© 1998 AlphaMed Press
Carboplatin in the Treatment of Small Cell Lung CancerThe Johns Hopkins Oncology Center Baltimore, Maryland, USA Correspondence: David S. Ettinger, M.D., F.A.C.P., Johns Hopkins Oncology Center, 600 North Wolfe Street, Baltimore, Maryland 21287-8936, USA. Telephone: 410-955-8847; Fax: 410-614-9424; e-mail: ettinger{at}welchlink.welch.jhu.edu
Small cell lung cancer (SCLC) will account for approximately 20%-25% of the 171,500 estimated new lung cancer cases in 1998. Combination cytotoxic therapies have yielded the best response rates in SCLC patients. Cisplatin in combination with etoposide is used routinely in the treatment of SCLC. Because of cisplatin's nonhematologic toxicities, carboplatin was developed and has far fewer nonhematologic toxicities. Carboplatin in combination with etoposide has been shown to be as effective as, but less toxic than, cisplatin/etoposide. Moreover, carboplatin/etoposide is a viable combination in the treatment of the elderly with SCLC, who can readily tolerate this combination. Concurrent radiation therapy with carboplatin and etoposide in patients with limited SCLC disease can be given safely and effectively. Carboplatin has been combined with several different chemotherapeutic agents, including ifosfamide, and, more recently, paclitaxel in hopes of improving the response rates and overall survival. In order to try to dose intensify carboplatin-based regimens, peripheral blood stem cells have been used to decrease the hematologic toxicities. Further studies are warranted to investigate these therapies as well as newer carboplatin combinations. Key Words. Carboplatin • Small cell lung cancer • Extensive disease • Limited disease
Lung cancer is the leading cause of death due to cancer in the United States [1]. Of new lung cancer cases, small cell lung cancer (SCLC) accounts for approximately 20%-25% of them [2]. Most cases of SCLC present with extensive disease due to the aggressive nature of SCLC. Extensive-stage disease (ED) and limited-stage disease (LD) are the two stages of SCLC that were identified by the Veterans Administration Lung Group (VALG). The VALG classifies LD as tumor growth limited to one hemithorax and its regional lymph nodes, while ED is tumor spread beyond these boundaries [2]. Combination cytotoxic therapies given at frequent intervals have yielded the best response rates in SCLC patients [2]. Although SCLC responds well to chemotherapy, the majority of patients will relapse and die within two years of diagnosis [3]. These poor survival rates are a result of the advanced stage of the disease at diagnosis, high recurrence rates associated with local therapy, and the inability of combination chemotherapy to prolong survival significantly [4]. Cisplatin in combination with etoposide is used routinely in the treatment of SCLC. The toxicity of cisplatin, including nausea, vomiting, nephrotoxicity, neurotoxicity, and ototoxicity, has led to the development and investigation of combination regimens with different toxicity profiles [5]. Carboplatin, a derivative of cisplatin, has far less nonhematologic toxicity, although myelosuppression may be slightly greater than that observed with cisplatin [5, 6]. The reduced toxicity and equivalent efficacy of carboplatin has resulted in increased use of carboplatin-based modalities for the treatment of SCLC [5]. Carboplatin, an analog of cisplatin, has similar activity in lung cancer but exhibits a more favorable toxicity profile and is easier to administer than cisplatin [5, 6]. Nonhematologic toxicities are less common in patients who receive carboplatin in comparison with those who receive cisplatin. The dose-limiting toxicity for carboplatin is myelosuppression, particularly thrombocytopenia [7].
