Context.-Aggressive treatment of medulloblastoma, the most common pediatric brain tumor, has not improved survival. Identifying better prognostic indicators may warrant less morbid therapy.Objective.-To investigate the role of sex on outcome of medulloblastoma. Design.-Retrospective study of significant factors for survival with a median follow-up of 82 months. Setting.-University medical center. Patients.-A total of 109 consecutive, pediatric patients treated for primary medulloblastoma from 1970 to 1995 with surgery and postoperative radiotherapy and, after 1979, chemotherapy. Main Outcome Measures.-Factors independently associated with survival. Results.-The final multivariate model predicting improved survival included sex (hazard ratio, 0.52; 95% confidence interval [CI], 0.29-0.92; P=.03; favoring female), metastases at presentation (hazard ratio, 2.01; 95% CI, 1.14-3.52; P=.02), and extent of surgical resection (hazard ratio, 0.60; 95% CI, 0.34-1.04; P=.07; favoring greater resection). The overall, 5-year freedom from progression was 40% and survival was 49%. Radiotherapy dose (P=.72), and chemotherapy (P=.90) did not significantly affect a disease outcome.Conclusions.-The sex of the child was an important predictor for survival of medulloblastoma; girls had a much better outcome. The difference in survival between sexes should be evaluated in prospective, clinical trials.

Although nonoptimized KVAT dose distributions presented here were of inferior quality to VMAT plans, this work has demonstrated the feasibility of delivering low-energy kilovoltage x-rays to lesions up to 4 cm in diameter to depths of 8.1 cm while sparing surrounding tissue.

At 4 years the freedom from PSA failure following radiotherapy for high grade prostate cancer was comparable to previously reported surgical series. The high failure rate among patients with PSA greater than 20 ng./ml. suggests that these patients should be considered for investigational approaches. The apparent improvement in freedom from progression with the use of higher doses provides reason for optimism.

The circular genome of Staphylococcus aureus was cut into two fragments by a simple enzymatic method that cleaves a 10-base-pair site. The recognition sequence, A-T-C-G-mA J T-C-G-mA-T, was created by the combined use of the methylase MoCla I (A-T-C-G-mA-T) and the restriction endonucleaseDpn I (G-mA 4 T-C). This technique is insensitive to CpG methylation and in human DNA is predicted to produce fragments that, on average, are greater than five million base pairs. The ability to create such long pieces of DNA should facilitate mapping of large, complex chromosomes.Cutting chromosomes into a few, long fragments would markedly simplify the task of organizing and mapping complex genomic DNA. To reduce significantly the number of pieces that result from the cleavage of large genomes, it will be necessary to cut DNA into fragments that are greater than one million base pairs (1 Mb) and perhaps as large as 10 Mb. Recent work using pulsed-field electrophoresis (PFE) has resulted in the separation of intact, chromosomal DNA as large as 12 Mb (1, 2). However, present methods of cleaving DNA do not take advantage of this separation technology because they cut too frequently. It is generally true that the longer the recognition sequence of a DNA cleavage system, the less frequently it will cut and the larger the resulting fragments. To generate consistently pieces of DNA that are longer than 1 Mb requires a cutting technique with a recognition specificity of 10 base pairs (bp). The longest recognition sequence found in the case of type II restriction endonucleases is 8 bp (3), and on most genomes, the average size of fragments produced by these rare-cutting restriction enzymes is <1 Mb (4). A number of other enzymatic and chemical techniques that cut DNA less frequently than restriction endonucleases are still in the early stages of development (e.g., refs. 5 and 6).We previously described a method that uses adenine methylases to generate novel 8-and 10-bp recognition sequences for the methyladenine-dependent endonuclease Dpn I (G-mA I T-C; mA, N6-methyladenine) (7). It was possible with this technique to cleave a 10-bp recognition sequence on a specially constructed plasmid by using the adenine methylase M-Cla I (A-T-C-G-mA-T) with Dpn I. However, the technique had not been successfully applied to genomic DNA. We now demonstrate the improvement of the M-Cla I/Dpn I system and show that it can produce specific fragments of 1 Mb and greater in size by using the genome of the bacteria Staphylococcus aureus as a model substrate.MATERIALS AND METHODS DNA Preparation. Genomic DNA was prepared from S. aureus ISP8 that was provided by Peter Pattee (Iowa State University, Ames). The cells were grown to stationary phase, pelleted, and washed. Blocks or "inserts"of agarose containing S. aureus were made by first mixing SeaPlaque agarose (FMC BioProducts, Rockland, ME), final concentration of 1%, with 1010 cells per ml and allowing the mixture to solidify in 100-ul rectangular molds. Genomic DNA was prepared within the agaro...