- Email: [email protected]
Functional MRI and PET FDG studies: Correlations with glial tumor prognosis
Proceedings
of the 34th Annual ASTRO
Meeting
143
occurred in the 313 patients who received local irradiation only. There was no statistical difference (2/19 versus 9/38, Fisher’s exact 2-tailed t test, p=O.3). Of 4 patients with poorly differentiated have died of recurrent disease.
in these recurrence rates histology, 3 are known 10
Conclusion: We conclude that for well or moderately differentiated ependymomas of the spinal cord, local irradiation only is indicated, and that treatment should be individualized for the rarer cases of poorlydifferentiated ependymomas of the spinalcord.
33 INTERSTITIAL I-125 TEMPORARY IMPLANTS D. Yakar,
M.D.1,
Department
IRRADIATION (TI)
L. Zamorano,
of Radiation
FOR MALIGNANT
M.D.2,
Oncology1
BRAIN
M. Dujovny,
TUMORS:
M.D.2,
and Neurosurgery2,
ADVANTAGE OF PERMANENT IMPLANT
M. Sheehan,
R.T.T.l,
Henry Ford Hospital,
J.H.
Detroit,
(PI)
OVER
Kim, M.D.1 Michigan.
Purpose: Interstitial irradiation using high activity I-125 TI gives a moderate increase in survival in patients with brain tumors (reported by us in ASTRO 1989). However, this was accompanied by poor quality of life due to local necrosis and brain edema. In order to decrease radiation injury we started in 1989 to use low activity I-125 PI. Materials & Methods: From 7/86 to 6/89, 63 patients were treated with TI and from 7/89 to 5/91, 57 The implants were performed with stereotatic technique using 3D treatment patients with low activity PI. planning. The I-125 seeds were inserted into the tumor in l-4 catheters both in TI and PI patients. The dose rate for TI was 40 cGy per hour with a total dose of 6000 cGy. The initial dose rate for PI was 5 cGy per hour with a total dose of 10,000 cGy in 1 year. The dose was calculated at the periphery of the tumor. All patients with primary brain tumors also received local external beam radiation therapy of 5000 cGy in 5 weeks except 9 patients with AA who received PI alone.
Results: m
Patient data are
characteristics and shown in the enclosed
table. Comparison of Karnofsky and neurological status of patients before implant and after 6 months was performed. Karnofsky status remained excellent or improved in 46 out of 54 (85%) patients with PI compared with 28 of 48 patients (58%) with TI. This was significant at p= 0.001. Neurological status was better in 47 of 54 patients (87%) with PI compared to 33 of 49 (67%) with TI (p= 0.016). Conclusion: Patients treated with permanent implant showed improved survival compared to patients treated with temAmong patients who surporay implant. vived 6 months or longer, those with PI had a better quality of life as reflected by their Karnofsky and neurological status
Mean N
Group
Age
Mean Tumor Vol.
TI+ all Primary GBM ** AA ***
:: 7
57 53 42
29 31 37
TI Rec.
22 13
43 46
55 64
TI Met.
1:
42 52
48 11
PI++ all Primary GMB AA
41 5 36
39 50 38
15 30 13
all GBM AA
I
Mean Initial Karn.
I
Median survival weeks *
I
Probability 1.5 years survival *
PI 19 13 42 all 15 Recur- GMB 4 52 .. HA rent 1” , _.__ : 66 40 27 6 , “.I , PI Met. ** Glioblastoma multiforme * Kaplan Meyer survival estimates *** Anaplastic Astrocytoma + lemporary implant ++ Permanent implar
34 LACK OF COMPLICATIONS ADJUVANT TO RADIATION
IN LONG-TERM SURVIVORS AFTER TREATMENT WITH FLUOSOL AND OXYGEN AS AN THERkPY FOR HIGH-GRADE MALIGNANT BRAIN TUMORS.
