Abstract
Background
The aim of the study is to demonstrate that intrapatient dose escalation of carboplatin would improve the outcome in ovarian cancer compared with flat dosing.
Patients and methods
Patients with untreated stage IC-IV ovarian cancer received six cycles of carboplatin area under the curve 6 (AUC 6) 3 weekly either with no dose modification except for toxicity (Arm A) or with dose escalations in cycles 2–6 based on nadir neutrophil and platelet counts (Arm B). The primary end-point was progression-free survival (PFS).
Results
Nine hundred and sixty-four patients were recruited from 71 centers. Dose escalation was achieved in 77% of patients who had ≥1 cycle. The median AUCs (cycle 2–6) received were 6.0 (Arm A) and 7.2 (Arm B) (P < 0.001). Grade 3/4 non-hematological toxicity was higher in Arm B (31% versus 22% P = 0.001). The median PFS was 12.1 months in Arm A and B [hazard ratio (HR) 0.99; 95% confidence interval (CI) 0.85–1.15; P = 0.93]. The median overall survival (OS) was 34.1 and 30.7 months in Arms A and B, respectively (HR 0.98; 95% CI 0.81–1.18, P = 0.82). In multivariate analysis, baseline neutrophil (P < 0.001), baseline platelet counts (P < 0.001) and the difference between white blood cell (WBC) and neutrophil count (P = 0.009) had a significant adverse prognostic value.
Conclusions
Intrapatient dose escalation of carboplatin based on nadir blood counts is feasible and safe. However, it provided no improvement in PFS or OS compared with flat dosing. Baseline neutrophils over-ride nadir counts in prognostic significance. These data may have wider implications particularly in respect of the management of chemotherapy-induced neutropenia.
Keywords: carboplatin, dose escalation, myelosuppresssion, ovarian cancer
introduction
Ovarian cancer remains the leading cause of death from gynecological cancer; the 5-year survival rate for advanced ovarian cancer is only 30%–40%. The international standard of care for first-line chemotherapy in patients with advanced ovarian cancer, based on the results of four phase III studies, is intravenous carboplatin and paclitaxel (Taxol) [1]; however in the ICON3 trial, carboplatin was associated with a similar survival outcome to the carboplatin–paclitaxel combination [2] and for some clinicians this has led to the use of single-agent carboplatin as an acceptable, alternative regimen, especially for those patients in whom the toxicity of the combination would be of particular concern.
The Calvert formula, based on renal function, has been universally adopted to determine the first dose of carboplatin [3]. It was originally based on a dose calculation expected to lead to a predictable degree of myelosuppression. However, in current practice, the majority of patients receive the same dose of carboplatin for subsequent cycles irrespective of nadir blood counts unless there is a significant change in renal function or clinically significant hematological toxicity. Several studies have indicated relationships between chemotherapy-induced myelosuppression and response to chemotherapy in a variety of malignancies including breast, lung, testicular cancer, melanoma and ovarian cancer [4–7].
In a previous randomized trial involving carboplatin, the Scottish Gynecologic Cancer Trials Group (SGCTG) reported that failure to achieve a significant degree of leucopenia with carboplatin was associated with worse progression-free survival (PFS, P < 0.001) in ovarian cancer patients [8]. However, the approach of tailoring and adjusting the carboplatin dose according to the extent of myelosuppression in an individual patient has not been tested prospectively in ovarian cancer.
Our hypothesis was that by determining the initial dose of carboplatin based on renal function, and then adjusting subsequent doses according to nadir blood counts, improved PFS for patients with ovarian cancer could be achieved. The aim of the SCOTROC-4 randomized trial was, therefore, to determine whether an intrapatient carboplatin dose-escalation strategy resulted in improved outcome compared with standard flat dosing.
methods
trial design
SCOTROC-4 was a phase III, international, multicenter, prospective, randomized trial of two carboplatin regimens in first-line chemotherapy of ovarian, fallopian and primary peritoneal cancers. The primary end-point was the comparison of PFS between carboplatin flat dosing (Arm A) and carboplatin with intrapatient dose escalation (Arm B). The secondary endpoints were comparisons of toxic effects, response rates (RECIST and CA125), overall survival (OS) and quality of life (QoL).
Patients were randomly allocated using the minimization technique [9]. Factors used for minimization were residual disease, center, stage, performance status, grade, interval debulking intention, elevated CA125 before treatment, cytological diagnosis, primary peritoneal/fallopian tube histology. Investigators were not blinded to allocation to Arm A or B.