Approximately 70% of an administered dose of carboplatin is excreted in the urine. Renal clearance of carboplatin is closely correlated with glomerular filtration rate (GFR) [7]. Therefore, calculation of the carboplatin dose should consider patients' renal function. The clear relationship between dose and hematologic toxicity for carboplatin justifies the need to pharmacokinetically determine the dose of carboplatin to both minimize toxicity and increase therapeutic efficacy. The ability to predict the area under the concentration versus time curve (AUC) following administration of carboplatin will allow consistent drug exposure for patients with either normal or impaired renal function. Calvert and colleagues have validated the use of the following formula to calculate carboplatin dosage in adults based on GFR [7] measured as 51Cr EDTA clearance: Total dose (mg) = (target AUC) x (GFR + 25) Note: With the Calvert formula, the total dose of carboplatin is calculated in mg, not mg/m2. The AUC of carboplatin appears to relate more closely to the therapeutic and toxic effects of the drug than does dose calculated on the basis of body surface area. Therefore, the use of AUC dosing avoids subtherapeutic doses and minimizes overdosage. Additionally, with the use of AUC rather than toxicity as the measure of carboplatin exposure, previous or concurrent myelosuppressive therapy will not have a significant influence [7]. Studies that were conducted prior to the widespread use of the Calvert formula do not provide standardized doses of carboplatin to patients of varying ages, performance status, and renal function. The importance of standardizing carboplatin dose based on renal function cannot be overlooked in future trials.
A number of studies have been conducted to evaluate the combination of carboplatin and etoposide in the treatment of SCLC in previously untreated patients (Table 1
A randomized phase III trial was conducted by the Hellenic Cooperative Oncology Group that compared the efficacy and toxicity of etoposide/cisplatin (EP) versus etoposide/carboplatin (EC) in previously untreated patients with SCLC [9]. Patients less than 75 years of age with SCLC confirmed by histology or cytology and WHO (World Health Organization) performance status less than 3 were eligible. Disease stage was classified as LD or ED according to VALG criteria. Patients were randomized to receive etoposide 100 mg/m2 i.v. on days 1 to 3 plus either cisplatin 50 mg/m2 i.v. on days 1 and 2 or carboplatin 300 mg/m2 i.v. day 1 [9]. Treatment cycles were administered every three weeks, for a maximum of six cycles. Following the third cycle, patients with LD received thoracic irradiation with the fourth cycle. For those LD patients who achieved a complete response, prophylactic cranial irradiation was given.
The response rates of patients receiving etoposide/cisplatin versus etoposide/carboplatin are shown in Table 1 In addition, patients who received etoposide/carboplatin experienced significantly less toxicity, both nonhematologic and hematologic [9]. WHO grade 3/4 leukopenia was seen in 37.5%/12.5% of the patients treated with etoposide and cisplatin in comparison with patients treated with etoposide and carboplatin, who only experienced 10.3%/6.8%. Thrombocytopenia (WHO grade 4) occurred in only 6% of patients receiving etoposide/cisplatin and 4% of patients receiving etoposide/carboplatin. Nausea and vomiting, nephrotoxicity, neurotoxicity, leukopenia, infection, mucositis, and allergic reactions were reported less frequently for patients in the carboplatin arm than for those in the cisplatin arm. Nausea and vomiting along with neurologic toxicities were significantly less in patients receiving etoposide and carboplatin. WHO grade 2/3 nausea and vomiting occurred in 71%/4% of patients receiving etoposide and cisplatin, where nausea and vomiting only occurred in 25%/0% of patients receiving etoposide and carboplatin (p = 0.0001). Neurologic toxicity (WHO grades 1/2) occurred in 41%/12% of patients receiving etoposide and cisplatin and only 18%/0% of patients receiving etoposide and carboplatin, which was statistically significant (p = 0.0002). Carboplatin was also easier to administer and resulted in fewer hospitalizations. Dose intensity was similar for both the carboplatin and the cisplatin regimens. This trial provides a rationale for the use of etoposide/carboplatin over etoposide/cisplatin because of similar efficacy and decreased toxicity. Similar toxicity profiles were