A phase 1 study of simvastatin in combination with topotecan and cyclophosphamide in pediatric patients with relapsed and/or refractory solid and CNS tumors

Abstract

Background:

3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors (statins) can inhibit tumor proliferation, angiogenesis, and restore apoptosis in preclinical pediatric solid tumor models. We conducted a phase 1 trial to determine the maximum tolerated dose (MTD) of simvastatin with topotecan and cyclophosphamide in children with relapsed/refractory solid and central nervous system (CNS) tumors.

Methods:

Simvastatin was administered orally twice daily on days 1–21, with topotecan and cyclophosphamide intravenously on days 1–5 of a 21-day cycle. Four simvastatin dose levels (DLs) were planned, 140 (DL1), 180 (DL2), 225 (DL3), 290 (DL4) mg/m²/dose, with a de-escalation DL of 100 mg/m²/dose (DL0) if needed. Pharmacokinetic and pharmacodynamic analyses were performed during cycle 1.

Results:

The median age of 14 eligible patients was 11.5 years (range: 1–23). The most common diagnoses were neuroblastoma (N = 4) and Ewing sarcoma (N = 3). Eleven dose-limiting toxicity (DLT)-evaluable patients received a median of four cycles (range: 1–6). There were three cycle 1 DLTs: one each grade 3 diarrhea and grade 4 creatine phosphokinase (CPK) elevations at DL1, and one grade 4 CPK elevation at DL0. All patients experienced at least one grade 3/4 hematologic toxicity. Best overall response was partial response in one patient with Ewing sarcoma (DL0) and stable disease for four or more cycles in four patients. Simvastatin exposure increased with higher doses and may have correlated with toxicity. Plasma interleukin 6 (IL-6) concentrations (N = 6) showed sustained IL-6 reductions with decrease to normal values by day 21 in all patients, indicating potential on-target effects.

Conclusions:

The MTD of simvastatin with topotecan and cyclophosphamide was determined to be 100 mg/m²/dose.

1 |. INTRODUCTION

3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors, or statins, lower cholesterol and reduce cardiovascular disease by inhibiting the enzyme that catalyzes the first step in cholesterol biosynthesis. Four different statins are Food and Drug Administration (FDA) approved for use in pediatrics, including simvastatin, which is indicated for patients with heterozygous familial hypercholesterolemia at doses of 10–40 mg daily.^1,2 Statins also have direct anti-inflammatory properties that lower heart disease risk by inhibiting interleukin 6 (IL-6) expression and plasma concentrations, and signal transducer and activator of transcription 3 (STAT3) phosphorylation to prevent recruitment of inflammatory cells to areas of myocardial injury.^3–6 The pleiotropic roles of statins that contribute to the reduction of heart disease also lower the incidence of cancer by targeting tumor-specific pathways.⁷

Statins inhibit tumor proliferation, metastasis, and angiogenesis and restore chemotherapy-induced apoptosis in many tumor models in vitro and in vivo, including glioblastoma, neuroblastoma, medulloblastoma, osteosarcoma, and rhabdomyosarcoma.^7–9 In these tumor models, lipophilic statins, for example, lovastatin and simvastatin, have been shown to inhibit tumor oncogenes like C-MYC, insulin-like growth factor receptor (IGFR), and STAT3 that are critical to tumor cell proliferation and survival.^4,9–11 Pediatric solid tumors such as neuroblastoma, rhabdomyosarcoma, medulloblastoma, osteosarcoma, and Ewing sarcoma have aberrant STAT3 signaling.^12,13 Statins exert antitumor effects by inhibiting IL-6-mediated STAT3 activation, and the multifactorial anticancer properties of statins suggest that they may be more effective than specific STAT3 or IL-6 inhibitors in the clinic.¹⁴ Statins also inhibit the isoprenylation and activation of critical oncogenes like RAS and RHO and inhibit tumor p-glycoprotein expression (p-gp, ABCB1), a multidrug resistance channel.⁷ Co-administration of simvastatin with conventional chemotherapy in vitro enhances chemotherapy-induced tumor cell death to restore chemotherapy response.^15,16 These data support the rationale for combining statins with chemotherapy.