The protocol was approved by the UK ethics committee and the study was conducted in accordance with the Principles of Good Clinical Practice and the Declaration of Helsinki. All patients provided written informed consent. An independent data monitoring committee (DMC) reviewed the interim outcome and safety analyses.
patient eligibility
Patients with histologically confirmed epithelial ovarian carcinoma, primary fallopian tube carcinoma or ovarian-type peritoneal carcinomatosis who were considered suitable for chemotherapy were eligible. In 2006, the protocol was amended to allow patients with a cytological diagnosis providing that a pelvic mass was present and that the CA125/CEA ratio was >25. In addition, the following criteria were required: (i) age ≥18, (ii) International Federation of Gynecology and Obstetrics stage IC to IV, and (iii) Eastern Cooperative Oncology Group performance status of 0–2. The exclusion criteria included (i) prior treatment with chemotherapy or radiotherapy; neutrophils <1.5 × 109/l; platelets <100 × 109/l; calculated creatinine clearance (Cockcroft-Gault equation) of <30 ml/min; bilirubin >upper limit of normal (ULN); ALT and AST > 2.5 ULN; ALP > 5 × ULN; peripheral neuropathy ≥NCI-CTC grade 2.
treatment and dose modification
Patients were randomly assigned within 8 weeks of surgery to receive carboplatin at a flat dose (Arm A) or with an intrapatient dose escalation (Arm B). For cycle 1 in both the arms, the dose, based on the glomerular filtration rate (Cockcroft-Gault formula), was area under the curve 6, AUC 6 (Calvert formula). In Arm A, six cycles of 3 weekly carboplatin at AUC 6 were administered. Dose escalation was not permitted. In Arm B, carboplatin was administered 3 weekly. Cycle 1 of carboplatin was administered at AUC 6 and the dose of subsequent cycles (2–6) was calculated according to the nadir blood count taken between days 14 and 18 (Table 1).
Table 1.
Dose modification (Arm B) according to nadir full blood count
| Nadir neutrophil × 109/l | Nadir platelets × 109/l | Dose modification on next cycle |
|---|---|---|
| ≥1.5 | ≥100 | Increase 20% |
| 76–99 | Increase 10% | |
| 50–75 | No change | |
| 11–49 | Decrease 10% | |
| ≤10 or with hemorrhage or requiring platelet transfusion | Decrease 20% | |
| 1.0–1.49 | ≥100 | Increase 10% |
| 50–99 | No change | |
| 11–49 | Decrease 10% | |
| ≤10 or with hemorrhage or requiring platelet transfusion | Decrease 20% | |
| ≥0.5–0.99 | ≥50 | No change |
| 11–50 | Decrease 10% | |
| ≤10 or with hemorrhage or requiring platelet transfusion | Decease 20% | |
| <0.5 | >10 | Decrease 10% |
| ≤10 or with hemorrhage or requiring platelet transfusion | Decease 20% | |
| <0.5 with complicationsa | any | Decrease 20% |
Febrile neutropenia or infection.
In both the treatment arms, carboplatin was delayed by a week if on day 1, ANC was <1.5 × 109/l and platelets <100 × 109/l. Subsequent doses were administered at the same dose or at a reduced dose according to the nadir counts. A 10% dose reduction was required following a delay for hematological toxicity of 2 weeks. Patients requiring longer delays came off protocol treatment. Dose escalations were not permitted following dose reductions for toxicity. Growth factor support with recombinant G-CSF was not permitted within the study. For NCI-CTC version 2, grade 2 non-hematological toxicity, treatment was delayed until recovery. In the event of NCI-CTC ≥grade 3 toxicity, protocol treatment was stopped.
clinical assessments
CT scans were carried out at baseline and after six cycles of treatment and CT also carried out if CA125 rose or clinical progression was suspected. CA 125 measurements were carried out at baseline, before each cycle of treatment, and then 2 monthly. Tumor response was determined according to RECIST version 1.0 criteria. Patients were followed up for 2 years every 2 months and then 3 monthly. The QoL was evaluated using the European Oncology Research and Treatment Committee questionnaires QLQ-C30 and QLQ-OV28 before each treatment cycle and at 6 months following randomization.
statistical considerations
The trial was designed to detect a 20% improvement in median PFS (flat dosing 12 months, intrapatient dose escalation 14.4 months) with an 80% power and a two-sided 5% level of significance, corresponding to a hazard ratio (HR) of 0.83. This required 950 progression or death events and 1300 patients were to be randomly assigned over 18 months. Efficacy analysis included all randomly assigned (intent-to-treat) patients. Analysis of treatment delivery, QoL and toxicity was restricted to those patients who started study treatment.
PFS and OS were defined as time from the date of randomization to progression or death (PFS) or death (OS) and estimated using the Kaplan–Meier method. A Cox proportional hazards model incorporating study minimization factors was used to compare the study arms.
