Original article. Randomized phase II trial of gemcitabine cisplatin with or without trastuzumab in HER2-positive non-small-cell lung cancer

Transcription

1 Original article Annals of Oncology 15: 19 27, 2004 DOI: /annonc/mdh031 Randomized phase II trial of gemcitabine cisplatin with or without trastuzumab in HER2-positive non-small-cell lung cancer U. Gatzemeier 1 *, G. Groth 1, C. Butts 2, N. Van Zandwijk 3, F. Shepherd 4, A. Ardizzoni 5, C. Barton 6, P. Ghahramani 6 & V. Hirsh 7 1 Krankenhaus Grosshansdorf, Grosshansdorf, Germany; 2 Cross Cancer Institute, Edmonton, Canada; 3 The Netherlands Cancer Institute, Amsterdam, The Netherlands; 4 Princess Margaret Hospital, Toronto, Canada; 5 Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy; 6 Roche Products Ltd, Welwyn Garden City, UK; 7 McGill University, Montreal, Canada Received 29 May 2003; revised 4 September 2003; accepted 30 September 2003 Introduction Background: Trastuzumab provides significant clinical benefits in HER2-positive metastatic breast cancer patients when administered in combination with chemotherapy. Chemotherapy has also been shown to be beneficial in some patients with advanced non-small-cell lung cancer (NSCLC). The present randomized phase II trial examined the effect of adding trastuzumab to a standard chemotherapeutic combination (gemcitabine cisplatin) in patients with HER2-positive NSCLC. Patients and methods: Patients with untreated stage IIIB/IV HER2-positive NSCLC received up to six 21-day cycles of gemcitabine 1250 mg/m 2 (days 1 and 8) and cisplatin 75 mg/m 2 (day 1). Patients in the trastuzumab arm received trastuzumab 4 mg/kg intravenously (i.v.) followed by 2 mg/kg/week i.v. until progression. Results: Of 619 patients screened, 103 were eligible. Fifty-one patients were treated with trastuzumab plus gemcitabine cisplatin and 50 with gemcitabine cisplatin alone. Efficacy was similar in the trastuzumab and control arms: response rate 36% versus 41%; median time to progression 6.3 versus 7.2 months; and median progression-free survival (PFS) 6.1 versus 7 months. Response rate (83%) and median PFS (8.5 months) appeared relatively good in the six trastuzumab-treated patients with HER2 3+ or fluorescence in situ hybridization (FISH)-positive NSCLC. Addition of trastuzumab to gemcitabine cisplatin was well tolerated, sideeffects were as expected, and trastuzumab did not exacerbate the known toxicity of gemcitabine and cisplatin. Symptomatic cardiotoxicity was observed in one trastuzumab-treated patient. Serum trastuzumab concentrations in the presence of gemcitabine cisplatin were comparable to those of trastuzumab alone. Conclusions: Trastuzumab plus gemcitabine cisplatin is well tolerated. Clinical benefit was not observed. Although HER2 3+/FISH-positive patients may benefit from trastuzumab, the subgroup is too small to provide definitive information. No significant effect of gemcitabine cisplatin on trastuzumab pharmacokinetics was observed. Key words: cisplatin, gemcitabine, HER2, non-small-cell lung cancer, trastuzumab *Correspondence to: Dr Ulrich Gatzemeier, Krankenhaus Grosshansdorf, Zentrum für Pneumologie und Thoraxchirurgie, Wöhrendamm 80, Grosshansdorf, Germany. Tel: ; Fax: ; u.gatzemeier@t-online.de Lung cancer is the most common cancer and the major cause of cancer-related death in North America and Europe [1, 2]. Fiveyear survival rates are <10% [2]. Non-small-cell lung cancer (NSCLC) represents 75% of lung cancers and survival of patients with metastatic NSCLC is poor, with a median of 7 8 months [3]. The poor prognosis of patients with NSCLC despite current therapy and the significant toxicity of combination therapies heighten the need for new therapeutic approaches. Trastuzumab (Herceptin ) is a humanized monoclonal antibody rationally developed to target HER2-overexpressing cells. Trastuzumab has additive or synergic antitumor activity in combination with various cytotoxic agents in preclinical models of breast cancer and NSCLC [4 6]. In the study by Bunn et al. [5], the synergy of trastuzumab with cisplatin or gemcitabine was greater in HER2-positive NSCLC cell lines than it was in breast cancer cell lines. In breast cancer, preclinical observations such as these have been used to identify regimens for clinical study. Such clinical studies have shown that the use of trastuzumab in combination with chemotherapy provides significant clinical benefits, including a significant increase in survival, in patients with HER2- positive metastatic breast cancer [7]. The feasibility of combining trastuzumab with platinum analogs and gemcitabine has also been demonstrated in patients with HER2-positive breast cancer [8, 9]. Several studies have shown that women with metastatic breast cancer that is HER2 3+ by immunohistochemistry (IHC) or shows HER2 gene amplification by fluorescence in situ hybridization (FISH) obtain the greatest benefit [7, 10, 11] European Society for Medical Oncology