Multiple clinical trials testing fixed doses of simvastatin, most commonly 40 mg daily (range 10–80 mg), in combination with chemotherapy, have been performed in adults with advanced solid tumors.^17–25 While these trials have broadly shown mixed results, nearly all of the randomized controlled trials have failed to show a benefit with the addition of simvastatin.^{17,19,20,24,25} Investigators from these trials hypothesized that the lack of clinical efficacy may have been due to the lower dosing of simvastatin that was insufficient to achieve serum concentrations needed to control tumor cell proliferation.^20,24 Preclinical studies have demonstrated that the anticancer effects of simvastatin are dose-dependent, and that the anti-angiogenic effects only occur at higher concentrations, suggesting that higher doses may be needed for simvastatin to achieve its maximum antitumor effects in clinical trials.^26–31

A phase I study in adults evaluated high-dose simvastatin in combination with vincristine, doxorubicin, and dexamethasone (VAD) or cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) for end-stage multiple myeloma and lymphoma, respectively.³² The maximum tolerated dose (MTD) of simvastatin in combination with those chemotherapy regimens was 15 mg/kg/day divided twice daily. Dose-limiting toxicities (DLTs) were grade-3 diarrhea and febrile neutropenia (FN) with sepsis, and no rhabdomyolysis was observed at any dose level (DL).³² Data from this study and others have demonstrated that simvastatin is very well-tolerated when given in combination with chemotherapy and has a low toxicity profile, making it ideal for heavily pretreated cancer patients.^{17,19–21,24,25,33} The primary objective of this pediatric phase 1 trial was to establish the MTD and recommended phase 2 dose (RP2D) of simvastatin in combination with topotecan and cyclophosphamide, and to define and describe the toxicities of this regimen, in children with relapsed/refractory (r/r) solid and central nervous system (CNS) tumors.

2 |. METHODS

2.1 |. Patient eligibility

Eligible patients were between the ages of 12 months and 29 years at the time of study enrollment and had a r/r solid or CNS tumor; measurable or evaluable disease; no known curative therapy; a Lansky (≤16 years) performance status 50% or higher, or Karnofsky (>16 years) 60% or higher; and had recovered from the acute toxic effects of all prior anticancer therapy. In addition, eligible patients had adequate organ function as defined in Appendix SA1, which included creatine phosphokinase (CPK) within normal parameters for age and gender. Patients were not eligible if they currently or previously received treatment with an HMG-CoA reductase inhibitor (any statin) or met any of the other exclusion criteria as defined in Appendix SA1.

This trial was conducted at the Aflac Cancer & Blood Disorders Center at Children’s Healthcare of Atlanta. The study was conducted in accordance with good clinical practices and the Declaration of Helsinki, and Emory University Institutional Review Board approval was obtained. Informed consent and assent, as appropriate per institutional guidelines, was obtained from patients and their guardians prior to enrollment. The trial was listed on clinicaltrials.gov as NCT02390843.

2.2 |. Trial design

The primary objective of this study was to estimate the MTD and RP2D of simvastatin administered orally (PO) twice daily in combination with topotecan and cyclophosphamide in children with r/r solid and CNS tumors, and to define and describe the toxicities of this regimen.

Secondary objectives were to preliminarily define the antitumor activity of the regimen, examine the relationship between serum cholesterol values and response, and to examine the relationship between serum levels of IL-6, soluble IL-6 receptor (sIL-6R), STAT-3, and phospho-STAT3 (pSTAT3) and clinical response.

Simvastatin was obtained via commercial supply as 5, 10, 20, 40, or 80 mg tablets, and was administered PO twice daily on days 1–21 of each 21-day cycle and at least 1 hour prior to chemotherapy on days 1–5. For patients unable to swallow tablets whole, simvastatin was dissolved in water and immediately administered either PO or through a nasogastric (NG) or gastric (G) tube. Missed doses could be given within 6 hours of the scheduled time, and doses were not repeated if the patient vomited after receiving the dose. All patients received topotecan (0.75 mg/m²/dose) and cyclophosphamide (250 mg/m²/dose) intravenously (IV) daily on days 1–5 of each cycle. Myeloid growth factor (granulocyte colony-stimulating factor [G-CSF] or pegylated G-CSF) was initiated 24–48 hours after the completion of topotecan and cyclophosphamide, and was continued until the absolute neutrophil count (ANC) was more than 2000/mm³. Supplementation with ubiquinone (coenzyme Q10) was allowed if patients developed symptoms of myopathy such as muscle weakness, cramping, pain, or tenderness, either with or without an elevation in CPK. Subsequent cycles could begin if the patient had at least stable disease (SD), met eligibility lab parameters, and did not meet criteria for removal from protocol therapy due to toxicity or other protocol-defined criteria. Cycles could be repeated every 21 days for up to 35 cycles or up to a total duration of therapy of approximately 24 months.

