Glioblastoma multiforme (GBM) is the most common and aggressive type of primary malignant brain tumor [1] . Prognosis has been extremely poor, with a median survival of 14.6 months from diagnosis despite surgery, radiation therapy (RT), adjuvant temozolomide (TMZ) and bevacizumab (BVZ) [2] . The 5-year survival rates are between 1% and 2% for newly-diagnosed GBM patients [3] [4] . Median progression-free survival (PFS) after standard therapy is 6 to 9 months [5] and the prognosis for patients with recurrent GBM (RGBM) appears much worse, with the majority of patients dying within 6 months.

Our team evaluated treatment of GBM and other primary brain tumors with sodium phenylbutyrate (PB), a histone deacetylase (HDAC) inhibitor as well as with antineoplastons (ANP), a group of peptides, amino acid derivatives and carboxylic acids that inhibit the growth of neoplastic cells without inhibition of the growth of normal cells [6] - [14] . A recently published phase II study evaluated the safety and efficacy of ANP A10 and AS2-1 in recurrent high-grade glioma, with special emphasis to RGBM. Objective responses (OR) were determined in 17% of eligible patients, overall survival (OS) was 65.5% at 6 months; 56.7% at 9 months, 39% at 1 year and 4.4% at 2, 5, and 10 years [12] .

Parallel to the ANP studies, PB was also used in the treatment of patients with primary brain tumors. PB, approved for urea cycle disorders has also been used for the treatment of glioma and acute promyelocytic leukemia [15] [16] . PB shares its metabolites phenylacetylglutaminate (PG) and phenylacetate (PN) with some ingredients of ANP. The results of testing of PG and PN on the GBM genome indicate that PG and PN affect over 100 abnormal genes [17] . In 2012, a GBM classification into six biological subgroups had been proposed [18] . Each of the subgroups had distinct molecular and biological characteristics. PB and ANP have similar spectrums of activity on the GBM genome. Published data indicate that only some GBM patients will respond to monotherapy of ANP and PB [12] [14] [19] [20] . However, the activity of PB may not be as potent, since it is administered orally. The effect on over 100 abnormal genes in laboratory models of GBM suggests that it may not be sufficient to control the GBM genome that has an average of 650 abnormal genes. It was thus decided to treat advanced GBM patients who were not candidates for phase II studies with a combination of PB and additional targeted agents. The treatment plans of these patients were based on genomic profiling and consisted of PB for broad-spectrum coverage and selected targeted agents indicated by genomic abnormalities [14] . A number of ORs were determined within the course of treatment as was long-term OS that was considered highly unusual in RGBM. This publication provides a brief description of 11 evaluable cases of GBM treated with PB in combination with pazopanib, mammalian target of rapamycin (mTOR) inhibitors and other agents.

Patients were diagnosed with inoperable, recurrent, persistent or progressed GBM and received the treatment at Burzynski Clinic (BC) in Houston, Texas. Pathology and radiologic evaluations were performed by institutions not associated with BC. Tests performed by both outside laboratories and by the laboratory at BC included standard blood and urine analyses as well as a determination of genomic markers. Molecular profiling of tissue samples was performed by Foundation Medicine of Cambridge, Massachusetts. All patients read, understood and signed informed consent documents, which explained in detail the treatment prior to admission. The treatment plan, based on molecular profiling, included PB given in combination with targeted and chemotherapeutic agents. Therapy was performed on an out-patient basis, and after the initial two to four weeks of treatment at BC, patients continued their management under the care of local oncologists. Prior to the start of treatment, a magnetic resonance imaging (MRI) with and without contrast and in some instances a positron emission tomography (PET) scan was undertaken. The products of the two largest perpendicular diameters (LPD) of measurable lesions were calculated and totaled, providing a baseline evaluation for each study subject and a reference for determining response outcome to the treatment. Additional pretreatment measurements included Karnofsky Performance Status (KPS), vital signs, clinical disease status, demographics, medical history and current medications, physical examination, and electrocardiogram (EKG). Toxicity was evaluated according to Common Terminology Criteria for Adverse Events v3.0 (CTCAE v3.0). Possible response to treatment included complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). CR required the disappearance of all lesions confirmed at the end of four weeks, PR required 50% or higher decrease of the LPD of measurable lesions. PD was determined when there was over 25% increase of the lesions or new lesions, and SD was the status between PR and PD. The duration of each response was measured from the date that the criteria of the outcome were first met and until the date that PD was documented. In the case of SD, the duration was measured from the time that therapy had commenced.

Numerous chemotherapy and targeted therapy regimens have been evaluated for the treatment of patients with recurrent GBM [3] [21] . Some studies demonstrated minor improvement in PFS, but no significant increase in survival [21] [22] . A successful therapy for recurrent GBM is thus, still desperately required [3] [21] [22] .

The authors of this article suggest that treatment consisting of a combination of available and approved drugs, including PB, pazopanib, dasatinib, everolimus, and BVZ is currently an option.

In phase II studies of GBM with ANP, we identified a small group of patients with unusually long overall survival. Eligible patients in RGBM study exhibited an OR of 16.6% and increased OS compared to other studies. Unfortunately, only a small percentage of the patients obtained long-term OS. Understanding the spectrum of the effects of constituents of ANP on the targets in the GBM genome, we understood that control of GBM may require more than two agents [17] . PB is a drug with a similar spectrum of activity as ANP, but does not have as prominent an effect, since it is orally administered. Nonetheless, we chose to use PB together with selected drugs for treatment of patients who, except for one case, had recurrent GBM.

