|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2006.06.3891 on September 5 2006 © 2006 American Society of Clinical Oncology. Correlations Between O6-Methylguanine DNA Methyltransferase Promoter Methylation Status, 1p and 19q Deletions, and Response to Temozolomide in Anaplastic and Recurrent Oligodendroglioma: A Prospective GICNO Study
From the Department of Medical Oncology and Pathology of Bellaria Hospital, Bologna; Department of Medical Oncology, Istituto Oncologico Veneto; Servizio Immunologia e Diagnostica Molecolare Oncologica, Istituto Oncologico Veneto; Pathology, Neurological Sciences, Azienda Ospedale-Università of Padova, Padova; Department of Oncology, San Raffaele Hospital, Milan; Department of Pathology Verona Hospital, Verona; and the Department of Pathology, Belluno Hospital, Belluno, Italy Address reprint requests to Alba A. Brandes, MD, Department of Medical Oncology, Bellaria Hospital, Via Altura 3, Bologna, Italy; e-mail: aa.brandes{at}yahoo.it
PURPOSE: To date, no data are available on the relationship between 1p/19q deletions and the response to temozolomide (TMZ) in primary anaplastic oligodendroglioma (AO) and anaplastic oligoastrocytoma (AOA) recurrent after surgery and standard radiotherapy. The aim of this study was to evaluate correlations between 1p/19q deletions, O6-methylguanine DNA methyltransferase (MGMT) promoter methylation, and response rate to TMZ in this setting. PATIENTS AND METHODS: From June 2000 to February 2005, 67 patients were enrolled; 39 patients (58%) had AO and 28 patients (42%) had AOA. All patients received 150 to 200 mg/m2 of TMZ every 28 days. Chromosome 1p and 19q deletions were detected by fluorescence in situ hybridization and MGMT promoter methylation was analyzed using methylation specific polymerase chain reaction. RESULTS: The overall response rate was 46.3% (17 complete responses and 14 partial responses). The response rate was higher in patients with AO than in those with AOA (61.5% v 25%, P = .003). Combined 1p/19q allelic loss was found in 32 patients (47.8%), while MGMT methylation occurred in 37 (68.5%) of 54 assessable patients. 1p/19q loss was significantly correlated with response rate (P = .04), time-to-progression (P = .003), and overall survival (P = .0001). Despite the significant concordance found between MGMT promoter methylation and 1p/19q deletions (P = .02), MGMT promoter methylation showed only a borderline correlation with overall survival (P = .09). CONCLUSION: TMZ is active in anaplastic oligodendroglial tumors treated at first recurrence. In this setting, 1p/19q allelic loss is an important predictive and prognostic factor. Further studies on MGMT promoter methylation should be performed in randomized trials to test its correlation with survival.
Oligodendrogliomas are widely believed to be chemotherapy sensitive. The procarbazine, lomustine, and vincristine (PCV) regimen1 was long used as standard chemotherapy for recurrent oligodendrogliomas, achieving an overall response rates of 76%,1 63%,2 and 59.4%,3 with a median time-to-progression (TTP) of 16 months, 10 months, and 12.3 months, respectively. It has been demonstrated that the combined loss of the short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q) are significantly associated with response to chemotherapy and longer progression-free survival (PFS) and overall survival (OS) after PCV chemotherapy. However, the administration of this regimen is troublesome, due to intercurrent cumulative hematological toxicity that often hampers drug administration3; temozolomide (TMZ)4 has therefore replaced PCV as standard first-line chemotherapy treatment because of its better toxicity profile. In patients with glioblastoma, a significant correlation was recently found between survival and O6-methylguanine-DNA methyltransferase (MGMT).5 The methylation of the promoter of MGMT turns off gene transcription, thus reducing the intracellular level of MGMT and inhibiting the repair mechanism. So far, no data are available on correlations between MGMT status, 1p and 19q loss and TMZ chemotherapy in patients with oligodendroglioma.