The starting dose of simvastatin was 140 mg/m²/dose (DL1), approximately 60% of the adult RP2D (225 mg/m²/dose) when administered in combination with chemotherapy.³² Dose escalations to 180 mg/m²/dose (DL2) and 225 mg/m²/dose (DL3) were planned with a possible dose de-escalation to 100 mg/m²/dose (DL0) if DLT occurred at the starting DL. There was no max simvastatin dose. A 3+3 design was used for dose escalation.³⁴ No intrapatient dose escalation was allowed. The MTD was defined as the maximum dose at which no more than one out of six patients experienced DLT during cycle 1 of therapy. Any patient who experienced a DLT during cycle 1 was considered evaluable for determination of the MTD. Patients without DLT who received at least 85% of the prescribed doses per protocol guidelines and had the appropriate toxicity monitoring studies performed during cycle 1 were also considered evaluable for determination of the MTD. Patients who were not evaluable for determination of the MTD at a given DL were replaced.

Toxicities were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. During cycle 1, patients were monitored with the following weekly evaluations: history and physical exam (PE) with vital signs; review of simvastatin subject diary; chemistries including liver function tests, magnesium, and phosphorous; CPK; and fasting cholesterol and triglyceride levels. Complete blood count with differential (CBCd) was obtained twice weekly during cycle 1. Patients also had myopathy screening performed with each PE that included evaluation for signs and symptoms of myopathy and/or rhabdomyolysis such as muscle weakness, cramping, muscle pain and/or tenderness, red or brown colored urine, or decreased urine output. A urinalysis was performed if any symptom was positive on the myopathy screening. All of the above evaluations occurred only at the beginning of the subsequent cycles, with the exception of CBCds, which were obtained weekly.

Nonhematologic DLT was defined as any grade 3 or higher nonhematologic toxicity that was possibly, probably, or definitely attributable to simvastatin, with exclusion of toxicities listed in Appendix SA2. Any grade 2 nonhematological toxicity that persisted for 7 or more days and was considered sufficiently medically significant or sufficiently intolerable to patients that it required treatment interruption was considered a DLT. Upon meeting eligibility parameters or returning to baseline, patients who experienced a nonhematologic DLT were able to continue protocol therapy with a one DL reduction of simvastatin. If the nonhematologic DLT recurred with subsequent cycles, the simvastatin dose remained at the lower DL, and the topotecan and cyclophosphamide were reduced by 25%. A subsequent recurrence of the nonhematologic DLT required removal from protocol therapy. Patients who had a nonhematologic DLT that did not resolve to baseline within 21 days after the planned start date of the next treatment cycle were removed from protocol therapy.

Hematologic DLT was defined as failure to recover ANC to 750/mm³ or higher or platelet count to 75,000/mm³ or higher within 14 days of the scheduled start date of the cycle. Patients who experienced hematologic DLT continued simvastatin at the same DL, but had the topotecan and cyclophosphamide dose reduced by 25%. Patients who had hematological DLT that did not resolve to baseline within 21 days after the planned start of the next treatment cycle were removed from protocol therapy.

Radiographic response was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST) guideline (version 1.1).³⁵ Tumor disease evaluations, including bone marrow evaluation, if applicable, were performed at the end of cycles 2, 4, and 6, and then every three cycles thereafter. Objective (partial or complete) responses had to be confirmed after the next consecutive cycle. Objective response rate (ORR) was defined as the number of patients experiencing a complete response (CR) or partial response (PR). Best overall response (BOR) was defined as best response recorded from the start of treatment until disease progression or recurrence. BOR was determined based on disease assessment at two time points, except for patients who progressed at the first disease evaluation, whose BOR was classified as progressive disease (PD).

2.3 |. Pharmacokinetic studies

Blood samples were drawn on cycle 1, day 1 at 0.5, 1, 2, 4, and 6 hours after the first dose of simvastatin. An additional sample was obtained on cycle 1, day 2 prior to taking the morning dose (24 hours after first dose of simvastatin). Blood samples (4 mL) were collected in sodium heparin tubes, gently inverted 5–10 times, and centrifuged (650–1450 × g) for 10–15 minutes at 2°C–8°C. Separated plasma was then immediately stored at −70°C until analysis. Simvastatin and simvastatin acid plasma concentrations were determined by PPD Laboratories using a validated liquid chromatography, tandem mass spectrometry (LC-MS/MS) method. Simvastatin pharmacokinetic (PK) parameters were derived from the data by noncompartmental methods with PK Solutions 2.0 (Summit Research Services, Montrose, CO, USA).

