Concurrent whole brain radiotherapy and bortezomib for brain metastasis
© Lao et al.; licensee BioMed Central Ltd. 2013
Received: 25 March 2013
Accepted: 16 August 2013
Published: 21 August 2013
Survival of patients with brain metastasis particularly from historically more radio-resistant malignancies remains dismal. A phase I study of concurrent bortezomib and whole brain radiotherapy was conducted to determine the tolerance and safety of this approach in patients with previously untreated brain metastasis.
A phase I dose escalation study evaluated the safety of bortezomib (0.9, 1.1, 1.3, 1.5, and 1.7 mg/m2) given on days 1, 4, 8 and 11 of whole brain radiotherapy. Patients with confirmed brain metastasis were recruited for participation. The primary endpoint was the dose-limiting toxicity, defined as any ≥ grade 3 non-hematologic toxicity or grade ≥ 4 hematologic toxicity from the start of treatment to one month post irradiation. Time-to-Event Continual Reassessment Method (TITE-CRM) was used to determine dose escalation. A companion study of brain diffusion tensor imaging MRI was conducted on a subset of patients to assess changes in the brain that might predict delayed cognitive effects.
Twenty-four patients were recruited and completed the planned therapy. Patients with melanoma accounted for 83% of all participants. The bortezomib dose was escalated as planned to the highest dose of 1.7 mg/m2/dose. No grade 4/5 toxicities related to treatment were observed. Two patients had grade 3 dose-limiting toxicities (hyponatremia and encephalopathy). A partial or minor response was observed in 38% of patients. Bortezomib showed greater demyelination in hippocampus-associated white matter structures on MRI one month after radiotherapy compared to patients not treated with bortezomib (increase in radial diffusivity +16.8% versus 4.8%; p = 0.0023).
Concurrent bortezomib and whole brain irradiation for brain metastasis is well tolerated at one month follow-up, but MRI changes that have been shown to predict delayed cognitive function can be detected within one month of treatment.
KeywordsRadiation Brain Melanoma Bortezomib Phase I TITE-CRM Diffusion tensor imaging MRI
The development of metastatic disease to the brain often reflects a poor prognosis, with a median survival measured in months. Survival of patients with brain metastasis from historically more radio-resistant malignancies (i.e., melanoma and renal cell carcinoma) is particularly dismal. Of patients who develop symptomatic melanoma brain metastases, the metastases are fatal in up to 95% of cases . Despite the recent improvements in treatment for metastatic melanoma death from melanoma brain metastasis continues to be a critical barrier to improved survival [2–4].
Whole brain radiotherapy (WBRT) can have an impact on CNS progression, neurologic decline, and the likelihood of death from cerebral metastases, but such treatment has not demonstrated an overall survival benefit as primary therapy . Radio-sensitization has been attempted in melanoma but the results have been underwhelming . The promise of such an approach, however, is still important to explore. Bortezomib (VELCADE®, Millennium Pharmaceuticals, Inc.) is a proteasome inhibitor with preclinical and clinical data supporting activity against a variety of neoplasms. For melanoma, there does not appear to be much activity as a single agent but its potential role as a radiosensitizer is promising [7–9]. Concurrent use of bortezomib and radiation to treat metastatic disease to the brain has not been assessed previously, thus, a phase I study of concurrent bortezomib and whole brain radiotherapy in untreated patients predominantly with melanoma was conducted.
Men and women aged 18 years of age or older with a histopathologically confirmed solid tumor malignancy and clinical evidence of metastatic disease to the brain were considered for enrollment. The study was initially designed only for patients with melanoma and renal cell carcinoma, but was later expanded to other solid tumors with documented brain metastases. The study was approved by the University of Michigan Institutional Review Board (IRB) and signed informed consent was obtained for study enrollment. Study participation required one or more brain metastases on contrast-enhanced brain MRI for which WBRT was a treatment option, and in the judgment of the treating physician beginning additional systemic therapy (e.g., chemotherapy) could wait at least 30 days from completion of WBRT. Patients who had brain metastases managed with radiosurgery or surgery in the past were eligible as long as they had not received WBRT. Other enrollment criteria included: an estimated survival of at least 8 weeks, Karnofsky performance score (KPS) of at least 70%, adequate hematologic, hepatic, cardiovascular and renal function, and completion of previous chemotherapy at least 2 weeks before starting WBRT. Females of child bearing potential must have had confirmation they were not pregnant and acceptable contraception was required for the duration of the study.