The influence of hematological parameters on PFS was also examined in an exploratory manner using Cox regression techniques. Logistic regression (again incorporating study minimization factors) was used to compare RECIST 1.0 and CA125 response rates between the study arms. Elsewhere proportions were compared using Pearson’s chi-square test and ordinal data were compared using the Mann–Whitney U-test. For QoL analyses, treatment comparisons during treatment and at 6 months were carried out (See supplementary methods, available at Annals of Oncology online).
results
patients
Between March 2004 and January 2009, 964 eligible patients from 71 centers (UK, Australia and New Zealand) were randomly assigned. The DMC met on six occasions. On 2 December 2008, the DMC recommended closure of the study for futility on the basis of no apparent improvement in PFS for the intrapatient dose-escalation arm; this recommendation was endorsed by the trial steering committee on 17 December 2008. Baseline demographics and clinical characteristics were balanced between the two arms (Table 2). Two patients (one on each study arm) were found to be strictly ineligible at baseline and a further three patients on each arm did not start protocol therapy. These eight patients were excluded from the population analyzed for treatment delivery, safety and QoL. The study flow of participants is shown in Figure 1.
Table 2.
Patient baseline characteristics
| Study arm |
|||||
|---|---|---|---|---|---|
| Flat dosing (N = 481) | Intrapatient dose escalation (N = 483) | ||||
| Median age (years) | 67.7 | 67.9 | |||
| Median GFR (ml/min) | 70.0 | 68.6 | |||
| No. of patients | % | No. of patients | % | ||
| Bulk of residual disease after surgery | None/microscopic | 157 | 32.6 | 159 | 32.9 |
| Macroscopic <2 cm | 92 | 19.1 | 93 | 19.3 | |
| Macroscopic >2 cm | 232 | 48.2 | 231 | 47.8 | |
| FIGO stage (from registration form) | Ic | 60 | 12.5 | 59 | 12.2 |
| II | 39 | 8.1 | 39 | 8.1 | |
| III | 318 | 66.1 | 318 | 65.8 | |
| IV | 64 | 13.3 | 67 | 13.9 | |
| Tumor grade | Well | 31 | 6.4 | 32 | 6.6 |
| Moderate | 97 | 20.2 | 100 | 20.7 | |
| Poorly/undifferentiated | 228 | 47.4 | 228 | 47.2 | |
| Unknown | 125 | 26.0 | 123 | 25.5 | |
| Eastern Cooperative Oncology Group performance status | 0 | 133 | 27.7 | 132 | 27.3 |
| 1 | 253 | 52.6 | 252 | 52.2 | |
| 2 | 86 | 17.9 | 87 | 18.0 | |
| 3 | 9 | 1.9 | 12 | 2.5 | |
| Intend to have interval debulking operation | No | 329 | 68.4 | 328 | 67.9 |
| Yes | 152 | 31.6 | 155 | 32.1 | |
| Tumor type | Epithelial | 407 | 84.6 | 407 | 84.3 |
| Peritoneal | 62 | 12.9 | 66 | 13.7 | |
| Fallopian | 12 | 2.5 | 10 | 2.1 | |
| Elevated CA125 | No | 62 | 12.9 | 63 | 13.0 |
| Yes | 419 | 87.1 | 420 | 87.0 | |
| Was the diagnosis of ovarian cancer based on histology? | No | 23 | 10.8 | 22 | 10.2 |
| Yes | 190 | 89.2 | 193 | 89.8 | |
FIGO, International Federation of Obstetricians and Gynaecologists.
Figure 1.
CONSORT diagram. Flow of study participants.
carboplatin exposure
A total of 77.4% of cycles delivered to patients in Arm B (after cycle 1) were dose modified according to the protocol. Eighty-two percent of patients who had >1cycle achieved at least one dose escalation; 41% had one escalation; 25% had two; 10% had three and 5% had four escalations.
Overall, the median AUC per patient (received 2–6 cycles of treatment) was significantly higher in the dose-escalation arm (Arm B) compared with flat dosing (Arm A) (6.0 versus 7.2, P < 0.001). A total of 23.1% of patients in Arm B received carboplatin at an AUC of ≥8.0 (≥33% dose increase) and a further 25.3% achieved an AUC between 7.2 and 7.9 (≥20%, <33% dose increase).
A total of 82.6% (394) of patients in Arm A and 78.7% (377) of patients in Arm B completed six cycles of planned treatment. Although there was little difference between the two arms with respect to the total number of cycles patients received (P = 0.070), the number of cycles delayed or requiring dose reduction was significantly greater in Arm B. A total of 69.8% of patients in Arm A and 84.6% in Arm B received delayed cycles (P < 0.001); 32.3% (Arm A) and 63.3% (Arm B) of patients received cycles with a dose reduction (P < 0.001). Overall, the median dose intensity achieved compared with a notional equivalent of AUC 2 weekly was 92.3% in Arm A and 100.7% in Arm B (P < 0.001).