2 20 HER2 overexpression has been reported in up to 59% of NSCLCs, and the 2+/3+ overexpression rate is 5 20% (up to 30% in adenocarcinomas) [12 17]. Such HER2 overexpression is an indicator of poor prognosis [18, 19] and has been shown to contribute to tumorigenesis in lung tumor cell lines [5, 20]. Trastuzumab causes cytotoxicity in HER2-positive NSCLC cell lines [5, 20]. These observations create a rationale for targeting HER2 in patients with HER2-positive NSCLC. However, HER2 overexpression in NSCLC differs from that in breast cancer. In NSCLC, the majority of tumors are IHC 2+ and FISH positivity is uncommon [16, 17]. When this trial was designed, it was not known whether the efficacy of trastuzumab in NSCLC would be limited to the relatively small subgroup of patients with IHC 3+/FISHpositive disease, or whether patients with lower-level overexpression would also benefit. Therefore, HER2 positivity was defined as IHC 2+/3+ overexpression or HER2 amplification or elevated circulating serum HER2 extracellular domain (ECD) in patients with no tissue available. Gemcitabine cisplatin combination therapy is commonly used in Europe and has efficacy superior to that of cisplatin monotherapy and other combinations [21, 22]. The known efficacy of trastuzumab in HER2-positive breast cancer and of the gemcitabine cisplatin combination in NSCLC, the preclinical synergy between trastuzumab and gemcitabine and cisplatin [5], and the differing toxicity of trastuzumab and gemcitabine cisplatin provide a clear rationale for investigating trastuzumab in combination with gemcitabine cisplatin in HER2-positive NSCLC. This randomized, open-label, phase II study was designed to assess the efficacy and safety of trastuzumab with gemcitabine cisplatin compared with gemcitabine cisplatin alone in patients with HER2-positive NSCLC. The study also aimed to collect information on the pharmacokinetics of trastuzumab in the presence of gemcitabine cisplatin. Patients and methods Study design This was a randomized, open-label, phase II study with two arms. The primary study objective was to evaluate objective response rate (complete response plus partial response). Secondary objectives were to investigate the tolerability and pharmacokinetics of the combination, and to compare progression-free survival (PFS) in the two arms. Patients Patients had to fulfill the following criteria: age 18 years with histologically or cytologically proven stage IIIB or IV NSCLC; Eastern Cooperative Oncology Group (ECOG) performance status 0 2; at least one bidimensionally measurable lesion; ability to comply with the protocol; and recovered from any surgical procedure(s) as judged by the investigator. Only patients who were suitable candidates for chemotherapy alone were recruited. Exclusion criteria included dyspnea at rest or requirement for supportive oxygen therapy; CNS metastases (trastuzumab is known not to penetrate the blood brain barrier) or invasive malignancies other than NSCLC; pregnancy; any investigational drug within 30 days before treatment start; prior chemotherapy or anti-her2 therapy; radiotherapy within 3 weeks of study start; or prior radiotherapy to indicator lesions unless objective recurrence or progression of these lesions was documented. Patients with New York Heart Association (NYHA) class III/IV congestive heart failure (CHF), known left ventricular ejection fraction (LVEF) <40% by multigated radionucleotide angiography (MUGA) or echocardiography (Echo), or myocardial infarction within 6 months of treatment were also excluded, as were patients with abnormal baseline laboratory values: hemoglobin <10 g/dl; neutrophils < /l; platelets < /l; serum total bilirubin >1.5 the upper limit of normal (ULN); alanine aminotransferase (ALT) or aspartate aminotransferase (AST) >2.5 ULN; alkaline phosphatase >2.5 ULN; and creatinine clearance <60 ml/min. The study was conducted in accordance with the Declaration of Helsinki and approved by the local ethics committees of the participating institutions. Written, informed consent was obtained from all patients. Patients fulfilling these general inclusion/exclusion criteria underwent HER2 testing as described below. Those with HER2-positive disease who were subsequently shown to have an LVEF >40% by MUGA or Echo were enrolled. HER2 testing HER2 status was assessed by IHC using the HercepTest kit (DAKO, Glostrup, Denmark), FISH or enzyme-linked immunosorbent assay (ELISA) to measure shed ECD concentration. Tumor specimens were paraffin-embedded and fixed in neutral buffered formalin. Tissue blocks were sectioned at 4 5 µm. HER2 overexpression/amplification was defined as a score of 2+ (complete, moderate membrane staining of >10% of tumor cells) or 3+ (complete, intense membrane staining of at least 80% of tumor cells) by IHC or gene amplification (HER2/chromosome 17 ratio 