✓ This study includes 182 patients with intracranial gliomas who received bromodeoxyuridine (BUdR), 200 mg/sq m intravenously, at the time of craniotomy but before tumor biopsy. The tumor specimens were stained for BUdR using the immunoperoxidase method; the BUdR labeling index (LI), or S-phase fraction, was calculated as the percentage of BUdR-positive cells. The median BUdR LI's for 127 primary moderately anaplastic astrocytomas, highly anaplastic astrocytomas, and glioblastomas (< 1%, 2.7%, and 7.3%, respectively; range 0% to 38.1%) were not significantly different from those of 55 similar recurrent tumors (< 1%, 4.3%, and 7.4%, respectively; range 0% to 30.5%). The mean LI was significantly higher in tumors from patients over 50 years of age than in tumors from younger patients (p < 0.001). This age-related difference in LI's was found in both groups of patients with astrocytomas but not in those with glioblastomas. Kaplan-Meier survival curves showed a significantly greater probability of survival among patients whose tumors had LI's of less than 1% than among those with LI's greater than 5%; survival probability of patients with tumor LI's of 1% to 5% was intermediate between the two extremes. Thus, the BUdR LI appears to reflect the proliferative potential more accurately than the histopathological diagnosis and should therefore be considered an important factor in determining the prognosis of individual patients with intracranial gliomas and in selecting their treatment.
Takao Hoshino, Michael Prados, Charles B. Wilson, Kyung Gi Cho, Kyu-Sung Lee and Richard L. Davis
Isabelle M. Germano, Masami Ito, Kyung G. Cho, Takao Hoshino, Richard L. Davis and Charles B. Wilson
✓ One hundred fifty-two intracranial gliomas of various types were reviewed in order to correlate the histopathological features with the proliferative potential of each tumor as reflected by the bromodeoxyuridine (BUdR) labeling index (LI). Patients undergoing surgical removal of gliomas were given a 30-minute intravenous infusion of BUdR (150 to 200 mg/sq m) to label S-phase tumor cells. The tumor specimens were stained immunohistochemically for BUdR and processed for routine histopathological diagnosis. The BUdR LI was calculated as the percentage of labeled cells among cells analyzed. Twenty-seven histological features in three categories (degenerative, vascular, and cellular changes) were considered. A significantly higher BUdR LI (p < 0.05) was found in tumors with necrosis than in those without this feature; tumors with both coagulative and liquefactive necrosis had the highest BUdR LI (p < 0.05). Increased vascularity was also associated with a higher BUdR LI (p < 0.05). Although tumors with abnormal mitotic figures had a significantly higher BUdR LI than those without, the number of mitoses did not correlate with a higher BUdR LI. These results suggest that the number of mitoses is not a good indicator of tumor growth rate. Necrosis and increased vascularity should be heavily weighted in predicting the proliferative potential of individual gliomas.
Takao Hoshino, Luis A. Rodriguez, Kyung G. Cho, Kyu S. Lee, Charles B. Wilson, Michael S. B. Edwards, Victor A. Levin and Richard L. Davis
✓ The proliferative potential of low-grade astrocytomas was estimated in 47 patients. Each patient received an intravenous infusion of bromodeoxyuridine (BUdR), 150 to 200 mg/sq m, at the time of craniotomy to label cells in deoxyribonucleic acid (DNA) synthesis; the percentage of S-phase cells, or BUdR labeling index (LI), of each tumor was determined immunohistochemically. In 29 patients (60%), the tumors had BUdR LI's of less than 1%, indicating a slow growth rate; only three (10%) of these patients died of recurrent tumor during a follow-up period of up to 3½ years. In contrast, of the 18 patients (40%) whose tumors had BUdR LI's of 1% or more, 12 (67%) had a recurrence and nine died during the same follow-up period. These results show that the proliferative potential, as reflected by the BUdR LI, is an important prognostic factor that separates low-grade astrocytomas into two groups and provides a more scientific rationale for selecting treatment for individual patients.