Neutropenia was the most frequent reason for both dose delays and treatment discontinuation due to toxicity in both the treatment arms. The most common reason for discontinuation of carboplatin before six cycles was disease progression in both the arms (Arm A 8.6%, Arm B 7.7%). 4.8% (Arm A) and 3.3% (Arm B) of patients continued to receive carboplatin beyond six cycles.
toxicity
As expected, the nadir counts for neutrophils, platelets, total white blood cell (WBC) and hemoglobin were similar in both the arms at cycle 1 but were significantly lower in Arm B for subsequent cycles. Similarly, the grades of toxicity for nadir hematological parameters, according to NCI-CTC version 2.0 criteria, were consistently higher in the dose-escalation arm (Arm B) (P < 0.001) (Table 3). Seven patients on arm B experienced neutropenic sepsis compared with four patients on arm A. The overall incidence of grade 3/4 non-hematological adverse events was ~10% higher in Arm B (22.0% versus 31.1%, P = 0.001). Nausea, vomiting, non-neutropenic-associated infection, blood transfusion requirements and dyspnoea were significantly higher in Arm B (Table 3). One patient on Arm B with febrile neutropenia died following bowel perforation.
Table 3.
Nadir hematological NCI CTC grading over all cycles received and common drug-related adverse events
| Study arm |
||||
|---|---|---|---|---|
| Flat dosing (n = 477) |
Intrapatient dose escalation (n = 479) |
|||
| % | Count | % | Count | |
| Neutropenia (P < 0.001) | ||||
| 2 | 33.1% | 158 | 34.2% | 164 |
| 3 | 17.2% | 82 | 31.9% | 153 |
| 4 | 3.1% | 15 | 8.1% | 39 |
| Total white blood cell (WBC, P < 0.001) | ||||
| 2 | 33.5% | 160 | 43.6% | 209 |
| 3 | 4.4% | 21 | 11.5% | 55 |
| 4 | 0.4% | 2 | 0.8% | 4 |
| Platelets (P < 0.001) | ||||
| 2 | 17.0% | 81 | 26.1% | 125 |
| 3 | 8.8% | 42 | 30.3% | 145 |
| 4 | 6.3% | 30 | 12.7% | 61 |
| Hemoglobin (P < 0.001) | ||||
| 2 | 49.7% | 237 | 61.2% | 293 |
| 3 | 8.8% | 42 | 15.0% | 72 |
| 4 | 1.0% | 5 | 3.1% | 15 |
| Blood transfusion (P < 0.001) | ||||
| Yes | 14.5% | 69 | 26.3% | 126 |
| Fatigue (P = 0.090) | ||||
| 2 | 23.5% | 112 | 25.7% | 123 |
| 3 | 2.5% | 12 | 4.4% | 21 |
| 4 | 0% | 0 | 0.4% | 2 |
| Constipation (P = 0.948) | ||||
| 2 | 19.7% | 94 | 18.4% | 88 |
| 3 | 0.8% | 4 | 2.1% | 10 |
| 4 | 0.2% | 1 | 0.2% | 1 |
| Nausea (P = 0.002) | ||||
| 2 | 11.7% | 56 | 17.3% | 83 |
| 3 | 1.3% | 6 | 2.9% | 14 |
| Stomatitis (P = 0.445) | ||||
| 2 | 3.8% | 18 | 4.8% | 23 |
| 3 | 0% | 0 | 0.2% | 1 |
| 4 | 0.2% | 1 | 0% | 0 |
| Vomiting (P < 0.001) | ||||
| 2 | 7.1% | 34 | 12.5% | 60 |
| 3 | 1.3% | 6 | 4.0% | 19 |
| 4 | 0.2% | 1 | 0.2% | 1 |
| Infection without neutropenia (P = 0.015) | ||||
| 2 | 2.7% | 13 | 4.8% | 23 |
| 3 | 0.4% | 2 | 1.5% | 7 |
| 4 | 0.0% | 0 | 0.2% | 1 |
| Dyspnoea (P = 0.004) | ||||
| 2 | 2.7% | 13 | 4.8% | 23 |
| 3 | 0.2% | 1 | 1.0% | 5 |
Common events were defined as those that occur in at least one study arm at grade ≥2 in ≥5% of patients.
efficacy
progression-free survival
In the futility analysis that led to study closure, the probability of a statistically significant result, assuming all the remaining events were to occur at the hypothesized HR of 0.83, was 0.12. In this final analysis, with 342 PFS events on arm A and 348 PFS events on arm B and after a median follow-up time of 26 months for living patients, there was no difference between the median PFS in Arm A [12.1 months, 95% confidence interval (CI) 10.9–13.4] and Arm B (12.1 months, 95% CI 11.1–13.4) (Figure 2A). The HR (dose escalation/flat dosing) is 0.99 (P = 0.93). The 95% CI (0.85–1.15) excludes the HR that the study was designed to detect (0.83).