2) by FISH (PathVysion HER-2 DNA Probe Kit; Vysis, Stuttgart, Germany). Serum HER2 ECD concentrations were measured only for patients with no tissue available for HER2 testing, and concentrations >15 ng/ml by ELISA were judged to be positive. HER2 status was determined both locally and at a central testing laboratory. A HER2-positive status determined locally was sufficient for trial enrollment, regardless of central test results. Patients with HER2-negative disease based on local testing could be enrolled if HER2 positive on central testing. Study treatment All patients received gemcitabine 1250 mg/m 2 intravenously (i.v.) over 30 min on days 1 and 8 of a 21-day cycle in combination with cisplatin 75 mg/m 2 i.v. over 1 h on day 1 with mannitol diuresis. Chemotherapy was scheduled for up to six cycles. Participating institutions used their standard administration regimen, including choice of anti-emetic agent and hydration. In the experimental arm, trastuzumab was administered before chemotherapy as a 4 mg/kg initial dose i.v. over 90 min on day 1, followed by 2 mg/kg i.v. over 30 min weekly until disease progression. Trastuzumab was administered alone following completion of chemotherapy. Gemcitabine and cisplatin doses were modified in response to non-hematological toxicities according to a predefined scheme. National Cancer Institute Common Toxicity Criteria (NCI CTC) grade 3 increases in bilirubin levels led to a 50% reduction in gemcitabine dose and grade 4 increases to gemcitabine being withheld. Creatinine clearance of 30 59% led to a 50% reduction in both gemcitabine and cisplatin dose, whereas clearance of <30% led to both drugs being withheld until recovery. For all other toxicities, gemcitabine and cisplatin dose was reduced by 50% for NCI CTC grade 2 toxicity and withheld for grade 3 or 4 toxicity. At the first occurrence of a grade 3 or 4 non-hematological toxicity, trastuzumab treatment was interrupted and resumed at the same dose after resolution to grade 2 or better. For hematological toxicities, chemotherapy dose modification was based on values from the end of each 21-day cycle. In patients with an absolute neutrophil count (ANC) < /l or platelet count < /l, treatment was delayed for 1 week or until ANC and platelets had returned to values /l and /l, respectively. If the preceding cycle was accompanied by febrile neutropenia, thrombocytopenia with bleeding, or platelets < /l, doses were reduced to 80% of the

3 21 preceding dose and not increased thereafter. Trastuzumab was continued in the event of chemotherapy dose adjustment. Evaluation of response and toxicity Tumors were assessed according to World Health Organization (WHO) criteria at baseline, every 6 weeks thereafter, and within 2 weeks of study completion. Adverse events were graded using the NCI CTC, version 2.0. Serious adverse events were defined as any event that was fatal, life-threatening, required hospitalization or prolongation of existing hospitalization, or resulted in a persistent or significant disability/incapacity. Evaluation of cardiac function Cardiac function was monitored by Echo or MUGA scans at baseline, after cycles 4, 8 and 16, and every eight cycles thereafter. Patients experiencing a decrease in LVEF of 15%, or an absolute value of <30% during the study, were withdrawn. Pharmacokinetic assessments In patients in the trastuzumab arm, pre-dose and end of infusion samples (5 ml) were taken to measure serum trastuzumab at baseline and in cycles 1 6, 8 and 14. In addition, pre- and post-dose blood samples were taken on day 8 of the specified cycles. In a separate subprotocol, pharmacokinetic samples were taken more frequently from 10 selected patients in each of three centers. Serial blood samples were collected on day 1 of cycles 1 and 4. These samples were collected pre-dose, immediately after the end of infusion, and at 4, 8, 12, 24, 48, 96 and 168 h thereafter. Additional pre- and post-dose samples were collected on days 1, 8 and 15 of cycles 1 3. Serum samples were prepared and analyzed using a validated ELISA assay (Genentech, San Francisco, CA, USA) (data on file). Assessment of ECD Serum HER2 ECD concentration was measured at baseline and on day 8 of cycles 1, 4, 8 and 14. Blood samples were taken and serum samples prepared and analyzed using a validated ELISA assay (data on file). Statistical analysis Sample size was calculated to ensure that the lower one-sided confidence limit would be 0.7% (i.e. above 0%, indicating a difference) if response rates of 10% and 24% were obtained for the control and trastuzumab arms, respectively. Assumptions about response rates were based on observations that patients with HER2-positive breast cancer have lower response rates than the overall population of patients with breast cancer. Thus, although response rates to gemcitabine plus cisplatin are generally in the range 20 40% for unselected patients