Eligibility Criteria Our series consisted of 67 patients enrolled between June 2000 and February 2005, with pathology findings of anaplastic oligodendroglioma (AO) or anaplastic oligoastrocytoma (AOA). Oligodendroglioma (presence of only oligodendroglial cells) was considered anaplastic if the tumor had at least three of the following histopathologic features: increased cellularity, marked cytologic atypia, high mitotic activity, microvascular proliferation, and/or necrosis with or without pseudopalisading.6 Oligoastrocytomas have oligodendroglial and astrocytic components. As there is no consensus regarding criteria for a diagnosis of mixed glioma in literature, we considered as mixed glioma, biphasic and intermingled tumors containing more than 25% oligodendroglial elements plus an astrocytic component, which was graded according to the WHO criteria.6 The oligodendroglioma component was graded separately using the same criteria.6 All pathology samples were reviewed by a panel of three neuropathologists (M.G., P.I., C.G.).
Patients accrued in the study presented disease progression at magnetic resonance imaging performed 3 months after the end of radiotherapy, with at least one enhancing bidimensionally measurable lesion of 2 cm or larger in diameter. In patients who underwent surgery for recurrence, brain imaging performed within 2 days after resection showing residual disease with the aforementioned characteristics, was mandatory. All patients were chemotherapy naïve, 18 years of age or older, and had a Karnofsky performance score (KPS) of 50 or higher, a life expectancy of at least 8 weeks, an adequate bone marrow reserve (leukocyte count 3.0 x 109/L or higher, granulocyte count 1.5 x 109/L or higher, and platelet count 100 x 109/L or higher), normal baseline liver (AST
Treatment Plan Hematologic parameters were assessed at baseline and on days 21 and 28 of each cycle. Electrolytes, renal function, and hepatic function were assessed at baseline and on day 21 of each cycle. The response was evaluated every two cycles, and patients with complete response had follow-up controls including magnetic resonance image scanning every 3 months. All response evaluations were performed according to Macdonald's criteria.7
Off-Study Criteria
Assessment of 1p and 19q Status Analysis was carried out with a fluorescence Zeiss microscope Axioskop 2 Plus (Carl Zeiss, Gottingen, Germany) equipped with single band filters. For each probe set, 60 nonoverlapping nuclei were enumerated and results were reported as the ratio between number of 1p/1q and 19p/19q signals (Figs. 1 and 2). Cutoff values for designation of deletion were reported following Smith recommendations.8
MGMT Status Assessment DNA from 10 µm paraffin sections of cerebral lesion was modified by sodium bisulfite, which converts unmethylated, but not methylated, cytosine to uracil, according to the procedure of Herman et al.9 Modified DNA was submitted to methylation specific polymerase chain reaction (MSP) after a nested-polymerase chain reaction protocol.10 Because the quality of DNA obtained from formalin-fixed paraffin-embedded tumor tissue affects the success rate of MSP, in some cases MGMT methylation status was determined using a different nested MSP approach, with a first pair of primers to obtain smaller amplicons (129 base pairs; Fig 3), for which forward and reverse primers have been described.10,11 We considered the hypothesis of restriction enzyme digestion by BST UI and Taq I, but we did not use these enzymes in our study.10 However, the restriction site for both enzymes was included in our forward primer for methylated sequences,12 and the restriction map of the amplified methylated region did not show any other site useful for demonstrating a C or T in a CpG site. We therefore verified our results using a second step of both modification and nested polymerase chain reaction; sometimes we repeated the entire process in triplicate.
Statistical Design The primary end point of the study was to evaluate the response rate of TMZ as first-line chemotherapy treatment in recurrent AO and AOA; the secondary end points were to study the correlations between 1p/19q deletions, MGMT status, and this treatment in terms of PFS and survival.
The Minimax Simon two-stage phase II design was utilized with the following: P0 = 0.35; P1 = 0.55; PFS and OS were measured from the start of chemotherapy. OS was measured until the date of death or last follow-up examination. PFS was measured until progression or otherwise to the last follow-up visit. OS and PFS were estimated using the Kaplan-Meier method.