Figure 2.
(A) Analysis of progression-free survival, PFS (primary end-point). (B) PFS according to baseline neutrophil count in a multivariable model of all patients.
overall survival
There is no statistically significant difference in median OS between the treatment arms (Arm A 34.1 months, 95% CI 29.8–38.4; Arm B 30.7 months, 95% CI 27.4–34.1). The HR is 0.98 (95% CI 0.81–1.18, P = 0.82). Overall, there were 222 deaths on Arm A, and 232 on Arm B. The majority of deaths were due to malignant disease; there was one death attributed to therapy in Arm B and none in Arm A.
treatment response
There was no statistically significant difference between the rates of overall (complete and partial) response by RECIST (Arm A 124/253 (49.0%), Arm B 128/240 (53.3%); estimated odds ratio 1.28, P = 0.183, 95% CI 0.89–1.86) or CA-125 criteria (78.9% (243/308) versus 75.8% (223/294); estimated odd ratio 0.87, P = 0.961, 95% CI 0.59–1.29) between Arms A and B, respectively.
quality of life
The QoL was assessed using serial EORTC QLQ-C30/OV28 questionnaires at pretreatment, each cycle and at 6 months after randomization (supplementary methods, S1 available at Annals of Oncology online). Completion rates for questionnaire declined from 87% at pretreatment to 64% at 6 months. The differences between the arms in the scale scores were small during treatment and at 6 months and no statistically significant differences were found.
prognostic factors
Hematological parameters were divided into quartiles and analyzed for associations with PFS. In the univariate analysis, the association between baseline neutrophils, WBC, difference between WBC and neutrophils, platelets and hemoglobin and PFS were all statistically significant (all P < 0.001); reduced PFS was associated with higher neutrophils, WBC, platelets and lower hemoglobin levels. Lower baseline difference between WBC and neutrophils was also associated with poor PFS.
Similarly low average nadir neutrophil and high average nadir hemoglobin values were associated with improved PFS (P < 0.001). Higher ratios of nadir neutrophils, platelets and WBC to their respective baselines were all associated with improved PFS (P < 0.001). Furthermore, a high average difference between WBC and neutrophil nadir (P < 0.001) was associated with improved PFS.
In a stepwise multivariable analysis, considering all the hematological parameters and baseline patient characteristics (stage, bulk of residual disease, grade, elevated CA125, tumor type), nadir counts were no longer significant (Table 4).
Table 4.
Results of multivariable analysis of the association between all hematological parameters, key baseline clinical characteristics and progression-free survival
| Variable | Category | N | HR | 95% CI for HR |
P value | |
|---|---|---|---|---|---|---|
| LL | UL | |||||
| Baseline neutrophils | ≤3.90 | 235 | 1 | – | – | <0.001 |
| 3.91–5.02 | 221 | 1.254 | 0.988 | 1.592 | ||
| 5.03–6.54 | 229 | 1.280 | 1.005 | 1.630 | ||
| 6.55+ | 228 | 1.647 | 1.281 | 2.117 | ||
| Difference between WBC and neutrophils | ≤2.00 | 244 | 1 | – | – | 0.009 |
| 2.01–2.46 | 211 | 0.801 | 0.647 | 0.992 | ||
| 2.47–2.99 | 229 | 0.940 | 0.759 | 1.163 | ||
| 3.00+ | 228 | 0.703 | 0.561 | 0.880 | ||
| Platelets before treatment | ≤304.00 | 228 | 1 | – | – | <0.001 |
| 304.01–389.00 | 229 | 0.971 | 0.760 | 1.242 | ||
| 389.01–498.00 | 228 | 1.497 | 1.181 | 1.899 | ||
| 498.01+ | 227 | 1.287 | 0.997 | 1.662 | ||
| Performance status | 0 | 242 | 1 | – | – | 0.001 |
| 1 | 484 | 1.188 | 0.968 | 1.458 | ||
| 2 | 167 | 1.578 | 1.226 | 2.032 | ||
| 3 | 19 | 2.108 | 1.230 | 3.614 | ||
| Stage | Ic | 108 | 1 | – | – | <0.001 |
| II | 76 | 1.421 | 0.840 | 2.402 | ||
| III | 603 | 3.113 | 2.016 | 4.806 | ||
| IV | 125 | 3.657 | 2.248 | 5.949 | ||
| Type | Epithelial | 769 | 1 | – | – | <.001 |
| Peritoneal | 122 | 1.742 | 1.399 | 2.169 | ||
| Fallopian | 21 | 1.778 | 1.028 | 3.075 | ||
| Bulk of residual disease after surgery | None/microscopic | 293 | 1 | – | – | <0.001 |
| Macroscopic < 2 cm | 179 | 1.274 | 0.980 | 1.656 | ||
| Macroscopic > 2 cm | 440 | 1.530 | 1.197 | 1.956 | ||
| Elevated CA125 | No | 113 | 1 | – | – | 0.008 |
| Yes | 799 | 1.564 | 1.124 | 2.177 | ||
HR, hazard ratio.