with NSCLC [21, 22], the response rate in patients with HER2-positive disease was assumed to be lower. The magnitude of benefit for the addition of trastuzumab was expected to be large, as seen in the pivotal randomized study of chemotherapy with or without trastuzumab in patients with HER2-positive breast cancer [7]. The planned sample size should provide a good indication of whether trastuzumab provides a similar level of benefit. Final analysis was performed when all patients had completed at least 1 year of treatment or had withdrawn. For analysis of efficacy, the best response rate was reported with 95% Pearson Clopper confidence intervals (CIs). Two separate analysis sets were used: safety analysis set comprising all patients who were randomized and received at least one dose of study medication (groups defined by actual medication received); and the full analysis set (intention-to-treat population) comprising all randomized patients. The primary focus was on descriptive statistics such as rates and two-sided 95% CIs. Significance level or type I error probability was set to 5%. Results A total of 619 patients met the initial eligibility criteria and were screened for HER2 status; 103 patients (17%) were enrolled into the study between August 1999 and August Detailed results of the HER2 screening program will be reported elsewhere. Two patients did not receive study drug, 50 were randomized to receive gemcitabine cisplatin and 51 to trastuzumab plus gemcitabine cisplatin (Figure 1). One patient randomized to gemcitabine cisplatin received trastuzumab and is included in the gemcitabine cisplatin arm for efficacy analysis (as randomized) but in the trastuzumab arm for safety analysis. One patient with stage IB disease was treated and is included in the analyses. Patient demographics and HER2 status are shown in Table 1. The median number of cycles of chemotherapy received was six, and 60% of patients completed six cycles of chemotherapy. Patients in the trastuzumab arm also received a median of six cycles (18 weekly doses) (range 1 29 cycles or 2 85 doses) of trastuzumab (25 of 51 patients continued trastuzumab after completion of chemotherapy). Overall, only six patients completed 1 year of therapy (Figure 2). Efficacy Investigator-assessed response rates were similar in the gemcitabine cisplatin and trastuzumab arms (41% and 36%, respectively; P = not significant). Efficacy data are summarized in Table 2. Median TTP and median PFS were also similar between the two arms (Table 2; Figure 2). Of the eight patients with FISH-positive or IHC 3+ disease, two were randomized to receive gemcitabine cisplatin alone and six to receive trastuzumab plus gemcitabine cisplatin. There was one partial response in the control arm and five in the trastuzumab arm. In the trastuzumab arm, median PFS was 8.5 (range ) months in this small subgroup of patients with IHC 3+/FISHpositive disease, compared with 6.1 (range ) months in the group as a whole. No additional benefit of adding trastuzumab to gemcitabine cisplatin was detected in patients who had HER2- positive disease based on high serum HER2 ECD levels only. Safety Treatment was well tolerated in both arms and toxicity profiles were similar (Figure 3). Gastrointestinal toxicities were the most frequent adverse events, with nausea and vomiting most common in both treatment groups. However, there was a higher incidence of fatigue in the trastuzumab arm than in the gemcitabine cisplatin arm. Hematological toxicities were well balanced between the two treatment groups, with grade 3/4 neutropenia (58% of patients in the gemcitabine cisplatin arm versus 57% in the trastuzumab arm), thrombocytopenia (34% versus 35%) and leukopenia (36% versus 33%) being the most common events. Grade 3/4 anemia was observed in 12% and 16% of patients, respectively. These results indicate that the major cause of toxicity was the chemotherapy regimen.

4 22 Figure 1. Patient disposition. Cardiac safety Baseline median LVEF was 64% (range 40 82%) in the control arm and 61% (range 41 83%) in the trastuzumab arm. During the study, 52% and 58% of patients in the control and trastuzumab arms, respectively, experienced at least one decrease in LVEF of <15% on at least one occasion (Table 3). No patient in the control arm developed a decrease in LVEF of >15% or symptomatic CHF. Decreases in LVEF of >15% occurred in three patients in the trastuzumab arm, one of whom developed symptomatic CHF (NYHA class IV). In addition, one other patient stopped trastuzumab treatment due to a decrease in LVEF of 14% during therapy (mild cough and dyspnea were noted but it was not clear whether these were pre-existing), and one patient developed CHF (diagnosed clinically; no LVEF data reported) several weeks after stopping trastuzumab. Details of these patients are summarized in Table 4. Therefore, two or possibly three patients treated with