Judith Murovic, Krzysztof Turowski, Charles B. Wilson, Takao Hoshino and Victor Levin
✓ Ninety-seven patients with supratentorial malignant gliomas who received postoperative radiation therapy and chemotherapy at the University of California, San Francisco, from 1977 through 1984 showed improvement in their follow-up computerized tomography (CT) scans. Twenty-one of these 97 “CT responders” were designated “complete responders” because on serial CT scans they had complete disappearance of the tumor mass and contrast enhancement, which had been present postoperatively. In the remaining 76 patients, CT scans showed reduction in the size, but not disappearance, of the lesions, and these were designated “partial responders.” Fifty-eight partial responders had glioblastoma multiforme (GM); their median survival time was 72 weeks. The median survival time for the 11 complete responders with GM has not yet been achieved, but survival at the 53rd percentile is 172 weeks. Among patients with highly anaplastic astrocytoma, the median survival time was 211 weeks for the 10 complete responders and 125 weeks for the 18 partial responders. Eleven of the 21 complete responders are alive at a median postoperative follow-up time of 163 weeks (range 114 to 470 weeks). Eighteen of these patients had subtotal resection of tumor; three patients had gross total tumor resections, but postoperative CT scans showed evidence of residual or possibly recurrent tumor within 1.5 to 4.5 months. Resolution of the tumor mass and contrast enhancement took 9 to 151 weeks; the time to resolution did not depend upon the configuration of the remaining tumor mass and contrast enhancement after surgery. In this study, patients with malignant gliomas whose CT scans eventually showed sustained complete disappearance of the tumor mass and contrast enhancement had a more favorable prognosis than did patients whose CT scans showed improvement, but not complete disappearance, of the tumor. These CT findings may prove useful in determining the prognosis of patients with malignant gliomas.
Cell kinetics studies with bromodeoxyuridine labeling
Kyung G. Cho, Takao Hoshino, Tadashi Nagashima, Judith A. Murovic and Charles B. Wilson
✓ Eight patients with recurrent meningiomas (four malignant, two hemangiopericytic, and two nonmalignant) were given intravenous bromodeoxyuridine (BUdR), 200 mg/sq m, at the time of surgery to label cells in the deoxyribonucleic acid (DNA) synthesis phase; labeled cells were detected in excised tumor specimens by immunoperoxidase staining using anti-BUdR monoclonal antibody. These tumors showed a wide range of BUdR labeling indices (LI's), calculated as the percentage of BUdR-labeled cells divided by the total number of cells scored, from 0.3% to 5.4%. The tumor doubling times (Td's), estimated from serial computerized tomography scans, ranged from 8 to 440 days and showed a close inverse correlation with the BUdR LI's. A semilogarithmic linear regression analysis of these values yielded a correlation coefficient of 0.99. Tumor doubling time (Td) can be estimated using the formula: Td = 500 × Exp (−0.73 × LI), where Exp signifies the natural log base. By predicting the growth rate of meningiomas, the BUdR LI may supplement histopathological diagnosis and improve both the determination of prognosis and the design of treatment modalities in individual patients.
Tadashi Nagashima, Judith A. Murovic, Takao Hoshino, Charles B. Wilson and Stephen J. DeArmond
✓ At the start of transsphenoidal microsurgery for removal of various types of pituitary adenomas, 21 patients received a 1-hour intravenous infusion of 5-bromodeoxyuridine (BUdR, 200 mg/sq m) to label tumor cells in the deoxyribonucleic acid (DNA) synthesis phase (S-phase). Excised tumor specimens were fixed in 70% ethanol and stained by the indirect peroxidase method using anti-BUdR monoclonal antibody as the first antibody. The percentage of BUdR-labeled cells, or S-phase fraction, was calculated for each specimen. The S-phase fraction was less than 0.1% in nine cases, 0.1% to 0.5% in seven, and greater than 0.5% in five. Except in two cases of Nelson's syndrome, in which it was greater than 1%, the S-phase fraction did not correlate with any other variable, including patient age, tumor size, or the duration of signs and symptoms.
The small S-phase fraction of most of the pituitary adenomas correlates well with the clinical behavior of these tumors, which grow much more slowly than other kinds of brain tumors such as gliomas. However, the S-phase fractions varied by as much as one order of magnitude. The higher S-phase fractions may reflect aggressive and invasive growth. These results indicate that immunohistochemical studies of cell kinetics using BUdR and anti-BUdR monoclonal antibodies may provide information about the biological characteristics of pituitary adenomas which could lead to the design of appropriate treatment regimens (including surgery, radiation therapy, and chemotherapy) for individual patients.
Takao Hoshino, Tadashi Nagashima, Judith A. Murovic, Charles B. Wilson, Michael S. B. Edwards, Philip H. Gutin, Richard L. Davis and Stephen J. DeArmond
✓ Thirty-eight patients undergoing surgical removal of neuroectodermal tumors of the central nervous system were given a 1-hour intravenous infusion of bromodeoxyuridine (BUdR), 150 to 200 mg/sq m, to label tumor cells in the deoxyribonucleic acid (DNA) synthesis phase (S-phase). The excised tumor specimens were divided into two portions: one was fixed with 70% ethanol and embedded in paraffin and the other was digested with an enzyme cocktail to make a single-cell suspension. The paraffin-embedded tissues were stained by an indirect peroxidase method using anti-BUdR monoclonal antibody (MA) as the first antibody. Single-cell suspensions were reacted with fluorescein isothiocyanate (FITC)-conjugated anti-BUdR MA's for flow cytometric analysis. S-phase cells that had incorporated BUdR into their DNA were well stained by both methods. The percentage of BUdR-labeled cells, or S-phase fraction, was calculated in tissue sections by microscopic examination and in single-cell suspensions by flow cytometric analysis. The biological malignancy of the tumors was reflected in the S-phase fractions, which were 5% to 20% for glioblastoma multiforme, medulloblastoma, and highly anaplastic astrocytoma, but less than 1% in most moderately anaplastic astrocytomas, ependymomas, and mixed gliomas. Two juvenile pilocytic astrocytomas and two low-grade astrocytomas from children had high S-phase fractions despite the fairly benign and slow-growing nature of these tumors. These results indicate that the S-phase fraction obtained immunocytochemically with anti-BUdR MA's may provide useful information in estimating the biological malignancy of human central nervous system tumors in situ.
Esref Tel, Takao Hoshino, Marvin Barker and Charles B. Wilson
✓ In Fischer 344 rats, 9L tumors were implanted in the left cerebral hemisphere. Compared with control animals not operated on, rats treated with an LD10 dose of BCNU 1 hour before or 1 or 12 hours after surgery on Day 16 postimplant had an increased life span of over 200% (greater than a 6 log cell kill). Minimal effect on survival was found when BCNU was administered during surgery. On the other hand, BCNU administered 12 hours before or 24 or 72 hours after surgery did not show any additive effect of surgery on BCNU treatment. These results suggest that in a clinical setting, a bolus of BCNU administered to tumor patients within 12 hours of surgery might increase substantially the total tumor cell kill compared with surgical resection alone.
Flow cytometric studies
Takao Hoshino, Kazuhiro Nomura, Charles B. Wilson, Kathy D. Knebel and Joe W. Gray
✓ Flow cytometry (FCM) is a technique that measures the quantity of DNA contained in individual nuclei and records a frequency distribution of the DNA content per nucleus in the sampled cell population. Nuclei from a variety of human brain-tumor types were isolated by means of tissue grinding, purified by centrifugation through 40% sucrose (15 minutes at 4000 rpm), fixed with 10% formalin, stained with acriflavin-Feulgen, and analyzed by FCM. Profiles of DNA distribution in histologically benign tumors, such as meningiomas, pituitary adenomas, neuroblastomas, and low-grade astrocytomas, revealed a large diploid population (2C) with a few nuclei in DNA synthesis, as well as a small premitotic population (G2 cells) that contains a 4C DNA complement. In contrast, malignant gliomas, including glioblastomas, consist of more cells in DNA synthesis; these tumor cells show a highly variable distribution of ploidy consisting not only of diploid, and/or aneuploid, but also of triploid, tetraploid, and possibly octaploid populations. Also, a large variability between different regions of each tumor was always observed. In contrast, metastatic brain tumors, despite the fact that they contain a considerable number of cells undergoing DNA synthesis, demonstrate little variability within each individual tumor. The ability to rapidly characterize the cell populations of human brain tumors with FCM may enhance the effectiveness of their clinical management.
Takao Hoshino, Charles B. Wilson, Mark L. Rosenblum and Marvin Barker
✓Four patients received 3H-thymidine 4 to 7 days and vinblastine 4 to 6 hours prior to operation for recurrent malignant gliomas (three glioblastomas and one anaplastic astrocytoma). Tumor biopsies obtained at operation were fixed for routine histological studies and radioautography. The tumors' growth fractions averaged 0.28 with a range of 0.14 to 0.39. The tumor cell cycle time calculated in three patients had a mean duration of 57 hours with a standard deviation of 6 hours. The authors concluded that: 1) single short-term courses of cell-cycle specific chemotherapeutic agents alone will probably fail to achieve either significant reduction in tumor mass or dramatic clinical improvement; 2) cell-cycle phase-specific drugs should be administered to maintain effective blood levels over 2 to 3 days for maximal tumor cell kill. Tumor growth rate appears to correlate with the fraction of proliferating cells rather than the length of the tumor cell cycle. The scientific basis for combination drug and multimodality therapy is discussed.