Differences with regard to survival and disease progression were tested for statistical significance using the log-rank test. To determine truly independent variables (ie, prognostic factors), multivariate analysis using the Cox proportional hazards model was performed. The
Patients' Characteristics From June 2000 to February 2005, 67 patients were enrolled; their characteristics are outlined in Table 1. Histology of AO was present in 27 of 40 patients with a frontal site and in 12 (44.4%) of 27 patients with a tumor elsewhere (P = .06). The mean age of patients with 1p and 19q loss (51 years ± 9.4) was higher than that of patients without deletions (44.8 ± 13 years; P < .03). The presence of 1p and 19 q loss was significantly greater in AO (P = .00003) and in patients who had tumors with a frontal localization (P = .003).
MGMT promoter methylation, present in 37 (68.5%) of 54 assessable patients, in particular, in 21 (84%) of 25 patients with 1p and 19q deletion and in 16 (55.2%) of 29 patients without deletion (P = .02), was higher in patients with AO (70.3%) than in patients with AOA (29.7%; P = .10). No differences were observed between the mean age of the patients with (49.7 ± 11) or without (44.7 ± 12) MGMT promoter methylation (P = .14). Eighty-one percent of the patients with methylated status were older than 40 years of age versus 65% in the nonmethylated group.
Response At univariate analysis (Table 2), categoric variables found to be significantly correlated with response (CR + PR) were 1p and 19q loss and histology of AO, unlike a history of seizures, MGMT status, tumor site, and previous low grade. Neither KPS nor age were found to be significantly related to response (Mann Whitney U test, P = .15 and P = .87, respectively).
At multivariate analysis, performed using logistic regression, both deletion and histology persisted as variables independently related to response.
PFS
At univariate analysis (Table 3), response to therapy (CR + PR), 1p/19q loss and KPS (
At multivariate analysis, response to treatment (CR + PR, P = .0002), 1p/19q loss (P = .02), and KPS (P = .02) persisted as significant independent prognostic factors.
Survival Analysis At univariate analysis (Table 4), 1p/19q loss, frontal localization, and response to therapy (CR + PR) were found to be significantly correlated to OS. Performance status and MGMT were near to significance, unlike epilepsy, age, histology, and previous low grade (Figs. 5 and 6).
At multivariate analysis, performed on the entire group, only 1p/19q loss (P = .00003) and performance status (P = .005) persisted as significant independent prognostic factors. No change was found in this result on considering the subgroup (54 patients) with MGMT status: methylation was not found to be a significant prognostic factor at multivariate analysis.
Toxicity
In oligodendroglioma patients, PCV chemotherapy is considered the standard chemotherapy regimen for both adjuvant therapy13,14 and for treatment of first recurrence.1,3,4 In this study, TMZ was administered to patients with demographics and disease characteristics similar those in our previous study with PCV.3 The response rate in this study was slightly lower (46%; 95% CI, 34% to 58%), the median PFS (12 months) achieved was comparable, and the percentage of patients alive at 1 year was higher (84% v 50%). Studies on the correlation between PCV and 1p/19q loss in AOs have reported response rates ranging from 93% to 100% in patients with 1p/19q loss tumors.15-17 Cairncross et al17 observed a 100% response rate in 22 AO patients with 1p/19q loss tumors. It should, however, be noted that ring enhancement at neuroimaging, although uncommon, can negatively affect the response rate. van den Bent et al16 used the PCV regimen in 15 recurrent oligodendroglioma patients with 1p/19q deletions and while the response rate (93%) was higher than ours (59.4%), the median TTP was comparable (18 months16 v 22 months). In our study, the response rate of 59.4% was significantly correlated with 1p/19q loss and histology (pure oligodendroglioma v oligoastrocytoma). The outcome of nonrandomized studies, such as phase II studies, depends on the specific characteristics of the patient population studied. Selection bias may be due not only to known prognostic