reported in other trials using etoposide/carboplatin for the treatment of SCLC [10-15]. Leukopenia and thrombocytopenia were most frequently reported, dose limiting, and dependent on the dosage regimen used. In another clinical trial, when carboplatin 125 mg/m2/d in combination with etoposide 200 mg/m2/d was administered for three days, grade 3/4 leukopenia and thrombocytopenia were reported in 81% and 76% of patients, respectively. Four patients died from myelosuppression [10]. However, only 19% of patients reported grade 3/4 nausea and vomiting, and there was no neurotoxicity. In contrast, when carboplatin 300 mg/m2 was administered on day 1, the incidence of grade 3/4 leukopenia and thrombocytopenia was reported at rates of 20% and 16%, respectively [12]. Similarly, when carboplatin 450 mg/m2 was administered on day 1, the incidence of grade 3/4 leukopenia and thrombocytopenia was 8% and 11%, respectively [11]. Even higher single doses of carboplatin are associated with increased hematologic toxicity, with approximately two-thirds of patients receiving carboplatin 600 mg/m2 on day 1 developing grade 3/4 leukopenia and/or thrombocytopenia [13]. In a dose-escalation trial, the maximum tolerated dose of carboplatin was 650 mg/m2 in patients younger than 70 years of age and 450 mg/m2 for those 70 years or older [14]. As discussed earlier, the use of AUC dosing may help reduce the hematologic side effects. A recent CALGB study attempted to dose-intensify the etoposide/carboplatin combination using GM-CSF support. Several dose levels of GM-CSF, carboplatin, and etoposide were evaluated. Dose levels evaluated were: carboplatin25 and 150 mg/m2/d x 3; etoposide200, 250, and 300 mg/m2/d x 3, and GM-CSF5, 10, 20 µg/kg/d and 5 µg/kg every 12 h. GM-CSF was found to support only the first cycle of carboplatin/etoposide with acceptable toxicities; further cycles were not given because of delayed platelet and neutrophil count recovery. The conclusion was that dose intensification of etoposide and carboplatin was not possible with GM-CSF support alone [15].
A phase III trial designed to examine the role of carboplatin in SCLC randomized patients to receive either etoposide 200 mg/m2 days 1-3 and vincristine 1.4 mg/m2 (max 2 mg) days 1 and 8 or carboplatin 300 mg/m2 i.v. day 1; etoposide 140 mg/m2 days 1-3; and vincristine 1.4 mg/m2 (maximum 2 mg) i.v. on days 1, 8, and 15 (CEV) (Table 2
Another study evaluated dose-intensified CEV given with G-CSF support. Fifty-two untreated patients (26 with LD and 26 with ED) were given carboplatin (300 mg/m2 on day 1), vincristine (1.4 mg/m2 on days 1, 8, 15) and etoposide (190 mg/m2 on days 1-3) every three weeks. Of 45 evaluable patients (22 ED and 23 LD), the overall response rate was 86% and 95% for ED and LD patients, respectively. The median survival for ED patients was 10 months. In patients with LD, the median survival time has not been reached [17]. An Italian group evaluated the feasibility of dose-intensifying etoposide (i.v. on days 1-3) in combination with carboplatin (300 mg/m2 i.v. on day 1) and epirubicin (75 mg/m2 i.v. on day 1) with G-CSF support. Fifteen patients with ED SCLC received three dose levels of etoposide. Three patients received 100 mg/m2, six patients received 120 mg/m2, and six patients received 140 mg/m2. Of three patients who received VP-16 at 100 mg/m2, two had a complete response and one had a partial response. Of six patients who received VP-16 at 120 mg/m2, two of six achieved a complete response and three of six achieved a partial response. One of six patients had a complete response, and four of six had a partial response of those who received VP-16 at 140 mg/m2. Survival was not reported. The maximum tolerated dose of etoposide was 140 mg/m2. Hematologic toxicities (neutropenia and thrombocytopenia) were the dose-limiting toxicities [18]. Skarlos et al. evaluated the activity of a four-drug regimen containing carboplatin and etoposide in patients with LD SCLC [19]. The regimen consisted of carboplatin (150 mg/m2 i.v. on day 1) with etoposide (75 mg/m2 i.v. for two days) given every other week. These two drugs alternated with epirubicin (30 mg/m2 i.v.) and ifosfamide (2 gm/m2 with mesna) given every other week. After 77 weeks, patients were divided into two groups, A and B. If a complete response occurred, a patient could be put into either group A or group B. Patients in group A received only additional radiotherapy and PCI; patients in group B received more chemotherapy using the