Patients were considered ineligible if they had received previous radiotherapy to the head or neck, WBRT, or prior bortezomib therapy. Female patients who were pregnant or breast-feeding or patients with ≥ Grade 2 peripheral neuropathy within 14 days before enrollment, class III or IV congestive heart failure or serious concurrent cardiac disease, hypersensitivity to boron or mannitol, or serious medical or psychiatric illness likely to interfere with participation were ineligible.
Within seven days of registration, a complete history and general physical exam, including neurological examination was performed, as well as determination of KPS immediately prior to the beginning of protocol treatment. Documentation of steroid and anticonvulsant doses, complete blood count with differential and platelet count (CBCDP), and comprehensive chemistry profile were obtained. An MRI with gadolinium contrast was performed within 3 weeks prior to the start of therapy. During the treatment period, all participants were evaluated weekly by physical exam, including neurological examination and examination of the skin within the radiation treatment portal. Documentation of toxicities, KPS, and both steroid and anticonvulsant use were performed at each evaluation. A complete blood count with differential was performed prior to each dose of bortezomib. At the completion of WBRT, patients were evaluated once every two weeks for an additional month which completed study participation. Brain imagingwas obtained one month after completion of radiotherapy and then as clinically indicated approximately every 2 to 3 months. Following the first month post-irradiation, patients were seen and treated as clinically indicated.
Adequate immobilization and reproducibility of position was ensured using thermoplastic masks. The target volume included the whole brain and meninges to the foramen magnum. Doses were specified at central axis at mid-plane. Radiation was delivered with a daily fraction size of 2.5-3.0 Gy per fraction given 5 days a week, for a total dose of 30–37.5 Gy. The Radiation Oncology treating physicians were allowed to determine the most appropriate treatment fractionation scheme (30 Gy in 3 Gy fractions or 37.5 Gy in 2.5 Gy fractions) for patients independent of bortezomib. Treatment was delivered using photon beam energies of 6 megavolts (MV).
Bortezomib for this study was provided by Millennium Pharmaceuticals, Inc. (Cambridge, Massachusetts). It is a small molecule proteasome inhibitor currently approved by the United States Food and Drug Administration and registered in Europe for the treatment of multiple myeloma. Bortezomib was given by rapid IV injection beginning on day 1 of radiation, at least one hour prior to radiation treatment. During the radiation therapy period, bortezomib was given twice per week (at least 72 hours apart) for a total of 4 doses (given on days 1, 4, 8, and 11).
The bortezomib dose was assigned using TITE-CRM method . The primary objective of the study was to determine the dose of bortezomib associated with a 20% probability of DLT. A DLT was defined as the development of any ≥ grade 3 non-hematologic toxicity or ≥ grade 4 hematologic toxicity by National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 3.0 that was possibly, probably or definitely attributed to bortezomib and/or WBRT identified up to thirty days following treatment. Before each drug dose, the patient was evaluated for possible toxicities that may have occurred after the previous dose(s). The trial planned to accrue 30 patients evaluable for toxicity.
As in all CRM and TITE-CRM trials, before the enrollment of the first patient, the expected probability of DLT at each dose was elicited from the clinical investigators by the statistician, who used this information to construct the prior estimate of the dose-toxicity function, a one-parameter logistic function. As each patient completed the trial, the dose-toxicity function was re-estimated, and the re-estimated function was used as patients presented for enrollment to calculate the estimated probability of DLT at each dose. Each newly enrolled patient was thereby assigned to the bortezomib dose with estimated probability of toxicity closest to but not exceeding the target rate, 0.20.
The primary objective was to estimate the dose-toxicity function of bortezomib by means of a Bayesian logistic regression model, logit(π) = a + b × d, where π is the probability of toxicity, a and b are logistic regression parameters estimated from the dose and toxicity data, and d is dose of bortezomib. The secondary objective was to assess the response to the combination of bortezomib and WBRT by MRI or contrasted CT. Response criteria followed guidelines proposed by MacDonald et al. .