2.4 |. Pharmacodynamic studies

One peripheral blood sample (5 mL) was collected in a K2EDTA tube on days 1, 5, and after cycle 1 (either day 1 of cycle 2 or day 22 of cycle 1 if patient delayed). Whole blood was processed using Leucosep (GreinerBio) tubes following the manufacturer’s provided protocol to isolate plasma for analysis of IL-6 and IL-6Rα content and peripheral blood mononuclear cells (PBMCs) to measure phosphorylation of STAT3. Plasma and PBMCs were analyzed according to Appendix SA3.

3 |. RESULTS

3.1 |. Patient characteristics

Fourteen patients were enrolled on study, six to DL1 (140 mg/m²/dose) and eight to DL0 (100 mg/m²/dose). Baseline characteristics for all eligible patients are presented in Table 1.

TABLE 1.

Characteristics of eligible patients (N = 14).

Characteristic	Number (%)
Median age (range), years	11.5 (1–23)
Sex
Male	6 (43)
Female	8 (57)
Race
White	5 (36)
Black	5 (36)
Asian	3 (21)
Unknown	1 (7)
Ethnicity
Non-Hispanic	12 (86)
Hispanic	2 (14)
Diagnosis
Clear cell sarcoma, extremity	1 (7)
Ewing sarcoma	3 (21)
Malignant peripheral nerve sheath tumor	1 (7)
Malignant rhabdoid tumor, kidney	1 (7)
Medulloblastoma	1 (7)
Neuroblastoma	4 (29)
Rhabdomyosarcoma	1 (7)
Synovial sarcoma	1 (7)
Wilms tumor	1 (7)
Prior therapy
Median prior systemic therapy regimens (range)	2 (1–5)
Median prior radiation courses (range)	1 (1–3)

3.2 |. Toxicity

Eleven patients were evaluable for DLT determination. Two patients were inevaluable due to PD occurring prior to starting therapy, and one patient due to refusal of further protocol therapy during cycle 1, which led to the patient not receiving the amount of simvastatin required to be DLT-evaluable. The median number of cycles received by evaluable patients was four (range: 1–6). Two of five patients on DL1 experienced DLT, grade 3 diarrhea, and grade 4 CPK increase (Table 2). Subsequent patients were enrolled on de-escalation DL0, where one of six patients experienced DLT, a grade 4 CPK increase, thus establishing 100 mg/m²/dose as the MTD. No DLTs occurred in patients after cycle 1.

TABLE 2.

Summary of dose-limiting toxicity.

Dose level	Simvastatin dose (mg/m2)	No. of patients enrolled	No. of patients evaluable	No. of patients with cycle 1 DLT	Cycle 1 DLT	No. of patients with later-cycle DLT
0	100	8	6	1	Gr 4 CPK increased	0
1a	140	6	5	2	Gr 3 diarrhea Gr 4 CPK increased	0

Abbreviations: CPK, creatine phosphokinase; DLT, dose-limiting toxicity; Gr, grade; No., number.

^aStartingdose level.

As expected in patients receiving cytotoxic chemotherapy, the most common toxicities were hematologic, and all patients experienced at least one grade 3/4 hematologic toxicity. Common grade 3/4 cycle 1 hematologic toxicities that were at least possibly attributable to treatment included neutropenia (100%), leukopenia (100%), thrombocytopenia (91%), lymphopenia (91%), and anemia (55%) (Table 3). Grade 3 anemia did occur more frequently in patients who received more than one cycle (89%). There were only three grade 3 non-DLT cycle 1 nonhematologic toxicities that were at least possibly attributable to treatment: grade 3 FN (55%), grade 3 nausea (9%), and grade 3 vomiting (9%). No patient experienced a grade 4 non-DLT cycle 1 nonhematologic toxicity. There was not a clear trend toward increased nonhematologic toxicity in patients receiving more than one cycle.

TABLE 3.

Hematologic and nonhematologic toxicities^a observed in evaluable patients across all dose levels.

