrated synergism in preclinical studies, which is likely related to the interaction of both drugs on the inhibition of DNA repair [5]. Many phase II studies have demonstrated excellent clinical activity of the combination, although most have been characterized by significant hematologic toxicity [6 /10]. This toxicity, namely thrombocytopenia, has often been attributed to the 4-week schedule [6,8,11]. More recently, the combination of cisplatin and gemcitabine fared well when compared with cisplatin-based chemotherapy using older chemotherapy (such as vindesine, mitomcycin, etoposide, ifosfamide) and, in particular, those using third-generation chemotherapy (such as gemcitabine, vinorelbine, taxanes, irinotecan) in large phase III trials [12 / 15]. Carboplatin, an analog of cisplatin, administered alone or in combination therapy, has lower nephroand neurotoxicity compared with cisplatin, has been proven as effective as cisplatin in NSCLC [16] and unlike cisplatin, offers the possibility of ambulatory administration. Carboplatin has been used extensively in combination with paclitaxel, especially in the US, producing a similar efficacy outcome but a different toxicity profile compared
with other regimens [14,15]. Indeed, recently, Rosell et al. [17] making a direct comparison between paclitaxel/cisplatin and paclitaxel/carboplatin, obtained a significantly longer survival in the cisplatin arm. Few studies have been conducted, however, using the combination of carboplatin and gemcitabine. Phase I/II trials have explored both 3- and 4-week schedules, but these studies have often produced contradictory results in terms of recommended doses and toxicity profiles [18 /21]. No differences between the 3and 4-week schedules in activity were observed, although less thrombocytopenia was detected with the 3-week regimen [22,23]. Having said that, we can consider gemcitabine/ carboplatin (GC) an attractive alternative to gemcitabine/cisplatin (GP) in the treatment of NSCLC. Therefore, we decided to evaluate the efficacy, with response as the primary endpoint and the toxicity of GP versus GC, administered as 3-week schedules to patients with advanced NSCLC, in a multicenter, randomized trial.
2. Methods 2.1. Patient selection Patients had to meet the following inclusion criteria to be enrolled: histologically or cytologically confirmed stage IIIB or IV NSCLC (according to the American Joint Committee on Cancer staging system, 1992); at least one measurable lesion; age 18 /75 years; no prior chemotherapy or immunotherapy; performance status (PS) of 0 /2 on the Eastern Cooperative Oncology Group (ECOG) scale; life expectancy /12 weeks; adequate bone marrow reserve (white blood cell (WBC) count ]/4000/ mm3, neutrophil count ]/2000/mm3, platelet count ]/100,000/mm3 and hemoglobin ]/10 g/
Immunohistochemical expression of integrins and ECM proteins in NSCLC
dl); renal and hepatic function within the normal range (bilirubin B/1.5 mg/dl, transaminase levels less than three times the upper limit of normal and serum creatinine 5/1.4 mg/dl); and normal cardiac function. Patients who had received previous radiotherapy were included if their assessable disease was outside of the radiation field. The main exclusion criteria were: symptomatic central nervous system metastases, second primary malignancy and serious systemic disorders. Written informed consent was obtained from each patient before enrollment. The study protocol was in accordance with the ethical standards of Human Investigation Committee of Ancona, the coordinating center.
2.2. Study design This was a phase II, prospective, randomized, multicenter trial. Eligible patients were randomized to one of two arms, GP or GC, using a concealed list of random numbers. The randomization algorithm, based on the Pocock and Simon method [24], included ECOG PS (0 /1 vs. 2) and disease stage (IIIB vs. IV) as stratification factors. The trial was planned as a randomized phase II study to obtain some useful information for further development in a controlled randomized phase III setting. Thus, the findings obtained from the treatment-arm comparisons of this phase II study should be considered exploratory. The primary endpoint was response rate. Assuming a response rate of 50% in the GP arm [6,10,25] and an expected response rate of 25% in the GC arm, a sample size of 63 patients in each arm was required to detect a difference of 25%. This sample size was determined using a statistical power (1-b) of 80% and a significance level (a ) of 5%. Two sided testing was applied. Secondary endpoints included duration of response, toxicity, time to progression and overall and 1-year survival.