A companion University of Michigan IRB approved imaging study was conducted to assess MRI findings that might reflect changes in blood brain/tumor permeability in patients with brain metastases during and after radiotherapy. Patients enrolled in the Phase I bortezomib study were eligible to participate in the concurrent imaging study. Diffusion tensor imaging (DTI), which measures water diffusivity around axons and reflects the degree of demyelination, was used to evaluate a subset of patients enrolled to the concurrent WBRT-bortezomib study. All participating patients provided separate written informed consent for this study. Patients underwent brain MRI scans at three time points: 1–2 weeks pre-RT, within one week of finishing RT (end-RT), and one month after finishing RT (post-RT). All MRI scans were done on a single 3 T Phillips Achieva scanner. Functional Imaging Analysis Tool (FIAT), an imaging software package developed at the University of Michigan, was employed prior to Tract Based Spatial Statistics (TBSS). Details of these techniques and analysis were presented or published previously [12–15].
Patient and treatment characteristics
Patient and treatment characteristics at baseline
No. of patients
Gender, n (%)
Median age, years (range)
Underlying malignancy, n (%)
Non-small cell lung cancer
Renal cell carcinoma
No. previous systemic therapies
No. of brain metastases
5 (1- ≥ 10)
Neurologic symptoms at baseline, n (%)
Dexamethasone use, n (%)
Craniotomy*, n (%)
Anti-epileptic medication use, n (%)
Bortezomib dose (mg/m2/dose), n (%)
Radiation dose, n (%)
30 Gy (3 Gy fractions)
37.5 Gy (2.5 Gy fractions)
The bortezomib dose was escalated per-protocol via TITE-CRM from 0.9 mg/m2 to 1.7 mg/m2, with the largest cohort of patients (n = 9) receiving the highest dose level. A total radiation dose of 30 Gy was administered to 79% of patients and 37.5 Gy was administered to 21% of patients. All patients treated with 37.5 Gy (n = 5) were in the highest dose cohort (Table 1).
Adverse events possibly to definitely related to treatment
Bortezomib dose level*
0.9 (n = 5)
1.1 (n = 4)
1.3 (n = 4)
1.5 (n = 2)
1.7 (n = 9)
Response and survival
Response and survival
Best response in brain#
Study population (n = 24)*
Bortezomib dose group
0.9 (n = 5)
1.1 (n = 4)
1.3 (n = 4)
1.5 (n = 2)
1.7 (n = 9)¶
30 Gy (3 Gy fractions)
37.5 Gy (2.5 Gy fractions)†
Diffusion tensor imaging bortezomib interaction effect. Changes in radial diffusivity (RD) values by treatment group and white matter structures
Pre-RT to End-RT*
Pre-RT to 1 month post-RT#
Change in RD
Change in RD
Other white matter
6.3 – 14.5%
1.1 – 10.5%
7.5 – 18.8%
−1.4 – 11.1%
Other white matter
4.6 – 10.8%
4.8 – 11.5%
13.0 – 21.5%
12.4 – 21.2%
Bortezomib is a proteasome inhibitor with independent activity against various cancers and has demonstrated activity as a radiosensitizer in preclinical models . Much of the preclinical work on proteasome inhibition and radiation sensitization has focused on the role of nuclear factor kappa B (NF- κB) . In many cancer cell models, radiation treatment results in the activation of NF- κB which might act to inhibit apoptosis. In melanoma, however, as well as other cancer cell types, NF- κB is constitutively activated which could further mediate resistance to radiation therapy. Proteasome inhibition can result in the prevention of NF- κB activation, which may result in improved responses and outcomes. Other effects on DNA repair and survival pathways may also contribute, including suppression of homologous recombination, down-regulation of Bcl2, Bxl1, survivin, XIAP and increases in Fas/Fas ligand [16, 17].