Richard G. Evans, Bruce F. Kimler, Robert A. Morantz, and Solomon Batnitzky. Departments of Radiation Oncology and Diagnostic Radiology, University of Kansas Medical Center, Kansas City, KS 66160-7321; and Brain Tumor Institute, Kansas City, MO. Purpose: Starting in 1987, a Phase I/II trial was conducted to help determine the toxicity/efficacy of the perfluorochemical emulsion Fluosol-DA 20% and 100% oxygen, as an adjuvant to conventional radiation therapy (RT) for high-grade maljgnant brain tumors. Perfluorocarbon emulsions are able to carry large amounts of oxygen and release it where tissue PO is low. In addition, the small particle size allows passage through partially collapsed'capillaries.
144
Materials tration
Radiation Oncology, Biology, Physics
Volume 24, Supplement
1
and Methods: 3 grade III and 15 grade IV patients received one Fluosol adminis(8 ml/kg) per week, with daily oxygen breathing prior to and during RT. Megavoltage was delivered to the whole brain at 25 x 1.8 Gy, followed by 10 x 2 Gy to a boost
radiation volume (presurgical CT contrast-enhancing thus 65 Gy in 7 weeks. No serious early finishing the 7-week course on schedule. Results: Of the 18 patients, 10 (9 grade surgery, 6 (all grade IV) patients lived
plus a 3-cm complications
margin). The total tumor dose were noted, with every patient
1 grade III) survived IV, more 2 years, 4 patients
more
than
was
1 year
post-
lived more than 3 years, and 3 patients are alive at times ranging from 201 to 227 weeks. A historical control group of 37 matched (age and tumor grade) patients treated with comparable radiation doses of 6070 Gy. The median survival time of the Fluosol group was 75 weeks, which is not statistically significantly different from the median survival time of 54 weeks for the control cohort. However, a Gehan-Wilcoxon test applied to those patients that survive greater than 1 year (the same proportion in both groups) reveals a statistically significant difference (P=O.Oll) in favor of the Fluosol group. Periodic clinical evaluations showed no evidence of any functional or structural brain abnormalities that could be attributed to RT and/or Fluosol. Radiographic studies (CT and MRI), especially those obtained for 6 patients more than 2 years post-surgery, were examined for evidence of any changes in brain tissue outside the area of original tumor involvement. For the 10 patients that survived past one year, none of the studies showed any indication of structural alterations other that those easily assignable to tumor effects or resulting from surgery. Conclusions: These results show that although Fluosol/oxygen added to standard radiation therapy does not enhance survival of patients that succumb to their disease early, it does confer an added benefit to patients that survive past one year. The minimal acute side effects and no long-term deleterious effects suggest that Fluosol/oxygen sensitizes only hypoxic tumor cells, with no effect on well-oxygenated normal tissues. Overall, we have been impressed by the quality of life of patients following radiation therapy with adjuvant Fluosol plus oxygen and are encouraged to pursue this therapeutic approach further.
35 PHASE I-II STUDY OF IODODEOXYURIDINE (IUdR) AND RADIOTHERAPY IN THE TREATMENT OF MALIGNANT GLIOMA (RTOG 86-12) D. Fultoul, D. Cosmatos2, S. Lester3, M. Sliverstein4, T. Kinsella5 and T. Wasserman6 R.C. Urtasunl, tUniv of Alberta, Cross Cancer Institute; ZRadiation Therapy Oncology Group, Statistical Unit; 30chsner CCOP, 4Montefiore Medical Center; XJniv of Wisconsin, 6Washington Univ Purpose: In August 1987 the RTOG initiated a prospective step-wise study in patients with newly diagnosed anaplastic asmxytoma (AAF) and glioblastoma. In this study we wanted to evaluate the toxicity and tumor DNA incorporation of the halopvrimidine IUdR (NSC #39661, IND 22475) as a chemical modifier of radiation response. The studv was designed to have thedoseschedules ofIUdR,keeping the total a fixed dose of radiation (60.16 Gy = 1.88 Gy in 32 fx in 6.5 weeks) but varying