However, of the hematological parameters, the association of PFS with baseline neutrophils (P < 0.001) (Figure 2B), difference between baseline WBC and neutrophils (P = 0.009) and baseline platelet count (P < 0.001) remained statistically significant. Higher baseline neutrophil counts were associated with worse prognosis (<3.90 HR 1.0, >6.55 HR 1.6, P < 0.001) irrespective of treatment arm. Higher stage, performance status, residual disease after surgery, elevated CA125 and tumor type were also all significant prognostic factors (all P < 0.001).
discussion
The efficacy of higher doses of platinum has been investigated in several randomized trials in ovarian cancer. Most studies evaluated up to a twofold increase in dose intensity by randomly assigning patients to receive two different, fixed doses of platinum. Overall, these studies indicate that increasing the dose intensity of platinum by twofold without growth factor support is achievable. However, only two studies demonstrated an improvement in survival, albeit modest, and this was at the expense of increased toxic effects [10, 11].
Individualized dosing by adjusting subsequent doses of chemotherapy based on the level of myelosuppression or ‘toxicity-adjusting dosing’ [12] has been proposed as a more rational method for optimizing chemotherapy dose [8]. This tailored approach was investigated in a randomized trial in primary breast cancer patients [13]. This method resulted in a threefold interpatient range of cyclophosphamide dose (450–1800 mg/m2) and a fourfold range for epirubicin (38–120 mg/m2) together with a significantly improved relapse-free survival compared with marrow-supported high-dose chemotherapy [13].
The SCOTROC-4 trial, involving almost 1000 patients, is the largest study ever conducted, aimed at assessing the impact of intrapatient dose escalation. There were three main observations from the SCOTROC-4 trial. First, carboplatin dose escalation as per protocol is feasible for the majority of patients; 82% received at least one dose escalation. Overall, this resulted in a significantly higher median AUC dose administered in the dose-escalation arm over cycles 2–6 (6.0 versus 7.2, P < 0.001). Furthermore, ~23% of patients received carboplatin at an average AUC >8.0 over cycles 2–6. Dose escalation was safe; although non-neutropenic infection (grade 3 or 4) was more common in the dose-escalation arm (0.4% versus 1.7%; P = 0.02), the rate remained very low, as were the rates of neutropenic sepsis (1.5% and 0.8%, respectively).
Second, the results confirm that single-agent carboplatin is a reasonable treatment choice for selected patients with ovarian cancer. The majority of patients (67%) had suboptimal debulking surgery and the median PFS was in the range observed in other randomized studies (10.3–16.1 months) with similar patient baseline characteristics [2, 14]. However, intrapatient dose escalation of carboplatin failed to increase efficacy; there were no differences in PFS (median 12.1 months both arms) or OS (34 months Arm A; 31 months Arm B) in the dose-escalation arm compared with flat dosing.
Third, our exploratory multivariable analyses indicate that baseline neutrophils, platelets and difference between baseline WBC and neutrophils are highly prognostic factors in ovarian cancer. The prognostic values of baseline neutrophils and the neutrophil-to-lymphocyte ratio have been investigated (mainly in retrospective series) in several malignancies including melanoma, pancreatic, colorectal and ovarian cancers [6, 15–18]. While the association between high platelets and clinical outcome is well recognized [19], to our knowledge, ours is the first report in a large, prospective trial to demonstrate a statistically significant association between high baseline neutrophils and poor PFS in ovarian cancer. Moreover, we found poor associations between baseline neutrophils and other factors such as advanced disease status (Spearman’s rank correlation (r) for bulk of residual disease r = 0.3; stage r = 0.2), performance status (r = 0.3) and CA125 level (r = 0.1).
The previously unrecognized prognostic importance of baseline neutrophil counts calls into question the relevance of neutropenia following chemotherapy in ovarian cancer patients in various studies including our own [8]. In two other studies, there were conflicting data concerning the relationship of neutropenia and outcome, but neither addressed baseline counts [20, 21]. Our finding that nadir counts were not statistically significantly associated with PFS in the multivariate model suggests that baseline counts over-ride the absolute nadir count which in itself is a poor indicator of the impact of chemotherapy.