Data on concurrent bortezomib and radiotherapy is limited. Pugh et al.  performed a phase I study of bortezomib and radiotherapy for patients who were candidates for palliative radiation to sites other than central nervous system and bone. Doses up to 1.6 mg/m2 IV for 4 doses starting on day 1 of radiotherapy were given. No DLTs were observed. O’Neil et al.  performed a phase I study of 5-fluorouracil and bortezomib concurrent with RT for rectal cancer. Diarrhea was the dose limiting toxicity at the 1 mg/m2/dose level but NF- κB activation did not appear to be suppressed at that dose. Kubicek et al.  performed a phase I of concurrent bortezomib (up to 1.3 mg/m2/dose) and temozolomide with WBRT for primary central nervous system malignancies (85% had glioblastoma). Forty-four percent had previously received RT. Grade 3 headache, syncope, neuropathy, hyponatremia, dyspnea, stupor were observed but none were considered DLTs.
In our study of concurrent bortezomib and whole brain irradiation, we administered one cycle of bortezomib and observed only two DLTs up to 1.7 mg/m2/dose up to one month post treatment in previously untreated patients with metastatic cancer. The patient who developed grade 3 encephalopathy was symptomatic at baseline from multiple brain metastases, was the oldest patient enrolled (76 years old), and was heavily pretreated for her disease (breast cancer, estrogen receptor/progesterone receptor positive, HER2/neu negative). These factors may have contributed to the severity of this adverse event and it is unclear how much the bortezomib added to radiation contributed to the observed encephalopathy.
Phase I clinical trials involving brain irradiation for metastatic disease can be challenging and alternative strategies to adequately monitor patients are critical . In this study, a Time-to-Event Continual Reassessment Method (TITE-CRM) design was used to optimize accrual with dose assignment determined by acute effects. Late radiation effects from brain irradiation, however, are exceptionally difficult to use as a dose limiting toxicity (DLT) in a phase I study. In the current study, the late radiation effects from irradiation were not formally assessed given the technical and practical difficulties in following a patient for months to determine toxicities prior to enrolling another patient. However, late effects are a critical part of assessing toxicities from brain irradiation as it is already recognized as a major issue in the pediatric survivorship population . A number of investigators have attempted to assess delayed toxicities from radiation using surrogate imaging markers . The understanding of the biologic causes of late cognitive decline are limited but evidence supports white matter loss resulting from demyelination and axonal degradation as key components which can be exploited as predictors of late cognitive impairment. Diffusion tensor imaging which assesses the diffusion of water around axons as a measure demyelination may be one of the best methods to determine this . In the current study, the number of patients participating in the companion study with adequate follow-up was limited but the findings are strongly suggestive of white matter damage that did not recover four weeks after completion of RT for those who received bortezomib. No formal assessment of cognition was included in this study, but late cognitive impairment could be expected if patients survived longer . With the expected improvement in therapeutic approaches for melanoma in particular, this issue will be critical .
Although bortezomib concurrent with WBRT did appear to have activity (59% had a PR, mPR or SD), which is historically better than what would be expected with WBRT alone for these typically radio-resistant malignancies, the median overall survival (5 months) was similar to what might be expected in this population at the time this study was conducted . It is not possible from this study to quantify the contribution of bortezomib to WBRT.
Bortezomib appears to be well tolerated up to 30 days after completion of administration of up to 1.7 mg/m2 when given concurrently with WBRT; however, changes in the normal brain were already evident at the end of RT and disease control appeared limited. Future studies testing concurrent brain irradiation with more promising therapeutic agents should employ techniques to monitor for changes in the normal brain that could predict delayed cognitive impairment.
Thank you to Patricia Jo Harvey, Alisa Stager, Peg Esper, Samkeliso Mawocha, Fabian Rollins, Anders Lindquist, Theresa Royce-Westcott, and Monica Orians for the coordination and care of all the patients.
Millennium Pharmaceuticals, Inc. (Cambridge, MA) provided bortezomib (VELCADE®) and limited research support to conduct study. No salary support was provided. The work was supported in part by NIH/NINDS/NCI RO1 NS064973 and NIH/NCI R21 CA113699.