dose constant between 24 and 21 g/m2. We wanted to asses the efficacy and level of DNA incorporation of a high dose short time infusion vs. the acceptable four day infusion. Materials and Methods: IUdR was delivered in a 96 hour, 48 hour, and 24 hour continuous intravenous infusion per week for the 6-week course of radiation treatment. The 96 hour infusion was started 4 days prior to the start of radiotherapy. On the 48 and 24 hour infusions, IUdR started on the same day that radiation therapy commenced. All eligible patients were required to have normal renal and liver functions and bone marrow reserve, a Kamofskv of equal or more than 70 and be less than 75 vears old. The study was closed for patient accrual on October 1, 1991. 22 patients were entered to the 96 hour, 32 to the 48 hour and 25 to the 24 hour schedules. The incidence of glioblastoma ranged between 68 and 75% in the three arms. 70% of the patients had a Kamofsky of 80-90% at the onset of treatment. Over 50% of the patient population was under age 55. Results: Drug tolerance appeared to be related to the length of schedule of administration of the drug. More patients on the 24 and 48 hour schedule received at least 80% of the IUdR dose specified per protocol. The main drug related toxicities were hematological and changes in liver enzymes. There was a marked increase in liver enzymes @GOT, alkaline phosphatase and LDH), by four months after the end of treatment these values returned to normal levels with no sequela. No life threatening or Grade IV toxicity was observed. Toxicities were most pronounced in the 96 hour IUdR infusion schedule where 27.4% of the patients reported a Grade III toxicity. We did not observe a trend in acute normal tissue radiation reactions in any of the three arms. Although not enough time has elapsed to assess the survival experience in the 48 and 24 hour schedules, the median survival calculated from the Kaplan-Meier plot are 13.4, 10.9 and 10.4 months respectively for the 96,48 and 24 hour infusions. The Cox Proportional Hazards model showed that any difference in survival can be described by histological grade, type of previous surgery and to some extent age group of the patient. Dose schedule was not a significant predictor of survival. Pretreatment incorporation of IUdR by tumor cells was studied in 3 patients with glioblastoma. IUdR was administered prior to surgery at a dose of 2.5 g/m2 in a 24 hour continuous infusion in two patients and in 48 hour infusion in one patient prior to surgery. The percentage of cells labelled was 0% in all three patients. Conclusions: So far the failure to show an overall improvement in survival could be due to the low IUdR cell labelling seen in the biopsy material and possibly low incorporation into DNA as has been reported by others. In addition, the administration of IUdR is time consuming, laborious and has shown significant toxicity.
of the 34th Annual ASTRO
Meeting
143
occurred in the 313 patients who received local irradiation only. There was no statistical difference (2/19 versus 9/38, Fisher’s exact 2-tailed t test, p=O.3). Of 4 patients with poorly differentiated have died of recurrent disease.
in these recurrence rates histology, 3 are known 10
Conclusion: We conclude that for well or moderately differentiated ependymomas of the spinal cord, local irradiation only is indicated, and that treatment should be individualized for the rarer cases of poorlydifferentiated ependymomas of the spinalcord.
33 INTERSTITIAL I-125 TEMPORARY IMPLANTS D. Yakar,
M.D.1,
Department
IRRADIATION (TI)
L. Zamorano,
of Radiation
FOR MALIGNANT
M.D.2,
Oncology1
BRAIN
M. Dujovny,
TUMORS:
M.D.2,
and Neurosurgery2,
ADVANTAGE OF PERMANENT IMPLANT
M. Sheehan,
R.T.T.l,
Henry Ford Hospital,
J.H.
Detroit,
(PI)
OVER
Kim, M.D.1 Michigan.
Purpose: Interstitial irradiation using high activity I-125 TI gives a moderate increase in survival in patients with brain tumors (reported by us in ASTRO 1989). However, this was accompanied by poor quality of life due to local necrosis and brain edema. In order to decrease radiation injury we started in 1989 to use low activity I-125 PI. Materials & Methods: From 7/86 to 6/89, 63 patients were treated with TI and from 7/89 to 5/91, 57 The implants were performed with stereotatic technique using 3D treatment patients with low activity PI. planning. The I-125 seeds were inserted into the tumor in l-4 catheters both in TI and PI patients. The dose rate for TI was 40 cGy per hour with a total dose of 6000 cGy. The initial dose rate for PI was 5 cGy per hour with a total dose of 10,000 cGy in 1 year. The dose was calculated at the periphery of the tumor. All patients with primary brain tumors also received local external beam radiation therapy of 5000 cGy in 5 weeks except 9 patients with AA who received PI alone.