same chemotherapeutic agents, but the dosages were increased by 25% except for etoposide. The etoposide dose was kept at 75 mg/m2, but given for three days instead of only two. Patients in group B were also given radiotherapy and PCI. Twenty-nine patients were included in group A. In group B, 37 patients were studied. The overall response rates in groups A and B were 79.5% and 92%, respectively. Group B had a higher complete response rate, 46.9%, compared with that of group A, at 28.6%. The median survival of 10.6 months for Group B was higher than that of Group A at 8.33 months. These results were not statistically significant. Thus, the study showed that this regimen was not superior to standard dose [19]. The Australian Lung Cancer Study Group evaluated etoposide 120 mg/m2 and carboplatin 100 mg/m2 (both i.v. days 1-3) plus cyclophosphamide 750 mg/m2 and vincristine 1.4 mg/m2 (both i.v. day 1) [20]. Patients were assessed after three cycles, and for those with LD who responded, mediastinal radiation was given. LD patients with a complete response also received PCI [20]. Of the 87 evaluable patients, complete response rates of 40% and partial response rates of 41% were observed for etoposide/carboplatin plus cyclophosphamide and vincristine. The median survival time was 13.3 months for patients with LD and 9.6 months for patients with ED [20]. Neutropenia was the dose-limiting toxicity, and nonhematologic toxicity was mild in these patients receiving the four-drug combination. A Japanese study evaluated the combination of carboplatin, etoposide, and cisplatin both in patients with LD and with ED [21]. Evaluation of increasing doses of carboplatin and cisplatin was performed in a stepwise fashion. The dose of etoposide was held at 100 mg/m2 i.v. on days 1, 3, and 5. The maximum tolerated dose of carboplatin and cisplatin was 250 mg/m2 i.v. on day 1 and 60 mg/m2 i.v. on day 1, respectively. Hematologic toxicity was dose-limiting. At this dose level, four of seven patients experienced grade 4 thrombocytopenia, and the overall response rate was 100%, with 2/7 patients achieving a complete remission. More patients are needed to evaluate the tolerability, response, and survival in patients with SCLC [21].
Recently, two different studies evaluating the carboplatin/etoposide combination with concurrent radiation therapy were presented at the IALSC Lung Cancer Conference held in Dublin in the summer of 1997 (Table 3
Another study evaluated induction chemotherapy followed by thoracic irradiation in LD SCLC patients [23]. The induction therapy consisted of cyclophosphamide, epirubicin, and vincristine. After two to three cycles of therapy, 46 of 64 patients achieved an objective response. Thirty-nine of those patients then went on to receive late intensification chemotherapy and concurrent radiotherapy. The chemotherapy consisted of carboplatin (150 mg/m2 on days 1-3) and VP-16 (100 mg/m2 i.v. days 1-5) given concurrently with thoracic radiation of 45 Gy in twice-daily fractions. Patients were supported with G-CSF (5 µg/kg/d) from days 7 to 16; this combination was given for three cycles. Of those 39 patients, 30 achieved a CR (77%) and nine achieved a partial response. The median survival was 18.9 months. There were no episodes of sepsis or bleeding complications recorded. However, grade 3 and 4 neutropenia and thrombocytopenia occurred in 22% and 27% of the courses, respectively. These studies show that concurrent thoracic radiotherapy can be given with carboplatin and etoposide safely and effectively.
Patients over 65 years of age frequently are not good candidates for systemic chemotherapy because of coexisting medical conditions, including chronic heart disease and renal impairment [24]. Furthermore, elderly individuals do not tolerate intensive chemotherapy as well as younger patients, and they attain only modest gains relative to increased toxicity [25]. As the population over the age of 70 continues to increase, and because one-third of lung cancers occur in this age group, treatment of elderly patients is increasingly important.
Results of studies evaluating the use of etoposide/carboplatin in elderly or medically compromised patients are shown in Table 4
The etoposide/carboplatin regimens were well tolerated in these patients [24, 25]. As might be expected, myelosuppression was the dose-limiting toxicity [24]. Grade 3/4 thrombocytopenia and leukopenia occurred in approximately 32% to 53% and 45% to 50% of these patients, respectively [26, 28].