- Sampson JH, Carter JH Jr, Friedman AH, Seigler HF: Demographics, prognosis, and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998, 88: 11-20. 10.3171/jns.1998.88.1.0011View ArticlePubMedGoogle Scholar
- Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, Gonzalez R, Robert C, Schadendorf D, Hassel JC, et al.: Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 2010, 363: 711-723. 10.1056/NEJMoa1003466View ArticlePubMedPubMed CentralGoogle Scholar
- Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, Dummer R, Garbe C, Testori A, Maio M, et al.: Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med 2011, 364: 2507-2516. 10.1056/NEJMoa1103782View ArticlePubMedPubMed CentralGoogle Scholar
- Margolin K, Ernstoff MS, Hamid O, Lawrence D, McDermott D, Puzanov I, Wolchok JD, Clark JI, Sznol M, Logan TF, et al.: Ipilimumab in patients with melanoma and brain metastases: an open-label, phase 2 trial. Lancet Oncol 2012, 13: 459-465. 10.1016/S1470-2045(12)70090-6View ArticlePubMedGoogle Scholar
- Kocher M, Soffietti R, Abacioglu U, Villa S, Fauchon F, Baumert BG, Fariselli L, Tzuk-Shina T, Kortmann RD, Carrie C, et al.: Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: results of the EORTC 22952–26001 study. J Clin Oncol 2011, 29: 134-141. 10.1200/JCO.2010.30.1655View ArticlePubMedPubMed CentralGoogle Scholar
- Margolin K, Atkins B, Thompson A, Ernstoff S, Weber J, Flaherty L, Clark I, Weiss G, Sosman J, W IIS, et al.: Temozolomide and whole brain irradiation in melanoma metastatic to the brain: a phase II trial of the Cytokine Working Group. J Cancer Res Clin Oncol 2002, 128: 214-218. 10.1007/s00432-002-0323-8View ArticlePubMedGoogle Scholar
- Markovic SN, Geyer SM, Dawkins F, Sharfman W, Albertini M, Maples W, Fracasso PM, Fitch T, Lorusso P, Adjei AA, Erlichman C: A phase II study of bortezomib in the treatment of metastatic malignant melanoma. Cancer 2005, 103: 2584-2589. 10.1002/cncr.21108View ArticlePubMedGoogle Scholar
- Pervan M, Pajonk F, Sun JR, Withers HR, McBride WH: Molecular pathways that modify tumor radiation response. Am J Clin Oncol 2001, 24: 481-485. 10.1097/00000421-200110000-00013View ArticlePubMedGoogle Scholar
- Russo SM, Tepper JE, Baldwin AS Jr, Liu R, Adams J, Elliott P, Cusack JC Jr: Enhancement of radiosensitivity by proteasome inhibition: implications for a role of NF-kappaB. Int J Radiat Oncol Biol Phys 2001, 50: 183-193. 10.1016/S0360-3016(01)01446-8View ArticlePubMedGoogle Scholar
- Normolle D, Lawrence T: Designing dose-escalation trials with late-onset toxicities using the time-to-event continual reassessment method. J Clin Oncol 2006, 24: 4426-4433. 10.1200/JCO.2005.04.3844View ArticlePubMedGoogle Scholar
- Macdonald DR, Cascino TL, Schold SC Jr, Cairncross JG: Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 1990, 8: 1277-1280.PubMedGoogle Scholar
- Cao Y: Development of image software tools for radiation therapy assessment. Med Phys 2005, 32: 2136.View ArticleGoogle Scholar
- Chapman CH, Nazem-Zadeh M, Lee OE, Schipper MJ, Tsien CI, Lawrence TS, Cao Y: Regional variation in brain white matter diffusion index changes following chemoradiotherapy: a prospective study using tract-based spatial statistics. PLoS One 2013, 8: e57768. 10.1371/journal.pone.0057768View ArticlePubMedPubMed CentralGoogle Scholar