		No. of patients in cycle 1 (%) (N = 11)				No. of patients in cycle 2+ (%) (N = 9)
		By maximum grade				Bymaximum grade
Toxicity type	Toxicity	Grade 1	Grade 2	Grade 3	Grade 4	Grade 1	Grade 2	Grade 3	Grade 4
Hematologic	Anemia	0	3 (27)	6 (55)	0	0	1 (11)	8 (89)	0
	Lymphocyte count decreased	0	1 (9)	6 (55)	4 (36)	0	0	3 (33)	6 (67)
	Neutrophil count decreased	0	0	1 (9)	10 (91)	1 (11)	0	1 (11)	7 (78)
	Platelet count decreased	1 (9)	0	4 (36)	6 (55)	1 (11)	0	0	7 (78)
	WBC decreased	0	0	1 (9)	10 (91)	0	1 (11)	1 (11)	7 (78)
Nonhematologic	Abdominal pain	1 (9)	0	0	0	1 (11)	0	0	0
	ALT increased	1 (9)	0	0	0	1 (11)	0	0	0
	Alopecia	2 (18)	1 (9)	0	0	1 (11)	2 (22)	0	0
	Anorexia	2 (18)	4 (36)	0	0	1 (11)	3 (33)	0	0
	AST increased	0	1 (9)	0	0	0	0	0	0
	Bilirubin increased	1 (9)	0	0	0	1 (11)	0	0	0
	CPK increased	0	0	0	0	0	0	0	0
	Creatinine increased	1 (9)	0	0	0	0	0	0	0
	Diarrhea	0	0	0	0	0	0	0	0
	Dysgeusia	0	1 (9)	0	0	0	0	0	0
	Fatigue	2 (18)	1 (9)	0	0	1 (11)	1 (11)	0	0
	Febrile neutropenia	0	0	6 (55)	0	0	0	3 (33)	0
	Headache	3 (27)	0	0	0	3 (33)	0	0	0
	Hematuria	1 (9)	0	0	0	0	0	0	0
	Hypermagnesemia	0	0	0	0	1 (11)	0	0	0
	Hypoalbuminemia	0	1 (9)	0	0	1 (11)	0	0	0
	Infection	0	3 (27)b	0	0	0	1 (11)c	0	0
	Mucositis, oral	0	0	0	0	0	1 (11)	0	0
	Muscle weakness	0	0	0	0	0	1 (11)	0	0
	Myalgia	1 (9)	0	0	0	1 (11)	0	0	0
	Nausea	1 (9)	3 (27)	1 (9)	0	3 (33)	1 (11)	1 (11)	0
	Pharyngitis	0	1 (9)	0	0	0	0	0	0
	Rash	1 (9)	0	0	0	0	0	0	0
	Rectal pain	0	0	0	0	0	1 (11)	0	0
	Vomiting	4 (36)	1 (9)	1 (9)	0	4 (44)	0	1 (11)	0
	Weight loss	3 (27)	1 (9)	0	0	2 (22)	3 (33)	0	0

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatine phosphokinase; No., number; WBC, white blood cell.

^aOnly includes toxicities attributed as possibly, probably, or definitely related to protocol therapy.

^bIncludes upper respiratory infection (N = 2) and parainfluenza infection (N = 1).

^cOtitis externa.

3.3 |. Response

Eleven patients were evaluable for response. Initial response evaluation revealed one patient (9%) with Ewing sarcoma who had a PR (treated at DL0), SD in seven patients (58%), and PD in three patients (33%) for an ORR of 9%. BOR was PR in one patient (9%), confirmed SD in four patients (37%), PD in four patients (37%), and unknown in two patients (18%). Two Ewing sarcoma patients, one with PR and one with SD, came off protocol therapy after cycle 4 to receive local control measures that were not allowed on protocol.

3.4 |. Pharmacokinetic and pharmacodynamics studies

The PK profiles of simvastatin were studied in 13 patients during cycle 1 (Table 4 and Figure 1). A comparison of PK parameters by formulation (tablet vs. dispersed in water) is provided in Table S1. Simvastatin acid concentrations were consistently ~10%–20% of simvastatin concentrations. We report a patient treated at 100 mg/m² separately given the extreme PK values observed. Samples were re-assayed to confirm these data. Unique characteristics of this patient relative to the rest of the subjects with available PK data were age of 3 years versus median 12 years (range: 1.8–23), weight of 14.2 kg versus median 36 kg (range: 8.4–96.5), body surface area (BSA) of 0.62 m² versus median 1.25 m² (0.42–2.1), and rapidly PD resulting in death during the first week of therapy. This patient received simvastatin PO (not by NG or G tube) and was not taking any CYP3A4 inhibitors.

TABLE 4.