Recently, experimental evidence has been presented, which supports a relationship between increased neutrophils and resistance to anti-vascular endothelial growth factor (VEGF) therapy. Ferrara et al. demonstrated that G-CSF was able to induce angiogenesis and render tumors refractory to anti-VEGF treatment [22, 23]. Indeed, in patients with metastatic renal cell carcinoma treated with VEGF-targeted agents, raised neutrophil count was a significant independent adverse prognostic factor [24]. Since endogenous G-CSF is likely to be a key factor in increasing neutrophils, this highlights the potential risks, in terms of resistance to antiangiogenic agents, associated with exogenous G-CSF use to support chemotherapy-induced neutropenia [23]. In the light of positive recent trials in ovarian cancer where bevacizumab was combined with chemotherapy (ICON7 [25], GOG 218 [26], OCEANS [27]), it is conceivable that this will become increasingly relevant.
In conclusion, the SCOTROC-4 study demonstrated that although intrapatient dose escalation of carboplatin is feasible, this did not improve clinical outcome. Dose escalation of carboplatin can therefore not be encouraged as a standard practice. Further studies are recommended, examining the therapeutic implications of the adverse prognostic significance of high baseline neutrophil counts.
Supplementary Material
acknowledgements
We would like to acknowledge the patients, investigators and staff at all participating sites. The study was presented at ASCO 2009; Kaye et al. J Clin Oncol 27:15s, 2009 (suppl; abstr 5537).
funding
The clinical trials office in Glasgow has core grant support from CRUK, supplemented by SGCTG funding. NCRN provides infrastructure support for all contributing sites in England. SBK is supported by Cancer Research UK and the Department of Health, through Experimental Cancer Medicine Centre and Biomedical Research Centre grants to the Royal Marsden Hospital and Institute of Cancer Research. JP and L-AL were supported by Cancer Research UK grant C973/A9894 and the Scottish Gynaecological Clinical Trial Group.
Footnotes
disclosure
The authors have declared no conflicts of interest.
references
- 1.du Bois A, Quinn M, Thigpen T, et al. 2004 consensus statements on the management of ovarian cancer: final document of the 3rd International Gynecologic Cancer Intergroup Ovarian Cancer Consensus Conference (GCIG OCCC 2004) Ann Oncol. 2005;16(Suppl 8):viii7–viii12. doi: 10.1093/annonc/mdi961. [DOI] [PubMed] [Google Scholar]
- 2.Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet. 2002;360(9332):505–515. doi: 10.1016/S0140-6736(02)09738-6. [DOI] [PubMed] [Google Scholar]
- 3.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(11):1748–1756. doi: 10.1200/JCO.1989.7.11.1748. [DOI] [PubMed] [Google Scholar]
- 4.Di Maio M, Gridelli C, Gallo C, et al. Chemotherapy-induced neutropenia and treatment efficacy in advanced non-small-cell lung cancer: a pooled analysis of three randomised trials. Lancet Oncol. 2005;6(9):669–677. doi: 10.1016/S1470-2045(05)70255-2. [DOI] [PubMed] [Google Scholar]
- 5.Horwich A, Sleijfer DT, Fossa SD, et al. Randomized trial of bleomycin, etoposide, and cisplatin compared with bleomycin, etoposide, and carboplatin in good-prognosis metastatic nonseminomatous germ cell cancer: a Multiinstitutional Medical Research Council/European Organization for Research and Treatment of Cancer Trial. J Clin Oncol. 1997;15(5):1844–1852. doi: 10.1200/JCO.1997.15.5.1844. [DOI] [PubMed] [Google Scholar]
- 6.Schmidt H, Suciu S, Punt CJ, et al. Pretreatment levels of peripheral neutrophils and leukocytes as independent predictors of overall survival in patients with American Joint Committee on Cancer Stage IV Melanoma: results of the EORTC 18951 Biochemotherapy Trial. J Clin Oncol. 2007;25(12):1562–1569. doi: 10.1200/JCO.2006.09.0274. [DOI] [PubMed] [Google Scholar]
- 7.Lee CK, Gurney H, Brown C, et al. Carboplatin–paclitaxel-induced leukopenia and neuropathy predict progression-free survival in recurrent ovarian cancer. Br J Cancer. 2011;105(3):360–365. doi: 10.1038/bjc.2011.256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rankin EM, Mill L, Kaye SB, et al. A randomised study comparing standard dose carboplatin with chlorambucil and carboplatin in advanced ovarian cancer. Br J Cancer. 1992;65(2):275–281. doi: 10.1038/bjc.1992.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics. 1975;31(1):103–115. [PubMed] [Google Scholar]