A key study conducted by the European Organization for Research and Treatment of Cancer and the National Cancer Institute of Canada Clinical Trials Group in patients with newly-diagnosed GBM provided the rationale for the current standard-of-care [4] [5] . Since 2005, this standard-of-care consists of tumor resection followed by standard RT with concomitant TMZ (75 mg/m²/d) for seven weeks and adjuvant TMZ (150 - 200 mg/m²/d) on 5 day therapy every 28 days. Based on these recommendations, a patient may expect a median OS of 14.9 months or a median PFS of 6.7 months. However, the recurrence rate is still over 90%, median OS is less than 18 months, and survival at 5 years is less than 2% of patients [3] [5] . Clinical studies on RGBM with chemotherapy and targeted therapy were recently compiled and published in excellent reviews [3] [21] . However, there has been only very modest progress in treatment of GBM and specifically there are no current standard recommendations for RGBM [3] [14] [21] [22] .

Based on research with PB from private practice patients and our experience in phase II studies with ANP, we propose a treatment plan for the upcoming phase I/II clinical studies of PB in combination with targeted agents, as well as with ANP [14] [17] . This treatment approach appears to address basic biological mechanisms in GBM, such as the control of growth, survival, invasion and migration of neoplastic cells, vascular effects, metabolism, maintenance and function of neoplastic stem cells, main signaling pathways, cell cycle mechanisms, apoptosis, autophagy, and drug resistance (Figure 7). The medications proposed regulate vascular endothelial growth factor (VEGF) signaling (BVZ, pazopanib), Src kinases (dasatinib), mTOR pathway (everolimus) and affect multiple targets of ANP and PB. These mechanisms have recently been discussed in detail elsewhere [14] [17] .

The treatment with PB with the combination of the four additional targeted medications resulted in a 54.5% rate of CRs and PRs and a 27.3% rate of SD. As described in Table 6, four patients fully complied with the treatment regimen and all achieved ORs. One, whose response was categorized as CR, is now leading a normal life for over one year and seven months since the treatment onset and three years and two months since tumor diagnosis. The remaining three patients are continuing treatment and appear to be improving on an ongoing basis. Molecular profiling did not provide extensive data to classify these patients in biological subgroups. Based on the available information, however, we can assign two patients to the isocitrate dehydrogenases 1 (IDH1) proneural subgroup (CR and PR), two additional patients to the ribosome biogenesis and trna synthetase-asso- ciated kinase (RTK1) proneural subgroup (2 PR), two to the classical subgroup (1 SD, 1 PD), one to the mixed tumor protein p53 (TP53) subgroup (PD) and one PR patient to an unclassified subgroup. These data suggest that IDH1 and RTK1 proneural subgroups may be associated with better treatment response. Recently, published data on long-term survival with GBM did not provide the information that would favor a longer survival in these subgroups [23] . None of the genomic characteristics associated with long-term survivals were present among our patients and the IDH mutation does not appear to impart a better survival benefit. Further studies are necessary in this area to afford a better understanding and relationship between genomic signature and treatment responses.

The authors thus propose new phase I/II clinical trials to be conducted with PB in combination with targeted agents BVZ, pazopanib, dasatinib and everolimus in patients with RGBM after failure of standard surgery, RT and TMZ, evaluating the survival, response to treatment and toxicity. Furthermore, molecular profiling on tumor tissue should be undertaken to identify “genomic signatures” of responders and non-responders. Based on the experience from private practice of the authors, some patients did not tolerate the proposed combination. Such patients may nonetheless still be helped by the substitution of pazopanib with sorafenib, a drug with a similar spectrum of action. As well, erlotinib or lapatinib can be used in cases with specific genomic targets for these drugs.

There is no established standard of care for RGBM. Numerous clinical studies with single chemotherapy and targeted agents or their combinations have shown some promising results, but progress has been modest at best. The treatment with PB in combination with targeted agents carries promise for a rapid and durable response in RGBM. The results reported in this paper provide evidence that it may be possible to accomplish a high response rate in RGBM with PB in combination with four targeted agents. The authors propose phase I/II clinical trials with PB in combination with targeted agents: BVZ, pazopanib, dasatinib and everolimus for patients with RGBM after failure of standard surgery, RT and TMZ. Caution should be exercised when combining these agents, since no clinical trials have yet been conducted with such combinations. With proper dose reduction, such treatment appears to be reasonably well-tolerated. Molecular profiling will ameliorate the selection of subgroups of RGBM having favorable genomic signature for future studies.

The authors express their appreciation to the additional physicians involved in the care of the patients: Drs. Zanhua Yi, Alejandro Marquis, Robert Weaver, Sheryl Acelar, Lourdes De Leon, and Mohammad Khan. Preparation of the manuscript was provided by Carolyn Powers, Jennifer Pineda and Adam Golunski.

Written informed consent was obtained from patients for publication of this article and accompanying images.

All authors are employed by Burzynski Clinic. Dr. Stanislaw R. Burzynski and Dr. Gregory S. Burzynski are shareholders and directors, and Dr. Tomasz J. Janicki is the Vice-President of Burzynski Research Institute, Inc. Dr. Stanislaw R. Burzynski is President of Burzynski Research Institute, Inc., Dr. Gregory S. Burzynski is Vice- President of Burzynski Clinic and Dr. Sheldon Brookman is Director of Pharmaceutical Development of Burzynski Clinic.