Simvastatin PK parameters in pediatric patients with recurrent or refractory solid and CNS tumors.^a

	Simvastatin								Simvastatin acid
DL (mg/m2) [n]	Cmax (ng/mL)	Tmax(h)	T1/2 (h)	AUC0–6 (ng h/mL)	AUCinfc (ng h/mL)	Cl (L/h/m2)	Cl (L/h/kg)	Cl (L/h)	Cmax (ng/mL)	AUC0–6 (ng h/mL)
100 [6]	62 (1.5)	2.5 (1.8)	1.7 (1.4)	174 (1.7)	220 (1.8)	438 (1.8)	13.1 (2.2)	576 (1.7)	11 (1.8)	34.5 (1.8)
140 [6]	101 (1.4)	1.2 (2.3)	1.8 (2.0)	271 (1.7)	323 (1.8)	434 (1.8)	14.4 (1.9)	500 (2.1)	15 (2.4)	47.0 (2.3)
100 [1]b	725	2.1	2.0	1992	2554	37.9	1.7	23.5	75	270

Abbreviations: AUC, area under the concentration versus time curve; Cl, clearance; C_max, peak plasma concentration; DL, dose level; PK, pharmacokinetic; T_1/2. half-life; T_max time of C_max.

^aShown as geomean (geometric SD).

^bOutlierdata reported separately.

^cExtrapolated portion of AUC: median = 11%; range = 4.5%–52%.

FIGURE 1.

Simvastatin pharmacokinetics in pediatric patients with recurrent or refractory solid and central nervous system (CNS) tumors. The maximum plasma concentration for each patient at both dose levels tested (A), and the AUC_0-t for each patient on both dose levels tested (B). Black bars represent the geometric mean value for each dose level; patients with dose-limiting toxicity (DLT) are indicated with an X; outlier not shown for visual purposes.

Peripheral blood samples for pharmacodynamic studies were available from six patients (Figure 2). For one patient (patient 008), the day 1 plasma specimen was not evaluable and so was excluded from interpretation. Patients 009 and 011 demonstrated high circulating levels of IL-6 at treatment onset, with a significant decrease toward normal values by day 21, indicating sustained IL-6 inhibition and potential on target effects (Figure 2A). In the remaining patients with samples available for analysis, circulating IL-6 levels were not elevated at treatment onset, so no appreciable decrease in response to simvastatin treatment could be observed (Figure 2A). IL-6Rα, another component of the IL-6 signaling cascade, remained constant in most patients throughout treatment with simvastatin supporting on target inhibition of only IL-6. Patient 010 was an outlier to this and demonstrated a significant decrease in circulating levels of IL-6Rα from day 1 to day 21 (Figure 2B). PBMC isolates for evaluation of STAT3 phosphorylation were available from four of the six patients with available samples (008, 009, 010, and 014). Each patient showed maximum pSTAT3 inhibition on day 5, with recurrence of phosphorylation by day 21 despite continued simvastatin dosing (Figure 2C). While patients’ cholesterol values did trend down while receiving simvastatin during cycle 1, there was no clear correlation between cholesterol and toxicity or outcome.

FIGURE 2.

Pharmacodynamic responses during cycle 1 in pediatric patients receiving simvastatin in combination with cytotoxic chemotherapy for relapsed or refractory solid and central nervous system (CNS) tumors. Circulating plasma levels of IL-6 (A) and IL-6Rα (B) were determined by commercially available quantitative ELISA on days 1, 5, and 21 (cycle 2, day 1) of treatment regimen. Each patient’s plasma specimen was tested in duplicate, and plasma concentration was determined from standard curve. Data are presented as mean with standard deviation. Statistical analysis represents results of unpaired t-test (**p < .01, *p < .05; ns, p > .05). (C) Percentage of PBMCs identified as positive for STAT3 phosphorylation (pSTAT3) determined by flow cytometry on days 1, 5, and 21 of treatment regimen for four patients.

4 |. DISCUSSION

In this first-in-pediatrics trial evaluating simvastatin in combination with chemotherapy in patients with r/r solid and CNS tumors, we determined that the MTD of simvastatin was 100 mg/m²/dose administered PO twice daily on days 1–21, with topotecan 0.75 mg/m²/dose IV and cyclophosphamide 250 mg/m²/dose IV on days 1–5. The three DLTs that occurred, grade 4 CPK increase (N = 2) and grade 3 diarrhea (N = 1), were predictable simvastatin-related toxicities. There were no later cycle DLTs. The non-DLT grade 3/4 treatment-related toxicity was almost exclusively hematologic, with limited nonhematologic toxicity. There was one objective response (ORR = 9%), with a BOR of SD in four patients. Pharmacodynamic correlative studies of plasma IL-6 concentrations and STAT3 phosphorylation support on target IL-6 inhibition both were decreased following simvastatin treatment.