- 10.Ngan HY, Choo YC, Cheung M, et al. A randomized study of high-dose versus low-dose cis-platinum combined with cyclophosphamide in the treatment of advanced ovarian cancer. Hong Kong Ovarian Carcinoma Study Group. Chemotherapy. 1989;35(3):221–227. doi: 10.1159/000238673. [DOI] [PubMed] [Google Scholar]
- 11.Kaye SB, Paul J, Cassidy J, et al. Mature results of a randomized trial of two doses of cisplatin for the treatment of ovarian cancer. Scottish Gynecology Cancer Trials Group. J Clin Oncol. 1996;14(7):2113–2119. doi: 10.1200/JCO.1996.14.7.2113. [DOI] [PubMed] [Google Scholar]
- 12.Gurney H. How to calculate the dose of chemotherapy. Br J Cancer. 2002;86(8):1297–1302. doi: 10.1038/sj.bjc.6600139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bergh J, Wiklund T, Erikstein B, et al. Tailored fluorouracil, epirubicin, and cyclophosphamide compared with marrow-supported high-dose chemotherapy as adjuvant treatment for high-risk breast cancer: a randomised trial. Scandinavian Breast Group 9401 study. Lancet. 2000;356(9239):1384–1391. doi: 10.1016/s0140-6736(00)02841-5. [DOI] [PubMed] [Google Scholar]
- 14.Vasey PA, Jayson GC, Gordon A, et al. Phase III randomized trial of docetaxel–carboplatin versus paclitaxel–carboplatin as first-line chemotherapy for ovarian carcinoma. J Natl Cancer Inst. 2004;96(22):1682–1691. doi: 10.1093/jnci/djh323. [DOI] [PubMed] [Google Scholar]
- 15.Thavaramara T, Phaloprakarn C, Tangjitgamol S, et al. Role of neutrophil to lymphocyte ratio as a prognostic indicator for epithelial ovarian cancer. J Med Assoc Thai. 2011;94(7):871–877. [PubMed] [Google Scholar]
- 16.Aliustaoglu M, Bilici A, Seker M, et al. The association of pre-treatment peripheral blood markers with survival in patients with pancreatic cancer. Hepatogastroenterology. 2010;57(99–100):640–645. [PubMed] [Google Scholar]
- 17.Cho H, Hur HW, Kim SW, et al. Pre-treatment neutrophil to lymphocyte ratio is elevated in epithelial ovarian cancer and predicts survival after treatment. Cancer Immunol Immunother. 2009;58(1):15–23. doi: 10.1007/s00262-008-0516-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Michael M, Goldstein D, Clarke SJ, et al. Prognostic factors predictive of response and survival to a modified FOLFOX regimen: importance of an increased neutrophil count. Clin Colorectal Cancer. 2006;6(4):297–304. doi: 10.3816/CCC.2006.n.048. [DOI] [PubMed] [Google Scholar]
- 19.Stone RL, Nick AM, McNeish IA, et al. Paraneoplastic thrombocytosis in ovarian cancer. N Engl J Med. 2012;366(7):610–618. doi: 10.1056/NEJMoa1110352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rocconi RP, Matthews KS, Kemper MK, et al. Chemotherapy-related myelosuppression as a marker of survival in epithelial ovarian cancer patients. Gynecol Oncol. 2008;108(2):336–341. doi: 10.1016/j.ygyno.2007.10.012. [DOI] [PubMed] [Google Scholar]
- 21.Kim JJ, Park JY, Kim DY, et al. Is chemotherapy-induced neutropenia a prognostic factor in patients with ovarian cancer? Acta Obstet Gynecol Scand. 2010;89(5):623–628. doi: 10.3109/00016341003674913. [DOI] [PubMed] [Google Scholar]
- 22.Shojaei F, Ferrara N. Refractoriness to antivascular endothelial growth factor treatment: role of myeloid cells. Cancer Res. 2008;68(14):5501–5504. doi: 10.1158/0008-5472.CAN-08-0925. [DOI] [PubMed] [Google Scholar]
- 23.Shojaei F, Wu X, Qu X, et al. G-CSF-initiated myeloid cell mobilization and angiogenesis mediate tumor refractoriness to anti-VEGF therapy in mouse models. Proc Natl Acad Sci USA. 2009;106(16):6742–6747. doi: 10.1073/pnas.0902280106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted agents: results from a large, multicenter study. J Clin Oncol. 2009;27(34):5794–5799. doi: 10.1200/JCO.2008.21.4809. [DOI] [PubMed] [Google Scholar]
- 25.Perren TJ, Swart AM, Pfisterer J, et al. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365(26):2484–2496. doi: 10.1056/NEJMoa1103799. [DOI] [PubMed] [Google Scholar]
- 26.Burger RA, Brady MF, Bookman MA, et al. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med. 2011;365(26):2473–2483. doi: 10.1056/NEJMoa1104390. [DOI] [PubMed] [Google Scholar]
- 27.Aghajanian C, Blank SV, Goff BA, et al. OCEANS: A randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer. J Clin Oncol. 2012;30(17):2039–2045. doi: 10.1200/JCO.2012.42.0505. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.