variables, such 1p/19q deletions, histology and neuroimaging features, but also to unknown or unstudied variables. Random fluctuations occur when measuring a biologic variable over time, this phenomenon being intensified in phase II studies on small series, in which random tumor responses greatly influence the response rate; this type of influence can be reduced by increasing the number of patients enrolled in phase II studies.18 To our knowledge, ours is the first prospective study to demonstrate a significant correlation between response to TMZ and 1p/19q loss in patients with AO tumors (P = .004) and, in agreement with a previous report,17 the chromosomal alterations were found to be correlated with a frontal lobe location, and a history of pure oligodendroglioma. In contrast with data reported by van den Bent et al,16 we found that patients with chromosomal deletions tended to be older than those without this genetic signature (P = .03); this finding suggests that 1p/19q deletions are better predictors of outcome than other known prognostic factors. PFS was correlated with response to treatment (P = .00003), 1p/19q loss (P = .003), KPS (P = .03), and age (P = .04), while OS was correlated only with 1p/19q loss (P = .0001) and performance status (P = .01). Moreover, we performed a simultaneous evaluation of the prognostic significance of MGMT promoter methylation status and 1p/19q loss. Epigenetic DNA alterations, such as methylation of promoter regions of many genes, are frequent events in all types of tumors. Hypermethylation of the gene promoter is associated with loss of gene function, methylated DNA being less accessible to transcription factors and resulting in gene silencing. The DNA-repair protein, O6-alkylguanine DNA alkyltransferase (AGT), produced by the MGMT gene, is a key factor in resistance to alkylating agents, because the transfer of alkyl groups to AGT prevents the formation of lethal cross links in DNA12,19; the negative prognostic effect of this epigenetic signature in patients with glioblastoma was recently reported.5 Few data with a wide range of MGMT methylation rates (25% to 85%) are currently available in AO tumors; those available have been extrapolated from studies on small series comprising mixed glioma patients. Brell et al20 found no correlation between survival or PFS and MGMT promoter methylation in 40 anaplastic glioma patients treated with nitrosourea-based chemotherapy; MGMT promoter methylation was observed in 50% of the tumors analyzed. However, in their study, only eight patients had anaplastic oligodendroglial tumors and no data on MGMT status relevant for survival and PFS in this subgroup of patients were available. In a study by other authors21 only 12 (30%) of 40 glioma patients treated with first-line TMZ had MGMT promoter methylation; of these patients, eight responded to treatment (67.7% v 25% in the nonmethylated population). No significant correlations with TTP and OS were found. In the subgroup of 20 oligodendroglial patients, MGMT promoter methylation occurred in six (25%), and neither data on the predictive or prognostic role of MGMT or correlations with 1p/19q loss were described. Recently, Molleman et al22 found that MGMT methylation, which occurred in 46 (88%) of 52 oligodendroglial tumors analyzed, was present in a significantly greater number of tumors with 1p and 19q loss of heterozygosity; however, no data were reported on the predictive or prognostic role of MGMT promoter methylation in this setting. Similar data were obtained by Alonso et al.23 Finally, Dong et al24 showed MGMT promoter methylation in 26 (60%) of 43 patients with oligodendroglial tumors. This molecular alteration was found in 66.7% of patients in the anaplastic tumors subgroup. Moreover, hypermethylation of MGMT was significantly associated with 1p/19q allelic loss. In our study, the overall MGMT promoter methylation rate was 68.5%; it was 70% for AO and 29.7% for AOA, the difference almost attaining statistical significance (P = .10), probably due to the small sample size. As reported by other authors,22-24 we found that the frequency of MGMT promoter methylation was significantly higher in patients bearing tumors with combined 1p/19q loss (84%; P = .02). In our series, the response rate and TTP