In a dose-finding study in adult patients with r/r multiple myeloma or lymphoma, the MTD of simvastatin was determined to be 7.5 mg/kg/dose (~225 mg/m²/dose) when combined with standard chemotherapy.³² The DLTs were FN with sepsis and gastrointestinal (GI) toxicity (grade 3 nausea and grade 3 diarrhea). Other frequent toxicities were FN, low-grade GI toxicities, and fatigue; no rhabdomyolysis occurred. In our trial, due to concern for the potential effects of high-dose simvastatin on the CNS of developing children, we conservatively started DL1 at 140 mg/m²/dose, ~60% of the adult MTD. At this DL, two patients experienced simvastatin-related DLTs, grade 3 diarrhea, and grade 4 CPK increase (rhabdomyolysis), which precluded further escalation of the simvastatin dose. An additional patient on DL0 experienced grade 4 CPK increase. No other patient on the study experienced any grade CPK increase or diarrhea. Similar to the adult trial, patients on our study experienced low-grade GI toxicity as well as FN, but no patient experienced sepsis. Given that FN is an expected toxicity in pediatric patients receiving cytotoxic chemotherapy, grade 3 FN was not considered dose-limiting.

Rhabdomyolysis is an uncommon yet well-known side effect of statin therapy that occurs in a dose-related fashion. Predisposing risk factors include higher statin dose, renal impairment, female gender, and co-administration of drugs that impair simvastatin metabolism.¹ Two patients in our study experienced rhabdomyolysis. The first, a 5-year-old female with relapsed alveolar rhabdomyosarcoma treated on DL1 who, while admitted for management of malignant ascites, experienced clinically apparent myoglobinuria and myalgias and was found to have grade 4 CPK elevation on day 26 of cycle 1 (cycle 2 delayed due to hospitalization); the patient’s renal function was normal. This patient was dose-reduced to DL0 and went on to receive five additional cycles of protocol therapy without subsequent CPK elevation or symptoms of rhabdomyolysis. The second patient was a 12-year-old female with multiply relapsed Wilms tumor who had received three prior lines of chemotherapy and a nephrectomy, and entered the study with grade 1 creatinine elevation. The patient had grade 4 CPK elevation detected on her cycle 2, day 1 labs. Upon further questioning, the patient had mild myalgias and was noted to have myoglobinuria on urinalysis. Because the patient was treated on DL0, a dose reduction was not allowed and she was removed from protocol therapy. For both patients, the rhabdomyolysis resolved with aggressive IV hydration without ubiquinone supplementation. Neither was receiving medications known to impact simvastatin metabolism.

Rhabdomyolysis has rarely been reported in adult trials combining simvastatin with chemotherapy.^{17,19–21,24,25,32,33} However, most adult studies administered fixed doses of lower dose simvastatin (40–80 mg daily) as opposed to the high-dose simvastatin tested in our trial. In addition, patients in our study received high-dose simvastatin concurrently with chemotherapy, which may have increased the risk for rhabdomyolysis. While the lipophilic statins like simvastatin seemed to be ideal for anticancer purposes, given the potential for wider bodily distribution, the DLTs of rhabdomyolysis that occurred in pediatric patients may support looking at less lipophilic, more hydrophilic statins in future trials. Careful consideration for choosing hydrophilic statins that are not CYP3A substrates and that can be dosed as a liquid formulation (like was done with simvastatin) are imperative to maintain safety and to ensure accessibility to all ages of pediatric solid tumor patients.

The combination of topotecan and cyclophosphamide (topo/cy) was selected as the chemotherapy backbone in our study due to its activity against a variety of r/r pediatric solid tumors, the fact that it is generally well-tolerated, and its ability to be administered in the outpatient setting.³⁶ In addition, it has safely been used as the chemotherapy backbone in prior pediatric clinical trials testing new agents or novel combinations.^37–41 A phase 2 trial conducted by the Pediatric Oncology Group (POG) demonstrated that the toxicity of topo/cy was primarily hematologic with limited nonhematologic toxicity, and that this regimen had clinical activity in certain tumor types such as rhabdomyosarcoma, neuroblastoma, and Ewing sarcoma.³⁶ While making comparisons across trials is challenging, particularly in a heterogeneous, heavily pretreated patient population being treated in the context of a phase 1 trial, the ORR of 9% in our study was substantially lower than the ORR of 30% seen in the POG trial, suggesting that the simvastatin did not improve the efficacy of the chemotherapy combination. The limited responses coupled with the simvastatin-related toxicities that were dose limiting are not supportive of further testing of this regimen in a phase 2 trial; hence, further testing of this regimen is not planned.