Ifosfamide, an analog of cyclophosphamide, has a broad range of activity in various tumor types, including lung cancer, sarcomas, lymphomas, breast cancer, gynecologic malignancies, and germ cell tumors [29]. It is one of the most active agents in SCLC, with objective response rates to single-agent therapy of approximately 50% in previously untreated patients [4]. In addition, ifosfamide is relatively nonmyelosuppressive as compared with cyclophosphamide, allowing combinations of ifosfamide with other myelosuppressive agents [29]. Other side effects include renal and neurologic toxicities as well as hemorrhagic cystitis, which may be reduced or prevented by the administration of mesna [6, 29]. As noted below, a number of studies have evaluated ifosfamide in combination with carboplatin and etoposide, with or without other agents, in previously untreated patients with SCLC (Table 5
The major toxicities of the ICE-containing regimens were neutropenia and thrombocytopenia, with a tendency for these toxicities to increase with subsequent cycles of treatment, and a 6% to 10% rate of death from infection [30-33]. The severe myelosuppression observed in most patients requires close monitoring, supportive care, and dose reduction [30]. Nausea and vomiting, which can be severe, were also common [30-33]. Some investigators have attempted to dose-intensify regimens by increasing the doses of the chemotherapeutic agents, but Hanauske et al. evaluated dose intensification by decreasing the intervals between cycles from 27 to 17 days using a combination of carboplatin, ifosfamide, etoposide, and vincristine [35]. The specific regimen consisted of carboplatin 250 mg/m2 i.v. on day 1, ifosfamide 2 g/m2 on days 1 and 2, etoposide 120 mg/m2 i.v. on days 1 and 2 with an oral dose on day 3, and vincristine 1.4 mg/m2 i.v. on day 14. G-CSF was given, allowing cycle intervals to be reduced to every 17 days. The toxicities were moderate, with 21% of patients developing febrile episodes and 48% of patients being withdrawn from the study. The shorter the cycle interval, the better the response rate, with patients receiving cycles greater than every 23 days achieving a 71% response rate and patients receiving cycles equal to or less than every 17 days having a 100% response rate [35]. Shevlin et al. modified the ICE regimen dosages (carboplatin 6 x (GFR + 25) mg, ifosfamide 3 g/m2 plus mesna 3 g/m2, mesna 1.8 g/m2 bolus, and 50 mg oral etoposide twice daily for seven days) [36]. The chemotherapy regimen was given every 28 days. Thirty patients (27 with LD and three with ED) took part in the study. The overall response rate was 83%, and the median survival time was 12.6 months. Only one-half of the patients were able to receive treatment without dose reduction or delay. The deviation in most patients was caused by WHO grade 3 or 4 thrombocytopenia or neutropenia. The dose-intensified ICE chemotherapy regimen did result in acceptable survival and response rates, but had a substantial amount of toxicity associated with it.
Several other studies have looked at whether dose-intensive ICE-containing regimens improve response rates [37-40]. Data from several ongoing studies were presented at the IASLC Lung Cancer Conference this past summer. These studies used peripheral blood stem cell (PBSC) support to theoretically reduce the cytopenias that occur with the dose-intensive therapy (Table 6
Dose-intensive therapy using PBSC transplant was studied using four drugs (etoposide, ifosfamide, carboplatin, and epirubicin). One such study took patients with LD SCLC and gave high-dose ifosfamide 4 gm/m2/d i.v. days 1-3 and epirubicin 30 mg/m2/d i.v. days 1-3 for cycles 1 and 3, then gave high-dose carboplatin AUC 5 mg/d i.v. days 1-2 and etoposide 120 mg/m2/d i.v. days 1-3 for cycles 2 and 4. These drugs were given currently with radiotherapy at fractions of 2 Gy once a day on days 1 through 5 of each cycle. G-CSF was given, and PBSC were reinfused on day 5 of cycles 2, 3, and 4. No prophylactic cranial irradiation was given, and no one died of acute toxicities. Of 35 evaluable patients, 65% achieved a complete response. However, 21 patients relapsed, mainly in the central nervous system. The median survival was 27 months [39]. Another study evaluated patients with LD and ED receiving first etoposide (500 mg/m2), ifosfamide (400 mg/m2), cisplatin (50 mg/m2), and epirubicin (50 mg/m2) with G-CSF. Then, thirty patients proceeded to receive high-dose chemotherapy consisting of etoposide 1,500 mg/m2, ifosfamide 12 gm/m2, carboplatin 750 mg/m2, and epirubicin 150 mg/m2, with autologous PBSC transplant. In contrast with the previous study, 13% of patients died of treatment-related complications. Median survival was 26 months in patients with LD SCLC and eight months with ED. Two-year survival also demonstrated the same trend, with 53% LD and 9% ED SCLC patients surviving at two years. Patients with LD were the only group that showed an improved survival from dose-intensive chemotherapy with PBSC support [40].