were not correlated with MGMT methylation status and survival almost attained significance (P = .09). This is in contrast with studies on MGMT status in glioblastoma, in which promoter methylation was found to be correlated with response to alkylating agents12,25 and with survival in glioblastoma patients treated with TMZ.5 A distinctive pattern of promoter methylation may be present in oligodendroglial and astrocytic cells. In line with this hypothesis, two studies23,24 reported that MGMT methylation is frequently associated with promoter methylation of different genes in patients with oligodendroglial or astrocytic tumors. The PCV regimen leads to cumulative and persistent toxicity, which reduces the relative dose intensity of this regimen, it started at cycle 3 and frequently called for discontinuation of treatment.3 Unlike nitrosoureas, TMZ does not incur cumulative toxicities, and is therefore tolerated better than PCV. In agreement with a previous report,4 TMZ toxicity was mild, and only five patients (7%) had grade 4 hematologic toxicity. Interestingly, all of these patients presented MGMT methylation status, thus suggesting that baseline AGT levels can vary26 in relation to tumor and bone marrow precursors, and that AGT protects bone marrow cells from the cytotoxic consequences of alkylating agents. Perhaps, in view of its greater tolerability and the lower risk of cumulative toxicity, TMZ should replace the PCV combination as first-line treatment for patients with AOs, even in absence of randomized phase III trials. In conclusion, 1p and 19q deletion and histology are predictive of response to TMZ chemotherapy; PFS is significantly correlated with performance status and age, while OS is predicted by 1p/19q deletions and performance status, MGMT promoter methylation being of borderline significance. The natural history of cancer is unpredictable. However, based on well-recognized prognostic factors, clinicians can now confidentially predict which patient will respond to treatment, thus designing and tailoring the best possible treatment for the individual patient. Notwithstanding the achievements made, many aspects of this issue remain controversial and require further investigation in order to improve on our knowledge of oligodendroglial tumors.
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Dollar Amount Codes (A) < $10,000 (B) $10,000-$99,999 (C)
published online ahead of print at www.jco.org on September 5, 2006 Terms in blue are defined in the glossary, found at the end of this article and online at www.jco.org. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Cairncross G, Macdonald D, Ludwin S, et al: Chemotherapy for anaplastic oligodendroglioma: National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 12:2013-2021, 1994 2. van den Bent MJ, Kros JM, Heimans JJ, et al: Response rate and prognostic factors of recurrent oligodendroglioma treated with procarbazine, CCNU, and vincristine chemotherapy: Dutch Neuro-oncology Group. Neurology 51:1140-1145, 1998 3. Brandes AA, Tosoni A, Vastola F, et al: Efficacy and feasibility of standard procarbazine, lomustine, and vincristine chemotherapy in anaplastic oligodendroglioma and oligoastrocytoma recurrent after radiotherapy: A phase II study. Cancer 101:2079-2085, 2004[CrossRef][Medline] 4. van den Bent MJ, Taphoorn MJ, Brandes AA, et al: Phase II study of first-line chemotherapy with temozolomide in recurrent oligodendroglial tumors: The European Organization for Research and Treatment of Cancer Brain Tumor group study 26971. J Clin Oncol 21:2525-2528, 2003 5. Hegi ME, Diserens AC, Gorlia T, et al: MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 352:997-1003, 2005 6. Kleihues P, Louis DN, Scheithauer BW, et al: The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 61:215-225, 2002; discussion 226-229, 2002[Medline] 7. Macdonald DR, Cascino TL, Schold SC Jr, et al: Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 8:1277-1280, 1990[Abstract] 8. Smith JS, Perry A, Borell TJ, et al: Alterations of chromosome arms 1p and 19q as predictors of survival in oligodendrogliomas, astrocytomas, and mixed oligoastrocytomas. J Clin Oncol 18:636-645, 2000 9. Herman JG, Graff JR, Myohanen S, et al: Methylation-specific PCR: A novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci U S A 93:9821-9826, 1996 10. Palmisano WA, Divine KK, Saccomanno G, et al: Predicting lung cancer by detecting aberrant promoter methylation in sputum. Cancer Res 60:5954-5958, 2000 11. van Engeland M, Weijenberg MP, Roemen GM, et al: Effects of dietary folate and alcohol intake on promoter methylation in sporadic colorectal cancer: The Netherlands cohort study on diet and cancer. Cancer Res 63:3133-3137, 2003 12. Esteller M, Garcia-Foncillas J, Andion E, et al: Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 343:1350-1354, 2000 13. Cairncross G, Seiferheld W, Shaw E, et al: An Intergroup randomized controlled clinical trial (RCT) of chemotherapy plus radiation (RT) versus RT alone for pure and mixed anaplastic oligodendrogliomas: Initial report of RTOG 94-02. J Clin Oncol 22:107s, 2004 (abstr 1500) 14. van den Bent MJ, Delattre J, Brandes AA, et al: First analysis of EORTC trial 26951, a randomized phase III study of adjuvant PCV chemotherapy in patients with highly anaplastic oligodendroglioma. J Clin Oncol 23: 114s, 2005 (abstr 1503) 15. Ino Y, Betensky RA, Zlatescu MC, et al: Molecular subtypes of anaplastic oligodendroglioma: Implications for patient management at diagnosis. Clin Cancer Res 7:839-845, 2001 16. van den Bent MJ, Looijenga LH, Langenberg K, et al: Chromosomal anomalies in oligodendroglial tumors are correlated with clinical features. Cancer 97:1276-1284, 2003[CrossRef][Medline] 17. Cairncross JG, Ueki K, Zlatescu MC, et al: Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 90:1473-1479, 1998 18. Zia MI, Siu LL, Pond GR, et al: Comparison of outcomes of phase II studies and subsequent randomized control studies using identical chemotherapeutic regimens. J Clin Oncol 23:6982-6991, 2005 19. Ludlum DB: DNA alkylation by the haloethylnitrosoureas: Nature of modifications produced and their enzymatic repair or removal. Mutat Res 233:117-126, 1990[CrossRef][Medline] 20. Brell M, Tortosa A, Verger E, et al: Prognostic significance of O6-methylguanine-DNA methyltransferase determined by promoter hypermethylation and immunohistochemical expression in anaplastic gliomas. Clin Cancer Res 11:5167-5174, 2005 21. Paz MF, Yaya-Tur R, Rojas-Marcos I, et al: CpG island hypermethylation of the DNA repair enzyme methyltransferase predicts response to temozolomide in primary gliomas. Clin Cancer Res 10:4933-4938, 2004 22. Mollemann M, Wolter M, Felsberg J, et al: Frequent promoter hypermethylation and low expression of the MGMT gene in oligodendroglial tumors. Int J Cancer 113:379-385, 2005[CrossRef][Medline] 23. Alonso ME, Bello MJ, Gonzalez-Gomez P, et al: Aberrant promoter methylation of multiple genes in oligodendrogliomas and ependymomas. Cancer Genet Cytogenet 144:134-142, 2003[CrossRef][Medline] 24. Dong SM, Pang JC, Poon WS, et al: Concurrent hypermethylation of multiple genes is associated with grade of oligodendroglial tumors. J Neuropathol Exp Neurol 60:808-816, 2001[Medline] 25. Balana C, Ramirez JL, Taron M, et al: O6-methyl-guanine-DNA methyltransferase methylation in serum and tumor DNA predicts response to 1,3-bis(2-chloroethyl)-1-nitrosourea but not to temozolamide plus cisplatin in glioblastoma multiforme. Clin Cancer Res 9:1461-1468, 2003 26. Spiro TP, Liu L, Majka S, et al: Temozolomide: The effect of once- and twice-a-day dosing on tumor tissue levels of the DNA repair protein O(6)-alkylguanine-DNA-alkyltransferase. Clin Cancer Res 7:2309-2317, 2001 Submitted February 28, 2006; accepted June 14, 2006.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 x normal and bilirubin 
= .05; power = .95. For the first phase, the enrollment of 49 patients was planned; if fewer than 20 patients had responded, the study would have been terminated. Otherwise, 17 more patients were to be enrolled. If at least 30 of a total of 66 patients responded, the treatment would have been judged effective. Moreover, the size of the series guaranteed that a hypothetical toxic effect of 5% probability would have been identified at least in one patient, with a 97% probability. 
2 test and the Mann-Whitney U test were used for univariate analysis to identify the variables significantly related to response, while logistic regression was used for multivariate analysis. 
$100,000 (N/R) Not Required