Simvastatin and simvastatin acid C_max (geomean 62–110 and 11–15 ng/mL) and AUC_0–6 (geomean 174–271 and 34.5–47.0 ng h/mL) values observed were in line with reported pediatric literature values (C_max 3.5 and 0.4–2.1 ng/mL; AUC_0–8 10.7 and 3.8 ng h/mL) after correction for the ~24-fold higher doses (3.77 vs. 0.16 mg/kg) used in our study.⁴² The increase in simvastatin dose was associated with an increase in exposure, though there was significant overlap between the two dose groups due to the variability within each group. In addition to the rapidly progressing patient with almost 10-fold higher exposures relative to the other patients, the two other patients who experienced DLT also had relatively high exposure to simvastatin, one with the highest AUC and another with an AUC above the geometric mean. It would thus appear that there may be an exposure toxicity relationship for simvastatin in this population, and that attempts to achieve higher exposures and doses are likely to be limited by exposure-related toxicity.

Six patients received tablets dispersed in water. When the PK were evaluated by formulation type (tablets vs. dispersion), similar BSA-normalized doses were received, as opposed to absolute doses, which were lower in the dispersion group because smaller children were given the dispersion formulation due to their inability to swallow tablets whole. While similar C_max was experienced, the AUC appeared to be higher in patients receiving tablets, though this was driven by the extrapolated portion of the AUC. Absolute apparent clearance was lower in patients receiving the dispersion, again due to smaller sized children who received this formulation. Size-indexed apparent clearance (normalized to BSA or weight) appeared to be approximately two-fold higher in patients receiving dispersion. This is likely due to either inherent differences in drug metabolism between these groups of different-sized patients or due to a true impact of formulation on bioavailability. Given the correlation between size and formulation in our study, further PK studies would be required to explain these differences.

Correlative results demonstrated pSTAT3 levels decreased by day 5 and then returned to near baseline by day 21. The fact that topo/cy was given on days 1–5, while simvastatin was given on days 1–21 may suggest that the chemotherapy or the combination of simvastatin plus chemotherapy possibly contributed to the transient pSTAT3 decline that recovered once the chemotherapy completed. Yet, we cannot completely rule out a compensatory mechanism of pSTAT3 escape from simvastatin suppression through parallel pathways regulating STAT3, which may explain the lack of overall clinical efficacy. To validate these potential mechanisms would require further preclinical testing to model the treatment regimen and evaluate additional complementary pathways or proteins regulating STAT3 activation.

In conclusion, the MTD of simvastatin in pediatric patients with r/r solid and CNS tumors is 100 mg/m²/dose administered PO twice daily on days 1–21 with topotecan and cyclophosphamide on days 1–5. This combination was generally well-tolerated, with predominantly hematologic toxicity and predictable DLTs related to simvastatin. Beside the known statin-related toxicities, simvastatin did not add to the known toxicity or improve the clinical activity of the chemotherapy backbone. Future studies could consider testing a lower simvastatin dose, an alternative administration schedule, such as simvastatin monotherapy prior to initiating chemotherapy, or an alternative chemotherapy backbone to improve tolerability and ultimately response in select cohorts.

Abbreviations:

ANC

absolute neutrophil count

BOR

best overall response

BSA

body surface area

CBCd

complete blood count with differential

CNS

central nervous system

CPK

creatine phosphokinase

DL

dose level

DLT

dose-limiting toxicity

FN

febrile neutropenia

G

gastric

G-CSF

granulocyte colony-stimulating factor

GI

gastrointestinal

HMG-CoA

3-hydroxy-3-methylglutaryl-coenzyme A

IL-6

interleukin 6

IV

intravenously

mL

milliliter

MTD

maximum tolerated dose

NG

nasogastric

ORR

objective response rate

PD

progressive disease

PE

physical exam

PK

pharmacokinetic

PO

orally

PR

partial response

r/r

relapsed/refractory

RP2D

recommended phase 2 dose

SD

stable disease

STAT3

signal transducer and activator of transcription 3

EXPECTS DATA SHARING

The anonymized data that support the findings of this study are available on request from the corresponding author.