- Mori S, Oishi K, Jiang H, Jiang L, Li X, Akhter K, Hua K, Faria AV, Mahmood A, Woods R, et al.: Stereotaxic white matter atlas based on diffusion tensor imaging in an ICBM template. Neuroimage 2008, 40: 570-582. 10.1016/j.neuroimage.2007.12.035View ArticlePubMedPubMed CentralGoogle Scholar
- Mazziotta J, Toga A, Evans A, Fox P, Lancaster J, Zilles K, Woods R, Paus T, Simpson G, Pike B, et al.: A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philos Trans R Soc Lond B Biol Sci 2001, 356: 1293-1322. 10.1098/rstb.2001.0915View ArticlePubMedPubMed CentralGoogle Scholar
- Wiegner EA, Daly ME, Murphy JD, Abelson J, Chapman CH, Chung M, Yu Y, Colevas AD, Kaplan MJ, Fischbein N, et al.: Intensity-modulated radiotherapy for tumors of the nasal cavity and paranasal sinuses: clinical outcomes and patterns of failure. Int J Radiat Oncol Biol Phys 2012, 83: 243-251. 10.1016/j.ijrobp.2011.05.044View ArticlePubMedGoogle Scholar
- Goel A, Dispenzieri A, Greipp PR, Witzig TE, Mesa RA, Russell SJ: PS-341-mediated selective targeting of multiple myeloma cells by synergistic increase in ionizing radiation-induced apoptosis. Exp Hematol 2005, 33: 784-795. 10.1016/j.exphem.2005.04.005View ArticlePubMedGoogle Scholar
- Pugh TJ, Chen C, Rabinovitch R, Eckhardt SG, Rusthoven KE, Swing R, Raben D: Phase I trial of bortezomib and concurrent external beam radiation in patients with advanced solid malignancies. Int J Radiat Oncol Biol Phys 2010, 78: 521-526. 10.1016/j.ijrobp.2009.07.1715View ArticlePubMedGoogle Scholar
- O’Neil BH, Raftery L, Calvo BF, Chakravarthy AB, Ivanova A, Myers MO, Kim HJ, Chan E, Wise PE, Caskey LS, et al.: A phase I study of bortezomib in combination with standard 5-fluorouracil and external-beam radiation therapy for the treatment of locally advanced or metastatic rectal cancer. Clin Colorectal Cancer 2010, 9: 119-125. 10.3816/CCC.2010.n.017View ArticlePubMedPubMed CentralGoogle Scholar
- Kubicek GJ, Werner-Wasik M, Machtay M, Mallon G, Myers T, Ramirez M, Andrews D, Curran WJ Jr, Dicker AP: Phase I trial using proteasome inhibitor bortezomib and concurrent temozolomide and radiotherapy for central nervous system malignancies. Int J Radiat Oncol Biol Phys 2009, 74: 433-439. 10.1016/j.ijrobp.2008.08.050View ArticlePubMedPubMed CentralGoogle Scholar
- Cao Y, Tsien CI, Sundgren PC, Nagesh V, Normolle D, Buchtel H, Junck L, Lawrence TS: Dynamic contrast-enhanced magnetic resonance imaging as a biomarker for prediction of radiation-induced neurocognitive dysfunction. Clin Cancer Res 2009, 15: 1747-1754. 10.1158/1078-0432.CCR-08-1420View ArticlePubMedPubMed CentralGoogle Scholar
- Chang EL, Wefel JS, Hess KR, Allen PK, Lang FF, Kornguth DG, Arbuckle RB, Swint JM, Shiu AS, Maor MH, Meyers CA: Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol 2009, 10: 1037-1044. 10.1016/S1470-2045(09)70263-3View ArticlePubMedGoogle Scholar
- Jeraj R, Cao Y, Ten Haken RK, Hahn C, Marks L: Imaging for assessment of radiation-induced normal tissue effects. Int J Radiat Oncol Biol Phys 2010, 76: S140-S144. 10.1016/j.ijrobp.2009.08.077View ArticlePubMedPubMed CentralGoogle Scholar
- Chapman CH, Nagesh V, Sundgren PC, Buchtel H, Chenevert TL, Junck L, Lawrence TS, Tsien CI, Cao Y: Diffusion tensor imaging of normal-appearing white matter as biomarker for radiation-induced late delayed cognitive decline. Int J Radiat Oncol Biol Phys 2012, 82: 2033-2040. 10.1016/j.ijrobp.2011.01.068View ArticlePubMedPubMed CentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.