Paclitaxel, isolated from the western yew tree (Taxus brevifolia), exerts its cytotoxic effect by interfering with microtubule structure and function. Dose-limiting toxicities include myelosuppression (neutropenia) and peripheral neuropathy [41]. In two clinical trials, as a single agent, paclitaxel 250 mg/m2 given over 24 h every three weeks resulted in an overall response rate of 34% to 41% in previously untreated patients with ED SCLC [41, 42].
The combination of carboplatin/paclitaxel/etoposide has been evaluated in several trials of patients with SCLC (Table 7
A complicated study done by Lassen et al. evaluated the efficacy and toxicity of cisplatin versus carboplatin in combination with teniposide and vincristine in patients with LD and ED SCLC [49]. Four hundred eighty-four patients were randomized to either arm 1 (three cycles of cisplatin) or arm II (three cycles of carboplatin), with teniposide and vincristine alternating with other combinations of chemotherapeutic agents such as: block Acyclophosphamide, etoposide, lomustine, and vincristine; block Bdoxorubicin and vincristine; and block Ccisplatin, hexamethylmelamine, and vindesine. Patients were also randomized to an arm III that only alternated blocks A, B, and C. While grade 3/4 leukopenia was more common in arm III, grade 3/4 thrombocytopenia was more common in arms I and II. The efficacy, toxicity, response rates, and survival rates were no different between arm I and II (cisplatin versus carboplatin). The platinum and teniposide combinations (arms I and II) showed only superior survival rates compared with arm III (without teniposide). For arms I, II, and III, the overall response rates were 63%, 72%, and 65%, respectively, while the two-year survival rates were 16%, 14%, and 9%, respectively [49]. A novel combination of carboplatin and vinorelbine has been evaluated recently [50]. Thirty-three chemo-naive patients with ED SCLC were given carboplatin (300 mg/m2) i.v. on day 1 and vinorelbine (25 mg/m2) i.v. on days 1 and 8 every four weeks. The overall response rate was 70%, with a complete response rate of 24%. The median survival rate has not yet been reached. This combination was relatively well tolerated. WHO grades 3 and 4 leukopenia were 14.2% and 3.5%, respectively. Grade 2 anemia was 10.5%, and grade 3 thrombocytopenia was 3.5%. Overall, the combination achieved a comparable response rate and was well tolerated. More studies of the carboplatin and vinorelbine combination are warranted.
The combination of carboplatin/etoposide results in comparable efficacy, has less nonhematologic toxicity, and is easier to administer than cisplatin/etoposide. In addition, carboplatin-based regimens are particularly useful for elderly patients and those who are medically compromised. Response rates with carboplatin/etoposide in patients with LD and ED range from 73% to 93% and from 50% to 85%, respectively. Median survival is up to 15 months in LD patients and up to 12 months in ED. Several investigators evaluated the ICE combination. The two-year survival rates with ICE-containing regimens were 24% to 32% in LD patients. The principal toxicity of ICE regimens is myelosuppression, particularly neutropenia and thrombocytopenia. Nausea and vomiting can be severe, but other nonhematologic toxicities are generally manageable. Dose-intensive ICE-containing regimens with PBSC support may decrease cytopenias related to this intensive therapy. Further studies are needed to evaluate this. Thus far, only one study showed that patients with LD may have an improved survival from dose-intensive chemotherapy with PBSC support. Further research has combined paclitaxel with carboplatin and etoposide in the treatment of SCLC. Preliminary results are encouraging, and they support additional, larger trials of this combination. The study of the vinorelbine/carboplatin combination shows promising results but needs further investigation. These multiple approaches and new combination chemotherapy regimens with carboplatin may further determine the ability of systemic therapy to improve survival in patients with SCLC, and in combination with radiation therapy, potentially cure those patients with LD.
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||