Results: m
Patient data are
characteristics and shown in the enclosed
table. Comparison of Karnofsky and neurological status of patients before implant and after 6 months was performed. Karnofsky status remained excellent or improved in 46 out of 54 (85%) patients with PI compared with 28 of 48 patients (58%) with TI. This was significant at p= 0.001. Neurological status was better in 47 of 54 patients (87%) with PI compared to 33 of 49 (67%) with TI (p= 0.016). Conclusion: Patients treated with permanent implant showed improved survival compared to patients treated with temAmong patients who surporay implant. vived 6 months or longer, those with PI had a better quality of life as reflected by their Karnofsky and neurological status
Mean N
Group
Age
Mean Tumor Vol.
TI+ all Primary GBM ** AA ***
:: 7
57 53 42
29 31 37
TI Rec.
22 13
43 46
55 64
TI Met.
1:
42 52
48 11
PI++ all Primary GMB AA
41 5 36
39 50 38
15 30 13
all GBM AA
I
Mean Initial Karn.
I
Median survival weeks *
I
Probability 1.5 years survival *
PI 19 13 42 all 15 Recur- GMB 4 52 .. HA rent 1” , _.__ : 66 40 27 6 , “.I , PI Met. ** Glioblastoma multiforme * Kaplan Meyer survival estimates *** Anaplastic Astrocytoma + lemporary implant ++ Permanent implar
34 LACK OF COMPLICATIONS ADJUVANT TO RADIATION
IN LONG-TERM SURVIVORS AFTER TREATMENT WITH FLUOSOL AND OXYGEN AS AN THERkPY FOR HIGH-GRADE MALIGNANT BRAIN TUMORS.
Richard G. Evans, Bruce F. Kimler, Robert A. Morantz, and Solomon Batnitzky. Departments of Radiation Oncology and Diagnostic Radiology, University of Kansas Medical Center, Kansas City, KS 66160-7321; and Brain Tumor Institute, Kansas City, MO. Purpose: Starting in 1987, a Phase I/II trial was conducted to help determine the toxicity/efficacy of the perfluorochemical emulsion Fluosol-DA 20% and 100% oxygen, as an adjuvant to conventional radiation therapy (RT) for high-grade maljgnant brain tumors. Perfluorocarbon emulsions are able to carry large amounts of oxygen and release it where tissue PO is low. In addition, the small particle size allows passage through partially collapsed'capillaries.
144
Materials tration
Radiation Oncology, Biology, Physics
Volume 24, Supplement
1
and Methods: 3 grade III and 15 grade IV patients received one Fluosol adminis(8 ml/kg) per week, with daily oxygen breathing prior to and during RT. Megavoltage was delivered to the whole brain at 25 x 1.8 Gy, followed by 10 x 2 Gy to a boost
radiation volume (presurgical CT contrast-enhancing thus 65 Gy in 7 weeks. No serious early finishing the 7-week course on schedule. Results: Of the 18 patients, 10 (9 grade surgery, 6 (all grade IV) patients lived
plus a 3-cm complications
margin). The total tumor dose were noted, with every patient
1 grade III) survived IV, more 2 years, 4 patients
more
than
was
1 year
post-
lived more than 3 years, and 3 patients are alive at times ranging from 201 to 227 weeks. A historical control group of 37 matched (age and tumor grade) patients treated with comparable radiation doses of 6070 Gy. The median survival time of the Fluosol group was 75 weeks, which is not statistically significantly different from the median survival time of 54 weeks for the control cohort. However, a Gehan-Wilcoxon test applied to those patients that survive greater than 1 year (the same proportion in both groups) reveals a statistically significant difference (P=O.Oll) in favor of the Fluosol group. Periodic clinical evaluations showed no evidence of any functional or structural brain abnormalities that could be attributed to RT and/or Fluosol. Radiographic studies (CT and MRI), especially those obtained for 6 patients more than 2 years post-surgery, were examined for evidence of any changes in brain tissue outside the area of original tumor involvement. For the 10 patients that survived past one year, none of the studies showed any indication of structural alterations other that those easily assignable to tumor effects or resulting from surgery. Conclusions: These results show that although Fluosol/oxygen added to standard radiation therapy does not enhance survival of patients that succumb to their disease early, it does confer an added benefit to patients that survive past one year. The minimal acute side effects and no long-term deleterious effects suggest that Fluosol/oxygen sensitizes only hypoxic tumor cells, with no effect on well-oxygenated normal tissues. Overall, we have been impressed by the quality of life of patients following radiation therapy with adjuvant Fluosol plus oxygen and are encouraged to pursue this therapeutic approach further.