trastuzumab developed symptomatic CHF during this study, an incidence of 4 6%. Pharmacokinetics Serum trastuzumab levels accumulated over time: a three-fold increase in mean pre-dose concentrations (29.7 µg/ml increasing to 90.7 µg/ml) was seen from cycle 2 to cycle 15, as expected (Figure 4). Mean post-dose concentrations increased by 30% from cycle 1 to cycle 9. Trastuzumab concentrations appeared to approach steady-state at around cycle 9, although the small number of patients completing cycle 15 (n = 4) makes this assessment inconclusive. When pre-dose serum concentrations of trastuzumab over 15 cycles in patients receiving trastuzumab in combination with gemcitabine cisplatin were compared with those of trastuzumab alone (using data from Cobleigh et al. [10]), there was no evidence to suggest a significant difference between the serum trough concentrations of trastuzumab. Furthermore, mean exposure between 0 and 42 weeks (AUC 0 42w ) was µg week/ml for trastuzumab in combination with chemotherapy, compared with µg week/ml for trastuzumab alone. ECD ECD levels tended to be low (median baseline level 9.5 ng/ml, range 5 50). Of the total of 103 patients for whom baseline ECD levels were available, only 19 had levels exceeding the 15 ng/ml positivity threshold (19.5 ng/ml, range 15 50). It is possible that this was related to the relatively low level of HER2 overexpression, with very few patients having HER2 3+ overexpression, in

5 23 Table 1. Patient demographics (safety population) this study. The relatively small number of patients and the homogeneity of ECD data do not permit a meaningful analysis of the relationship between ECD levels and baseline disease characteristics or outcome in this study. An analysis of pooled data from several studies may be warranted to examine the relationship between ECD and serum trastuzumab levels. Discussion Parameter Gemcitabine cisplatin Trastuzumab plus gemcitabine cisplatin No. of patients (n) Male/female 30/20 33/18 Median age [years (range)] 60 (35 76) 57 (35 76) HER2 status [patients (%)] FISH positive a 2 (4) 5 (10) IHC 3+ a 1 (2) 4 (8) Other (IHC 0 2+/shed ECD positive) 48 (96) 45 (88) Disease stage [patients (%)] IB 0 (0) 1 (2) IIIB 6 (12) 9 (18) With effusion 5 (10) 3 (6) IV 39 (78) 38 (74) Metastases (%) Median no. of sites 4 4 Regional nodes/metastatic nodes 24/20 26/21 Prior radiotherapy (%) 9 (18) 10 (20) Prior surgery (%) 26 (52) 25 (49) a Based on central laboratory testing results. Three of four patients with IHC 3+ disease in the trastuzumab plus gemcitabine cisplatin arm were also FISH positive; the patient with IHC 3+ disease in the gemcitabine cisplatin arm was FISH positive. Therefore, a total of eight patients had IHC 3+/ FISH positive disease. Figure 2. Progression-free survival with gemcitabine cisplatin (dashed line) and trastuzumab plus gemcitabine cisplatin (solid line). The poor prognosis of NSCLC and the lack of effective, welltolerated treatments have led to trials involving new agents and combinations [23]. One of these agents, trastuzumab, is effective in patients with HER2-positive breast cancer [7] and has the potential to be used to treat patients with HER2-positive NSCLC, which is also known to be aggressive [18, 19]. The documented improvement in survival observed when trastuzumab is added to chemotherapy in metastatic breast cancer [7], preclinical synergy between trastuzumab and various chemotherapeutic agents in NSCLC cell lines [5, 6], and the favorable tolerability profile of trastuzumab [10] provide a rationale for studying trastuzumab in NSCLC. This study evaluated the efficacy and safety of trastuzumab in combination with gemcitabine cisplatin compared with that of gemcitabine cisplatin alone. The majority of patients (>90%) in this study had HER2 2+ NSCLC as determined by IHC. In this population, the addition of trastuzumab to gemcitabine cisplatin did not appear to provide any benefit, and the response rate to gemcitabine cisplatin therapy was within the expected range based on data from prior trials in patient populations unselected for HER2 status [21, 22]. However, there is increasing evidence to demonstrate that those patients with breast cancer with very strong HER2 overexpression (IHC 3+) or HER2 gene amplification (FISH positive) respond best to trastuzumab [10, 24]. In this study, few patients had HER2 3+/FISH-positive disease, although those treated with trastuzumab had a relatively high response rate (five of six patients responded) and PFS (8.5 versus 6.1 months in the group of patients with IHC 2+ NSCLC) compared with other patients in the trial. However, this can only be considered to be an interesting observation. Further evaluation in a larger patient population