2.3. Treatment plan On the GP arm, patients received gemcitabine administered at 1200 mg/m2 over 30 min in 500 ml of normal saline solution on days 1 and 8 and cisplatin administered at 80 mg/m2 over 45 min in 500 ml of normal saline solution on day 2. On the GC arm, patients received the same gemcitabine regimen, with carboplatin at an area under the curve (AUC) [26] of 5 mg/ml per minute given over 60 min in 500 ml of normal saline solution on day 2. Drugs were administered over a 21-day cycles. Prophylactic antiemetics for cisplatin and carbo-
83
platin were i.v. ondansetron 8 mg or i.v. granisetron 3 mg and i.v. dexamethasone 8 mg. Drugs were administered on an inpatient or outpatient basis according to institutional and/or patient preference. Patients completed treatment upon receiving six cycles unless one of the following occurred: disease progression, unacceptable toxicity, withdrawal of consent by the patient, physician’s discretion, external beam radiotherapy in patients with stage IIIB disease or surgical evaluation in downstaged patients. Post-study therapy was permitted at the discretion of the investigator. The day-8 gemcitabine dose was omitted either if the WBC count was B/2500/mm3 or the neutrophil count was B/1500/mm3 and/or the platelet count was B/100,000/mm3. Patients with febrile neutropenia, grade 4 thrombocytopenia or thrombocytopenia with bleeding had doses of all drugs reduced by 25% for subsequent cycles. If the WBC count was B/3500/mm3 and/or the platelet count was B/100,000/mm3 at the beginning of each cycle, a 1-week delay was required. In the event of a peak creatinine value of /5.0 /N or World Health Organization (WHO) grade 3/4 peripheral neurotoxicity, treatment was discontinued. The use of granulocyte-colony stimulating factors was not allowed during treatment, except for patients with febrile neutropenia, WHO grade 4 leukopenia or grade 3/4 neutropenia, according to investigator decision. Supportive care, including administration of blood transfusions, erythropoietin, antibiotics, antiemetics and analgesics was provided if judged appropriate by investigators.
2.4. Efficacy and tolerability assessments Staging assessments included a physical examination, complete blood count, electrocardiogram, chest computed tomography (CT) scan, chest X-ray, brain CT or MRI scan, bone scan and abdominal echography or CT scan, or, if appropriate, other radiological assessments (X-ray or MRI for bone lesions) and endoscopy. Routine blood tests for blood chemistry and hematological toxicity were performed before each chemotherapy administration. Response was assessed with CT scan or other staging methods after three and six cycles of treatment, using WHO objective response criteria [27]. The duration of response was measured from the date of randomization until documented progressive disease for responders only. Toxicity was graded according to WHO criteria and was assessed on days 1, 8 and 21 of each cycle by physical
84
examination, direct questioning and measurement of hematological and biochemical variables.
2.5. Statistical analysis Response rates, including 95% confidence intervals (CI), were calculated according to an intentto-treat (ITT) principle and proportions were compared using either a Pearson’s x2-test (corrected according to the Yates method) or the Fisher’s exact test, as appropriate. Survival endpoints were estimated using the Kaplan /Meier product-limit method and compared using the log-rank test. The prognostic significance of baseline, treatment and post-treatment variables on overall survival and time to disease progression was assessed using Cox’s regression model, after verification of the assumptions of proportional hazards. Incidences of toxicities were compared using Pearson’s x2-test. For all assessments, differences between treatment groups were considered significant at P B/ 0.05.
3. Results
P. Mazzanti et al.
treatment was discontinued for renal toxicity after the first cycle. According to an ITT analysis, objective responses were seen in 44 patients (26 GP and 18 GC; no complete responses), for an overall response rate of 36.7% (95% CI, 28.1 / 45.3%), with no significant difference found between the GP arm (41.9%) (95% CI, 29.6 /54.2%) and the GC arm (31.0%) (95% CI, 18.2 /42.8%) (P / 0.29). Most responses occurred within the first three cycles of treatment, with a median time to response of 10 weeks (range: 8 /25 weeks). The overall median duration of response was 43 weeks (range: 19 /141/), which was also the case for each chemotherapy arm (43 weeks). For patients with stage IIIB disease, response rates were 46.1% on the GP arm and 45.4% on the GC arm. Respective rates for patients with stage IV disease were 38.9 and 22.2%. When evaluated by metastatic sites, lung metastases were the most likely to respond to chemotherapy (31.0%), while liver metastases were the least likely (11.1%). Stable disease, which was recorded in 37 (30.8%) patients (24.2% GP and 37.9% GC), lasted /6 months for about half (18) of the patients and 36 (30.0%) patients (29.0% GP and 31.0% GC) progressed.
3.1. Patient and treatment characteristics 3.3. Survival and time to progression A total of 125 patients (62 GP and 63 GC) were enrolled in the study between January 1998 and August 2001. Five of the patients (two with stage IIIB and three with stage IV disease) were randomly assigned to the GC arm, but were ineligible to receive treatment (three with an ECOG PS of three at baseline, one pretreated with chemotherapy and one affected by a serious cardiac disease). Baseline patient and disease characteristics for the 120 patients treated with study drug (62 GP and 59 GC) are summarized in Table 1. Histological classification and differentiation information was missing in :/22 and 60% of the patients, respectively. This missing information was primarily related to patients who did not undergo surgical resection of the primary tumor and received only a cytological evaluation. Between stage IIIB patients, four had pleural effusions (two in GP and two in GC arm). Although there were some discrepancies between arms in terms of patient and disease characteristics, no differences reached statistical significance.
The median follow-up period was 10 months for all 120 patients, and 23 months for the 19 patients still alive as of April 2002. Eight patients (five in the GP arm, three in the GC arm) were lost to followup. The median survival time for all patients was 11.0 months (95% CI, 8.5 /13.4), with 1- and 2-year survival rates of 43 and 20%, respectively. Median survival times for the GP and GC arms were 10.4 months (95% CI, 8.0 /12.8) and 10.8 months (95% CI, 7.9 /13.8), respectively (P /0.39). Respective 1-year survival rates were 42 and 43% (P /0.92). There were no statistically significant differences between the two arms in overall survival (Fig. 1). As of this writing, 88.7% of the patients on the GP arm and 84.5% on the GC arm have progressed. Median time to disease progression for all treated patients was 5.4 months (95% CI, 4.0 /6.8), 6.7 months (95% CI, 5.4 /8.1) for the GP arm and 5.1 months (95% CI, 4.0 /6.2) for the GC arm (P /0.77) (Fig. 2).
3.4. Toxicity 3.2. Response Three (2.5%) patients (all on GP) were considered unassessable for response: two patients were lost to follow-up before restaging and one patient’s
In general, both treatments were well tolerated and WHO toxicities recorded at each cycle were mild (Table 2). The most frequently observed toxicities were leukopenia, anemia, thrombocyto-
Immunohistochemical expression of integrins and ECM proteins in NSCLC
85
Table 1 Patient characteristics Characteristic
Overall
No. of patients
120
GP arm
GC arm
62
58
Gender Male Female
94 (78.3%) 26 (21.7%)
45 (72.6%) 17 (27.4%)
49 (84.5%) 9 (15.5%)
Age, years Median Range
63 40 /75
60 40 /75
65 45 /75
ECOG performance status 0 1 2
25 (20.8%) 74 (61.7%) 21 (17.5%)
13 (21.0%) 36 (58.0%) 13 (21.0%)
12 (20.7%) 38 (65.5%) 8 (13.8%)
Differentiation Well/moderate Poor Unknown
10 (8.3%) 40 (33.3%) 70 (58.3%)
5 (8.1%) 16 (25.8%) 41 (66.1%)
5 (8.6%) 24 (41.4%) 29 (50.0%)
Histology Adenocarcinoma Squamous Unclassified Othera
54 (45.0%) 34 (28.3%) 27 (22.5%) 5 (4.2%)
24 (38.7%) 20 (32.3%) 15 (24.2%) 3 (4.8%)
30 (51.7%) 14 (24.1%) 12 (20.7%) 2 (3.5%)
13.5 8900 380
13.8 8700 428
13.0 9300 353
Stage IIIB IV
48 (40.0%) 72 (60.0%)
26 (41.9%) 36 (58.1%)
22 (37.9%) 36 (62.1%)
Metastatic sites (stage IV) Lung Bone Adrenals Distant nodes Liver Asymptomatic CNS Otherb Primary
29 (40.3%) 29 (40.3%) 17 (23.6%) 14 (19.4%) 9 (12.5%) 7 (9.7%) 3 (4.2%) 67 (93.0%)
15 (41.7%) 18 (50.0%) 7 (19.4%) 6 (16.7%) 4 (11.1%) 2 (5.6%) 2 (5.6%) 33 (91.7%)
14 (38.9%) 11 (30.6%) 10 (27.8%) 8 (22.2%) 5 (13.9%) 5 (13.9%) 1 (2.8%) 34 (94.4%)
No. metastatic sites (stage IV, primary not included) 1 43 (59.7%) 2 20 (27.8%) 3 or more 9 (12.5%)
19 (52.8%) 13 (36.1%) 4 (11.1%)
24 (66.7%) 7 (19.4%) 5 (13.9%)
Median pre-treatment values Hemoglobin (g/dl) White blood cells (/mm3) Lactate dehydrogenase (U/l)
GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; ECOG, Eastern Cooperative Oncology Group. a Large cell, adenosquamous. b Skin, spleen.
penia, nausea/vomiting, neuropathy and alopecia. Only one patient had febrile neutropenia (on the GP arm), no patients on either arm required platelet transfusions and there were no treatment-related deaths. As shown in Table 2, a significantly higher percentage of cycles on the GP arm were asso-
ciated with nausea/vomiting, renal toxicity and neuropathy versus the GC arm (P B/0.0005). When toxicity was evaluated by a patient the GP arm produced significantly higher incidences of nausea/ vomiting and renal toxicity compared with the GC arm (P B/0.02). When only grade 3/4 toxicities were compared (by cycle or by patient), however,
86
P. Mazzanti et al.
five (8.1%) patients on the GP arm and seven (12.1%) patients on the GC arm.
3.5. Dose administration and intensity Dose administration and intensity data are presented in Table 3. The total number of cycles administered (533) and the median number of cycles per patient were well balanced between the treatment groups. Twenty-two (35.5%) patients in the GP arm and 28 (48.3%) in the GC arm received the full six cycles of therapy. The day-8 gemcitabine dose was omitted in 55 (10.3%) cycles (31 on the GP arm and 24 on the GC arm), mainly because of hematological toxicity; the difference between arms was not statistically significant (P / 0.35). Dose intensities were similar across arms and well maintained during the first three cycles and over the course of the study.
Fig. 1 Overall survival.
3.6. Post-study therapy After primary chemotherapy, 14 (28.6%) patients with stage IIIB disease underwent explorative thoracotomy, but only nine patients (five GP and four GC) underwent radical resection (pneumonectomy and lobectomy in six and three patients, respectively). Two other patients deemed resectable did not receive surgery because of patient refusal. Two (2.8%) patients with stage IV disease, both with primary tumor and lung metastases,
Fig. 2 Overall time to progression.
no significant differences between the two arms were found. Growth factors were administered in 9.8 and 8.2% of cycles on the GP arm and GC arms, respectively, generally for grade 4 leukopenia or neutropenia. Erythropoietin was administered to Table 2 WHO toxicity evaluated by cycle (N/533) Toxicity
Anemia Neutropenia Granulocytopenia Thrombocytopenia Nausea/vomiting Mucositis Diarrhea Hepatic Renal Neurologic Constipation Hypersensitivity Skin Othersb
GP arm (264 cycles)
P valuea
GC arm (269 cycles)
G 1/2
G 3/4
Total
%
G 1/2
G 3/4
Total
%
69 47 27 28 79 7 2 8 15 32 6 2 6 12
3 6 10 17 11 0 0 0 1 2 0 0 0 0
72 53 37 45 90 7 2 8 16 34 6 2 6 12
27.3 20.1 14.0 17.0 34.1 2.6 0.7 3.0 6.1 12.9 2.3 0.7 2.3 4.5
61 39 26 30 38 7 4 4 1 11 15 3 2 5
6 7 7 12 5 0 0 0 0 0 1 0 0 0
67 46 33 42 43 7 4 4 1 11 16 3 2 5
24.9 17.1 12.3 15.6 16.0 2.6 1.5 1.5 0.4 4.1 5.9 1.1 0.7 1.8
0.601 0.440 0.639 0.741 B/0.0001 0.814 0.698 0.363 0.0005 0.0005 0.056 0.983 0.273 0.077
WHO, World Health Organization; GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; G, grade. a Differences between arms in WHO toxicities (all grades combined) were detected using a x2-test; P B/0.05 was considered significant. b Thrombocytosis, cardiac toxicity, anorexia.
Immunohistochemical expression of integrins and ECM proteins in NSCLC
87
Table 3 Dose administration and intensity
No. of cycles delivered Total Median Range Mean delivered dose/patient/week Cycles 1 /3 mg/m2 % Of planned All cycles mg/m2 % Of planned
GP arm
GC arm
264 4 1 /6
269 5 2 /6
G
P
G
C
649.0 81
23.3 87
643.0 80
AUC 86
631.0 79
22.9 86
629.0 78
AUC 86
GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; AUC, area under curve.
underwent radical surgery after chemotherapy disease shrinkage. Twenty-six patients with stage IIIB disease (14 GP and 12 GC) were treated by curative external beam radiotherapy of the chest. Radiotherapy for symptom palliation was administered to 21 patients. A total of 45 patients (37.5%, 22 GP and 23 GC arm) received second-line chemotherapy and eight (6.7%) patients received third-line chemotherapy.
3.7. Prognostic factors The multivariate analysis according to the Cox regression model, identified an ECOG PS of 0 or 1, stage IIIB disease, higher hemoglobin levels at baseline, response to treatment and second-line
chemotherapy as significant independent prognostic indicators of better overall survival (Table 4). Moreover, when the variables, ‘radical surgery’ and ‘external beam chest radiotherapy’ (both performed after chemotherapy) were included in the model, they were associated with a higher survival (P /0.03 and P /0.001, respectively); whereas, the variable, ‘stage,’ lost its significance (P / 0.134). For time to disease progression, significant favorable prognostic factors were stage IIIB disease, higher hemoglobin levels at baseline and response to treatment. Median survival and median time to progression were not affected after adjustment for prognostic factors.
Table 4 Prognostic factors for overall survival and time to disease progression: results of the multivariate analysis Factors
Survival Wald test (x2)
Age (continuous variable) 0.312 Sex (male vs. female) 3.500 Performance status (ECOG 0 /1 vs. 2) 3.843 Hemoglobin (continuous variable) 4.607 White blood cells (continuous variable) 2.730 Histology (adenocarcinoma vs. squamous vs. unclassi4.448 fied vs. other) Stage (IIIB vs. IV) 11.413 Type of treatment (GP vs. GC) 0.0007 Weekly dose intensity for platinum compounds (contin- 1.821 uous variable) Weekly dose intensity for gemcitabine (continuous 1.294 variable) Treatment response (PR vs. SD vs. PD) 25.111 Second-line chemotherapy (yes vs. no) 10.357
Time to disease progression dof P value
Wald test (x2)
dof P value
1 1 1 1 1 3
0.577 0.061 0.049 0.032 0.098 0.217
1.431 1.815 3.875 5.544 0.004 5.197
1 1 1 1 1 3
0.231 0.177 0.586 0.018 0.944 0.157
1 1 1
0.0007 0.980 0.177
8.689 0.988 2.335
1 1 1
0.003 0.320 0.126
1
0.255
0.542
1
0.461
B/0.0001 66.499 0.001 /
2 /
2 1
B/0.0001 /
dof, Degrees of freedom; ECOG, Eastern Cooperative Oncology Group; GP, gemcitabine/cisplatin; GC, gemcitabine/carboplatin; PR, partial response; SD, stable disease; PD, partial disease.
88
4. Discussion Cisplatin-based chemotherapy is a reasonable treatment option for patients with advanced NSCLC. Old cisplatin-based regimens have given way to those using newer third generation chemotherapy (such as gemcitabine, paclitaxel, vinorelbine) that have demonstrated higher activity and better tolerability and palliation capacity [28]. None of these newer regimens, however, have clearly demonstrated superior efficacy in phase III trials [14,15]. Therefore, tolerability has become an important consideration when making treatment decisions. In our trial in advanced NSCLC, there were no statistical differences between GC and GP in terms of efficacy. We obtained similar response rates in the two arms in stage IIIB patients and a slight, but not statistically significant, advantage for stage IV patients in the GP arm (38.9 vs. 22.2%, P /0.20). The similarity of the two regimens was underscored by the results of the Cox’s regression analysis, which did not identify the type of treatment as a prognostic factor for overall survival (P /0.98). The efficacy results obtained with GC are consistent with those reported in previous phase I/II trials applying the same regimen [18]. Of interest, the overall response rate of 31%, the median survival time of 10.8 months and the 1-year survival rate of 43% obtained with the GC regimen in this study compared well to those reported for other active third-generation regimens [14 /16]. The importance of clinical factors as prognostic indicators is generally well known and should not be overlooked when making treatment choices for patients with NSCLC. Through the multivariate analysis, we found that a better survival outcome was predicted for patients with stage IIIB disease as opposed to stage IV disease; however, the outcome for patients with stage IIIB disease was strictly dependent on the patient’s opportunity to receive a radical resection of down-staged pulmonary disease or external-beam chest radiotherapy after chemotherapy. Our analysis further indicated that second-line chemotherapy was an independent prognostic factor for improved survival in patients with advanced NSCLC. For time to disease progression, stage IIIB disease, an ECOG PS of 0 or 1, high baseline hemoglobin and response to treatment were identified as having favorable independent prognostic potential. We observed grade 3/4 thrombocytopenia in 16.1% of patients and 6.4% of cycles on the GP arm; however, no patients on this arm required platelet transfusions. These results compare well with Spanish [10] and Italian [13] trials employing
P. Mazzanti et al.
the same 3-week schedule. The incidence of grade 3/4 thrombocytopenia on our GC arm (13.8% of patients and 4.5% of cycles) did not differ significantly from that of the GP arm. The use of growth factors was very limited. Better overall tolerability was seen in the GC arm for nausea/vomiting, renal and neurological toxicity (all grades combined). Compared with previous studies applying the same regimens, our study had a lower overall incidence of toxic events. In addition, hematological toxicity had a minimal effect on the day-8 delivered dose intensity of gemcitabine, only affecting :/10% of the doses missed for both arms combined. One explanation for this could be our choice of regimen (both the 3-week schedule and the administration of platinum-compounds on day 2). An assessment of the data from previous trials revealed that an early administration of cisplatin is associated with a low incidence of neutropenia [29], a finding that is supported in the current study, together with a low incidence of thrombocytopenia.
5. Conclusion In this multicenter randomized phase II trial, GP and GC did not provide any significant difference in response rate in patients with advanced NSCLC, with GC exhibiting better overall tolerability than GP. The potential therapeutic advantages of GC in terms of overall tolerability as well as ease of administration, make this novel combination a valid alternative in the palliative treatment especially for the ‘unfit’ patients.
Acknowledgements We are grateful to Mary Alice Miller, PhD, Rachel Shemer, Scientific Communications Consultants and Noelle Gasco, Scientific Communications Associate, of Eli Lilly and Co. for general preparation of this manuscript. We also thank Dr Filippo de Braud, Director of New Drugs Unit and Dr Tommaso De Pas, Vice-Director of Medical Oncology, of European Institute of Oncology for critically reviewing the manuscript.
References [1] Rapp E, Pater JL, Willan A, et al. Chemotherapy can prolong survival in patients with advanced non-small-cell lung cancer: report of a Canadian multicenter randomized trial. J Clin Oncol 1988;6:633 /41. [2] Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis
Immunohistochemical expression of integrins and ECM proteins in NSCLC
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
using updated data on individual patients from 52 randomized clinical trials. Br Med J 2000;311:899 /909. Bunn PA, Jr, Kelly K. New chemotherapeutic agents prolong survival and improve quality of life in non-small cell lung cancer: a review of the literature and future directions. Clin Cancer Res 1998;4:1087 /2100. Sandler A, Ettinger DS. Gemcitabine: single-agent and combination therapy in non-small cell lung cancer. Oncologist 1999;4:241 /51. Bergman AM, Ruiz van Haperen VW, Veerman G, et al. Synergistic interaction between cisplatin and gemcitabine in vitro. Clin Cancer Res 1996;2:521 /30. Crino ` L, Scagliotti G, Marangolo M, et al. Cisplatin / gemcitabine combination in non-small-cell lung cancer. J Clin Oncol 1997;15:297 /303. Abratt RP, Bezwoda WR, Goedhals L, et al. Weekly gemcitabine with monthly cisplatin: effective chemotherapy for advanced non-small-cell lung cancer. J Clin Oncol 1997;15:744 /9. Sandler AB, Ansari R, McClean J, et al. A Hoosier Oncology Group phase II study of gemcitabine plus cisplatin in nonsmall cell lung cancer. Cancer Therapeut 1998;1:158 /63. Anton A, Diaz-Fernandez N, Gonzalez Larriba JL, et al. Phase II trial assessing the combination of gemcitabine and cisplatin in advanced non-small cell lung cancer (NSCLC). Lung Cancer 1998;22:139 /48. Castellano D, Lianes P, Paz-Ares L, et al. A phase II study of a novel gemcitabine plus cisplatin regimen administered every three weeks for advanced non-small-cell lung cancer. Ann Oncol 1998;9:457 /9. Soto Parra HJ, Cavina R, Antonelli G, et al. Superiority of three-week vs four-week schedule of cisplatin (CDDP) and gemcitabine (GEM): results of a randomized phase II study. Proc Am Soc Clin Oncol 2000;19:546a (abstract 2152). Cardenal F, Lopez-Cabrerizo MP, Anton A, et al. Randomized phase III study of gemcitabine-cisplatin versus etoposide-cisplatin in the treatment of locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 1999;17:12 /8. Crino ` L, Scagliotti GV, Ricci S, et al. Gemcitabine and cisplatin versus mitomycin, ifosfamide, and cisplatin in advanced non-small-cell lung cancer: a randomized phase III study of the Italian Lung Cancer Project. J Clin Oncol 1999;17:3522 /30. Scagliotti GV, De Marinis F, Rinaldi M, et al. Phase III randomized trial comparing three platinum-based doublets in advanced non-small cell lung cancer. J Clin Oncol 2002;20(21):4285 /91. Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. New Engl J Med 2002;346:92 /8.
.
89
[16] Klastersky J, Sculier JP, Lacroix H, et al. A randomized study comparing cisplatin or carboplatin with etoposide in patients with advanced non-small-cell lung cancer: European Organization for Research and Treatment of Cancer Protocol 07861. J Clin Oncol 1990;8:1556 /62. [17] Rosell R, Gatzeimer U, Betticher DC, et al. Phase III randomized trial comparing paclitaxel/carboplatin with paclitaxel/cisplatin in patients with advanced non-smallcell lung cancer: a cooperative multinational trial. Ann Oncol 2002;13:1539 /49. [18] Langer CJ, Gandara DR, Calvert P, et al. Gemcitabine and carboplatin in combination: an update of phase I and phase II studies in non-small cell lung cancer. Semin Oncol 1999;26(Suppl 4):12 /8. [19] Iaffaioli RV, Tortoriello A, Facchini G, et al. Phase I /II study of gemcitabine and carboplatin in stage IIIB /IV non-smallcell lung cancer. J Clin Oncol 1999;17:921 /6. [20] Ng EW, Sandler AB, Robinson L, Einhorn LH. A phase II study of carboplatin plus gemcitabine in advanced non-small-cell lung cancer (NSCLC): a Hoosier oncology group study. Am J Clin Oncol 1999;22:550 /3. [21] Gajra A, Mehdi SA, Kirshner J, et al. Phase I study of gemcitabine and carboplatin in advanced non-small cell lung carcinoma. Lung Cancer 2001;32:189 /96. [22] Carrato A, Garcia-Gomez J, Alberola V, et al. Carboplatin (CARBO) in combination with gemcitabine (GEM) in advanced non-small cell lung cancer (NSCLC). Comparison of two consecutive phase II trials using different schedules. Proc Am Soc Clin Oncol 1999;18:498a (Abstract 1922). [23] Masotti A, Zannini G, Gentile A, Morandini G. Activity of gemcitabine and carboplatin in advanced non-small cell lung cancer: a phase II trial. Lung Cancer 2002;36:99 /103. [24] Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 1975;31:103 /15. [25] Mazzanti P, Delprete S, Battelli N, et al. Combination of gemcitabine and cisplatin in advanced non-small cell lung cancer (NSCLC): a phase II study. Proc Am Soc Clin Oncol 1999;18:517a (Abstract 1993). [26] Calvert AH, Newell DR, Gumbrell LA, et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989;7:1748 /56. [27] Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981;47:207 /14. [28] Gebbia V, Galetta D, Riccardi F, et al. Vinorelbine plus cisplatin versus cisplatin plus vindesine and mitomycin C in stage IIIB /IV non-small cell lung carcinoma: a prospective randomized study. Lung Cancer 2002;37:179 /87. [29] Shepherd FA, Abratt R, Crino ` L, et al. The influence of gemcitabine and cisplatin schedule on response and survival in advanced non-small cell lung cancer. Lung Cancer 2000;30:117 /25.