35 PHASE I-II STUDY OF IODODEOXYURIDINE (IUdR) AND RADIOTHERAPY IN THE TREATMENT OF MALIGNANT GLIOMA (RTOG 86-12) D. Fultoul, D. Cosmatos2, S. Lester3, M. Sliverstein4, T. Kinsella5 and T. Wasserman6 R.C. Urtasunl, tUniv of Alberta, Cross Cancer Institute; ZRadiation Therapy Oncology Group, Statistical Unit; 30chsner CCOP, 4Montefiore Medical Center; XJniv of Wisconsin, 6Washington Univ Purpose: In August 1987 the RTOG initiated a prospective step-wise study in patients with newly diagnosed anaplastic asmxytoma (AAF) and glioblastoma. In this study we wanted to evaluate the toxicity and tumor DNA incorporation of the halopvrimidine IUdR (NSC #39661, IND 22475) as a chemical modifier of radiation response. The studv was designed to have thedoseschedules ofIUdR,keeping the total a fixed dose of radiation (60.16 Gy = 1.88 Gy in 32 fx in 6.5 weeks) but varying
dose constant between 24 and 21 g/m2. We wanted to asses the efficacy and level of DNA incorporation of a high dose short time infusion vs. the acceptable four day infusion. Materials and Methods: IUdR was delivered in a 96 hour, 48 hour, and 24 hour continuous intravenous infusion per week for the 6-week course of radiation treatment. The 96 hour infusion was started 4 days prior to the start of radiotherapy. On the 48 and 24 hour infusions, IUdR started on the same day that radiation therapy commenced. All eligible patients were required to have normal renal and liver functions and bone marrow reserve, a Kamofskv of equal or more than 70 and be less than 75 vears old. The study was closed for patient accrual on October 1, 1991. 22 patients were entered to the 96 hour, 32 to the 48 hour and 25 to the 24 hour schedules. The incidence of glioblastoma ranged between 68 and 75% in the three arms. 70% of the patients had a Kamofsky of 80-90% at the onset of treatment. Over 50% of the patient population was under age 55. Results: Drug tolerance appeared to be related to the length of schedule of administration of the drug. More patients on the 24 and 48 hour schedule received at least 80% of the IUdR dose specified per protocol. The main drug related toxicities were hematological and changes in liver enzymes. There was a marked increase in liver enzymes @GOT, alkaline phosphatase and LDH), by four months after the end of treatment these values returned to normal levels with no sequela. No life threatening or Grade IV toxicity was observed. Toxicities were most pronounced in the 96 hour IUdR infusion schedule where 27.4% of the patients reported a Grade III toxicity. We did not observe a trend in acute normal tissue radiation reactions in any of the three arms. Although not enough time has elapsed to assess the survival experience in the 48 and 24 hour schedules, the median survival calculated from the Kaplan-Meier plot are 13.4, 10.9 and 10.4 months respectively for the 96,48 and 24 hour infusions. The Cox Proportional Hazards model showed that any difference in survival can be described by histological grade, type of previous surgery and to some extent age group of the patient. Dose schedule was not a significant predictor of survival. Pretreatment incorporation of IUdR by tumor cells was studied in 3 patients with glioblastoma. IUdR was administered prior to surgery at a dose of 2.5 g/m2 in a 24 hour continuous infusion in two patients and in 48 hour infusion in one patient prior to surgery. The percentage of cells labelled was 0% in all three patients. Conclusions: So far the failure to show an overall improvement in survival could be due to the low IUdR cell labelling seen in the biopsy material and possibly low incorporation into DNA as has been reported by others. In addition, the administration of IUdR is time consuming, laborious and has shown significant toxicity.