6 24 Table 2. Efficacy data (intention-to-treat population) a Response data not available for one and three patients, respectively, due to drop out prior to initial efficacy analysis. These patients were recorded as not responding. b The P value for overall response rate was not significant. c Median overall survival duration has not been reached in the gemcitabine cisplatin arm. CI, confidence interval; TTP, time to progression; PFS, progression-free survival. would be desirable, but <2% of the patients screened for this trial had IHC 3+ or FISH-positive disease, and ongoing trials have either not reported the proportions of 2+ and 3+ patients or have similarly small numbers [25 27]. Tran et al. [26], using the regimen and doses described here, reported a response rate of 42% and disease stabilization in a further 42% in HER2-positive patients (IHC 1+ and/or 15 ng/ml serum HER2 ECD). The proportion and outcomes of patients with IHC 3+ NSCLC were not reported. Trastuzumab has also shown promise in combination with paclitaxel and carboplatin, producing a response rate of 18% at 6-month follow up in an ongoing ECOG trial [25]. Although this response rate does not appear to be higher than would be expected with chemotherapy alone, it is interesting that eight of 10 patients (80%) with IHC 3+ disease on study treatment were still alive at 6-month follow-up, compared with 64% of the overall population. The mature results of this trial will help to identify the benefit of combining trastuzumab with chemotherapy for patients with IHC 3+ disease. However, overall the data suggest that any benefit of trastuzumab in NSCLC will be confined to a group of patients (IHC 3+ or FISH positive) likely to constitute <5% of the Gemcitabine cisplatin a No. of patients (n) Trastuzumab plus gemcitabine cisplatin a Overall response rate [% (95% CI)] 41 ( ) b 36 ( ) b Complete response (%) 1 (2) 2 (4) Partial response (%) 20 (39) 16 (32) Stable disease (%) 27 (53) 22 (44) Progressive disease (%) 2 (4) 7 (14) Median TTP (months) Median PFS (months) (95% CI) 7.0 ( ) 6.1 ( ) Median overall survival [months (range)] ( ) c 12.2 ( ) Table 3. Changes in cardiac function LVEF increase or no change (%) + H LVEF, left ventricular ejection fraction;, gemcitabine cisplatin; H, trastuzumab. LVEF decrease <15% LVEF decrease >15% LVEF <30% + H + H Cycle 4 (n = 32, 36) Cycle 8 (n = 12, 20) Worst value (n = 33, 36) H Figure 3. Incidence of treatment-related adverse events (all grades) occurring in >10% of patients treated with gemcitabine cisplatin (open bars) or gemcitabine cisplatin plus trastuzumab (filled bars).

7 Table 4. Outcome in patients with LVEF decrease 15% and/or leading to treatment withdrawal and/or symptomatic congestive heart failure (CHF) Patient no. Cycle LVEF (%) Comment Relevant medical history Outcome Value Change from baseline 25040/0661 Baseline 72 Discontinued due to cardiac failure (with pulmonary edema, requiring ventilation) Follow-up (after 3 months) /0506 Baseline 49 Trastuzumab discontinued due to LVEF decrease at cycle 4 (with mild cough and dyspnea); chemotherapy continued Follow-up (4 months later) /904 Baseline 77 Discontinued due to (asymptomatic) LVEF decrease Follow-up (2 weeks later) 69 8 Follow-up (3 weeks later, 5 weeks after stopping) /0905 Baseline 53 Discontinued due to (asymptomatic) LVEF decrease Follow-up (6 weeks later) /1023 Baseline 60 CHF diagnosed clinically (no LVEF decrease), in association with atrial fibrillation and tumor progression in the right lung, 13 weeks after stopping trastuzumab and chemotherapy Follow-up (13 weeks post trastuzumab) Unknown Pneumonectomy Atherosclerosis, previous MI, cardiac dysrhythmias (receiving amiodarone and acebutalol at baseline), pericardial effusion Pericarditis (not active at baseline) MI and dyspnea, receiving diltiazem Hypertension, type II diabetes mellitus, alcohol abuse, obesity Improved with treatment Unchanged Unchanged Unchanged Improved LVEF, left ventricular ejection fraction; MI, myocardial infarction. 25

8 26 Figure 4. Comparison of mean (SD) pre-dose serum trastuzumab concentrations in combination with cisplatin and gemcitabine (open squares) compared with historical data for trastuzumab administered alone (filled squares) (1 cycle = 1 week). total population of patients with advanced NSCLC. This will make recruitment to future trials problematic and the likely clinical utility of trastuzumab in NSCLC limited. Pharmacokinetic analysis of trastuzumab in combination with gemcitabine cisplatin demonstrated that trastuzumab serum concentrations rise steadily from cycle 1, potentially achieving steady state concentration at around cycle 9. Data from beyond cycle 9 are not reliable because of the small number of samples (n = 4). Comparison of concentrations over time in this study with that in the pivotal monotherapy study suggests that co-administration of gemcitabine and cisplatin does not significantly affect exposure to trastuzumab [10]. It is notable that the trastuzumab-containing combination was generally as well tolerated as chemotherapy alone, and that trastuzumab did not appear to exacerbate the toxicity of gemcitabine cisplatin. This supports findings from trials in breast cancer and other tumor types, where trastuzumab has demonstrated favorable tolerability both as monotherapy and in combination with chemotherapy [7, 10, 28, 29]. The absence of tumor-specific trastuzumabassociated adverse events is encouraging for the development of studies in other indications in which a proportion of tumors are strongly HER2 positive. The most clinically significant adverse event that is known to be associated with trastuzumab therapy is cardiac toxicity. This was initially observed in a retrospective analysis [30] and has since been observed in other trials, but at a lower rate [31, 32]. Data from trials in breast cancer suggest that trastuzumab-related cardiac toxicity occurs primarily when the drug is used in combination with anthracyclines and tends to be both manageable and reversible [33]. In this trial in patients with NSCLC, only patients in the trastuzumab arm experienced large decreases in LVEF (>15%), decreases to an absolute LVEF value of <40% or CHF. However, symptomatic CHF occurred in only two or possibly three patients, and there were other significant factors contributing to these patients symptoms (previous pneumonectomy; previous myocardial infarction, cardiac dysrhythmias and pericardial effusion; progressive lung cancer and new onset atrial fibrillation). No correlation was observed between serum trastuzumab concentrations and cardiotoxicity. These cardiac toxicity data are important for several reasons. First, this population was not previously or concurrently exposed to anthracyclines, which are known to cause cardiotoxicity and to increase the likelihood of trastuzumab-associated cardiotoxicity [30, 34]. Because neither cisplatin nor gemcitabine is particularly associated with cardiotoxicity, these data provide an indication of the cardiac effects of the addition of trastuzumab to relatively noncardiotoxic drugs in patients not exposed to anthracyclines. However, it should be noted that although not anthracycline exposed, the population is likely to be at increased risk for cardiotoxicity due to the high percentage of smokers enrolled. This may explain some of the decrease in cardiac function observed in the control arm. Furthermore, the inclusion criteria allowed patients with an LVEF of only 40% at baseline to be enrolled. In most trials of trastuzumab to date, an LVEF of at least 50% has been stipulated. Current recommendations state that a careful risk benefit assessment needs to be made in patients with a baseline LVEF of <50%, because such patients are at increased risk of cardiac events [33]. Finally, this is one of the few trastuzumab studies to include men, and it appears that the effects on cardiac function are similar in men and women. Thus, although trastuzumab appears to have a negative effect on cardiac function in a population that is at increased risk due to the incidence of smoking and the inclusion of patients with LVEF <50%, even in the absence of anthracycline exposure, the effects are limited to relatively few patients, do not tend to be symptomatic, and are reversible with appropriate management. In conclusion, HER2 overexpression was seen in 17% of NSCLCs screened, although only <2% of the population screened showed IHC 3+ and/or HER2 amplification (as measured by FISH). The overall population did not appear to obtain any benefit from the addition of trastuzumab to gemcitabine cisplatin, and the small numbers of patients with strongly HER2-positive disease makes it difficult to assess whether it might improve efficacy in these patients. ECD levels do not appear to predict therapeutic response, although a pooled analysis of several studies is needed to provide more conclusive data. The addition of trastuzumab to gemcitabine cisplatin was well tolerated and the safety profile of the combination was as expected. Interestingly, cardiotoxicity, the most clinically significant adverse event associated with trastuzumab therapy, occurred infrequently (6%). Ongoing studies with other chemotherapeutic agents may further define the role of trastuzumab in NSCLC, but future trials should focus on patients with high HER2 overexpression or gene amplification. References 1. American Cancer Society. Cancer facts and figures Atlanta, GA: American Cancer Society Micheli A, Mugno E, Krogh V et al. Cancer prevalence in European registry areas. Ann Oncol 2002; 13: Schiller JH, Harrington D, Belani CP et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 2002; 346: Pegram M, Hsu S, Lewis G et al. Inhibitory effects of combinations of HER-2/neu antibody and chemotherapeutic agents used for treatment of human breast cancers. Oncogene 1998; 18: Bunn PA Jr, Helfrich B, Soriano AF et al. Expression of Her-2/neu in human lung cancer cell lines by immunohistochemistry and fluorescence in situ hybridization and its relationship to in vitro cytotoxicity by trastu-

9 27 zumab and chemotherapeutic agents. Clin Cancer Res 2001; 7: Hirsch FR, Helfrich B, Franklin WA et al. Preclinical studies of gemcitabine and trastuzumab in breast and lung cancer cell lines. Clin Breast Cancer 2002; 3 (Suppl 1): S12 S Slamon DJ, Leyland-Jones B, Shak S et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: Burstein HJ, Bunnell CA, Winer EP. New cytotoxic agents and schedules for advanced breast cancer. Semin Oncol 2001; 28: O Shaughnessy JA, Vukelja S, Marsland T et al. Gemcitabine and trastuzumab for HER-2 positive metastatic breast cancer: preliminary results of a phase II study. Breast Cancer Res Treat 2001; 69: 302 (Abstr 523). 10. Cobleigh MA, Vogel CL, Tripathy D et al. Multinational study of the efficacy and safety of humanized anti-her2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 1999; 17: Mass RD. The role of HER-2 expression in predicting response to therapy in breast cancer. Semin Oncol 2000; 27 (Suppl 11): Shi D, He G, Cao S et al. Overexpression of the c-erbb-2/neu-encoded p185 protein in primary lung cancer. Mol Carcinog 1992; 5: Harpole DH Jr, Marks JR, Richards WG et al. Localized adenocarcinoma of the lung: oncogene expression of erbb-2 and p53 in 150 patients. Clin Cancer Res 1995; 1: Salgia R, Skarin AT. Molecular abnormalities in lung cancer. J Clin Oncol 1998; 16: Scheurle D, Jahanzeb M, Aronsohn RS et al. HER-2/neu expression in archival non-small cell lung carcinomas using FDA-approved HercepTest. Anticancer Res 2000; 20: Hirsch FR, Franklin WA, Veve R et al. HER2/neu expression in malignant lung tumors. Semin Oncol 2002; 29 (Suppl 4): Cox G, Vyberg M, Melgaard B et al. Herceptest: HER2 expression and gene amplification in non-small cell lung cancer. Int J Cancer 2001; 92: Tateishi M, Ishida T, Mitsudomi T et al. Prognostic value of c-erbb-2 protein expression in human lung adenocarcinoma and squamous cell carcinoma. Eur J Cancer 1991; 27: Kern JA, Slebos RJ, Top B et al. C-erbB-2 expression and codon 12 K-ras mutations both predict shortened survival for patients with pulmonary adenocarcinomas. J Clin Invest 1994; 93: Noguchi M, Murakami M, Bennett W et al. Biological consequences of overexpression of a transfected c-erbb-2 gene in immortalized human bronchial epithelial cells. Cancer Res 1993; 53: Sandler AB, Nemunaitis J, Denham C et al. Phase III trial of gemcitabine plus cisplatin versus cisplatin alone in patients with locally advanced nonsmall-cell lung cancer. J Clin Oncol 2000; 18: Cardenal F, López-Cabrerizo MP, Antón 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: Dy GK, Adjei AA. Novel targets for lung cancer therapy: part 1. J Clin Oncol 2002; 20: Mass RD, Press M, Anderson S et al. Improved survival benefit from Herceptin (trastuzumab) in patients selected by fluorescence in situ hybridization (FISH). Proc Am Soc Clin Oncol 2001; 20: 22a (Abstr 85). 25. Langer CJ, Adak S, Thor A et al. Phase II Eastern Cooperative Oncology Group (ECOG) pilot study of paclitaxel (P), carboplatin (C), and trastuzumab (T) in HER-2/neu (+) advanced non-small cell lung cancer (NSCLC): early analysis of E2598. Proc Am Soc Clin Oncol 2001; 20: 315a (Abstr 1257). 26. Tran HT, Herbst RS, Glisson BS et al. Herceptin in combination with cisplatin and gemcitabine in patients (pts) with HER2 overexpressing, untreated, advanced, non-small-cell lung cancer (NSCLC): final report of a phase II trial. Proc Am Soc Clin Oncol 2002; 21: 307a (Abstr 1226). 27. Zinner R, Glisson BS, Pisters KM et al. Cisplatin and gemcitabine combined with Herceptin in patients (pts) with her2 overexpressing, untreated, advanced, non-small-cell lung cancer (NSCLC): a phase II trial. Eur J Cancer 2001; 37 (Suppl 6): S154 (Abstr 560). 28. Safran H, Ramanathan R, Schwartz J et al. Herceptin and gemcitabine for metastatic pancreatic cancers that overexpress HER-2/neu. Eur J Cancer 2001; 37 (Suppl 6): S310 (Abstr 1148). 29. Small EJ, Bok R, Reese DM et al. Docetaxel, estramustine, plus trastuzumab in patients with metastatic androgen-independent prostate cancer. Semin Oncol 2001; 28 (4 Suppl 15): Seidman A, Hudis C, Pierri MK et al. Cardiac dysfunction in the trastuzumab clinical trials experience. J Clin Oncol 2002; 20: Seidman AD, Fornier MN, Esteva FJ et al. Weekly trastuzumab and paclitaxel therapy for metastatic breast cancer with analysis of efficacy by HER2 immunophenotype and gene amplification. J Clin Oncol 2001; 19: Vogel CL, Cobleigh MA, Tripathy D et al. Efficacy and safety of trastuzumab as a single agent in first-line treatment of HER2-overexpressiong metastatic breast cancer. J Clin Oncol 2002; 20: Cook-Bruns N. Retrospective analysis of the safety of Herceptin immunotherapy in metastatic breast cancer. Oncology 2001; 61 (Suppl 2): Singal PK, Iliskovic N. Doxorubicin-induced cardiomyopathy. N Engl J Med 1998; 339: