OF TOTAL IRRADIATION. 0. RINGDEN, I. BARYD, B. JOHANSSON, G. GAHRTON, C. G. GROTH, G. LUNDGREN and B. LONNQVIST

Transcription

1 Acra Radiologica Oncology (983) Fasc. 6 FROM THE DEPARTMENT OF TRANSPLANTATION SURGERY, AND DIVISION OF CLINICAL HEMATOLOGY AND ONCOLOGY, DEPARTMENT OF MEDICINE, HUDDINGE SJUKHUS, S-486 HUDDINGE. AND DEPARTMENT OF HOSPITAL PHYSICS AND RADIUMHEMMET, KAROLINSKA SJUKHUSET, S-040 STOCKHOLM, SWEDEN. INCREASED MORTALITY BY SEPTICEMIA, INTERSTITIAL PNEUMONITIS AND PULMONARY FIBROSIS AMONG BONE MARROW TRANSPLANT RECIPIENTS RECEIVING AN INCREASED MEAN DOSE RATE OF TOTAL IRRADIATION 0. RINGDEN, I. BARYD, B. JOHANSSON, G. GAHRTON, C. G. GROTH, G. LUNDGREN and B. LONNQVIST Total body irradiation (TBI) is effective in destroying leukemic cells and has therefore been used in most protocols to prepare leukemic patients for bone marrow transplantation. However, radiation in high doses causes acute toxicity and late side effects. The highest tolerable dose and dose rate is limited by tissue damage especially to lung and gut. Recently, several articles have been published regarding a correlation between high dose rate of TBI and the appearance of interstitital pneumonitis. From the International Bone Marrow Registry it was reported that TBI given for the conditioning of bone marrow transplant recipients was associated with a higher risk of interstitial pneumonitis at a dose rate above Gy/min in comparison with lower dose rates (BORTIN et coll. 98). A relationship between the dose absorbed by the lung and interstitial pneumonitis has been reported by others (VAN DYK et coll. 98, KEANE et coll. 98). In a cooperative European survey of TBI for bone marrow transplantation it was found that the dose rate may affect the incidence of interstitial pneumonitis at total lung doses exceeding 8 Gy (BARRETT 98). A correlation between lung damage and the dose rate of TBI has also been shown in mice (DEPLEDGE & BARRETT 98). A markedly increased mortality in a group of bone marrow transplant recipients, who were accidentally given an increased dose rate of TBI, is now reported. Materials and Methods From February 980 until September bone marrow transplant recipients with acute leukemia were conditioned using either cyclophosphamide (60 mg/kg day 5 and 4, 3 patients) or nitrogen mustard ( mg/kg day 4, 3 patients). All of them received 0 Gy of TBI as a single fraction dose, referred to the abdomen midline of the patient and 9 Gy to the middle of the lungs. The TBI was delivered by a 4 MeV linear accelerator (Philips SL 75-5) with horizontal beam, 45" rotated collimator and a maximum field size of 60 cm x 60 cm at 400 cm from focus to the patient's midline. The treatment was given a.p.-p.a. with the patient lying with bended knees alternately on the right and left side. The pulse repetition frequency was chosen in such a way that the midpoint dose rate in a 0 cm thick patient is less than 0. Gy/min. The overall treatment time including interruptions was determined not to be shorter than 3 hours. During a part of the treatment, the lungs were shielded with lead blocks to receive no more than 9 Gy. Entrance and exit dose measurements were performed with TLD (thermoluminescent dosimeter) and a direct reading diode system. The treatment was subdivided in 6 sessions and after each session Accepted for publication 8 April

2 44 0. RINGDEN, I. BARYD, B. JOHANSSON ET COLL. the TLD were sent for reading. A typical treatment plan is shown in Table. Following transplantation, 3 of the patients were treated with methotrexate for prophylaxis against graft-versus-host disease (GVHD), using the Seattle protocol (STORB et coll. 970). Three patients were treated with cyclosporin A7.5 mg/kg/day intravenously given over 4 h during the first weeks when parenteral nutrition was needed. Thereafter, cyclosporin A was given.5 mglkglday orally in divided doses for 6 months, then being tapered. A detailed description of the treatment schedule has recently been published (RINGD~~N et coll. 98). In order to have the results from the first two TLD measurements before giving the last treatment fields, the lung shielding arrangements were performed during the middle of the TBI. Thus there was an interruption in the irradiation of the lungs lasting from h and 37 min to h and 3 min (mean I h 47 mink3 min, SE). The patients changed positions 5 times and had at least this number of interruptions during the irradiation. The total procedure took 4 h and resulted in an overall mean dose rate to the lungs and the pelvic midpoint of approximately 0.04 Gylmin. Ten patients were subjected to this TBI schedule and the dose calculations were found to be appropriate (group I). This schedule was later changed because a careful adjustment of the position of the lung shield blocks is much easier to perform at the beginning of the TBI, when the patient is unaffected by the radiation. Thus, the long mid-time break was eliminated and, in accordance with the new protocol, the patients changed position only 3 times. The total procedure took about 3 4 h (Table ). The overall mean dose rate to the lungs remained approximately 0.04 Gy/min. For the pelvic midpoint it was 0.05 Gy/min. Most of the time during the session (Table ) is occupied by lung shielding arrangements which include development of films that are used to check the position of the lung shields. The radiation dose is delivered only during the last 0 minutes of the session. This means that about 9 Gy of the irradiation to the pelvic midpoint and the lungs were delivered during /4 h at a mean dose rate of 0.07 Gy/min. Seven patients (group ) followed this schedule; it was discontinued when it became apparent that the patients showed more severe signs of irradiation toxicity than the previous patients. The following patients were therefore treated according to the original schedule. There are now an additional 7 patients Table Original freatment plan of radiation Treatment session Treatment field A.p. P.a. A.p. P.a. A.p. P.a. Lung shield x x Pelvic midline dose (Gy) Lung midlinedose (Gy) Elapsed time (min) Table Radiation plan leading to an increased mean dose rate Treatment session Treatment field A.p. P.a. P.a. A.p. A.p. P.a. Lung shield x x Pelvic midline dose (Gy) Lung midline dose (Gy) Elapsed time (min) who have received the initial protocol (group I). The characteristics of all patients are shown in Table 3. Since group I only consisted of patients with acute lymphoblastic leukemia (ALL), this group was also compared with 0 patients from group I who had ALL. Results The clinical outcome in the patients is given in Table 4. Five of the 7 patients (7%) in group I developed six early septicemias compared with 8/7 (30%) among the other patients which was not a significant difference. Among the ALL-patients in group I the incidence of early septicemia was also 30 per cent. In group I receiving the higher dose rate of TBI there was an increased death rate (pco.0) compared with group I (Figure, Table 4). The incidence of death in early septicemia, interstitial pneumonitis and pulmonary fibrosis, causes that may be due to irradiation toxicity, were statistically significantly increased in group compared with group I (p<o.ol) as well as the ALL-patients in group I (p<0.0). In group I 3 patients died of septicemia: a 9- year-old male with ALL transplanted in an unstable second remission died on the 5th day of serratia

3 INCREASED MORTALITY IN MARROW RECIPIENTS BY A HIGHER DOSE RATE OF TBI 45 Table 3 Patient characteristics, total body irradiation, age, diagnosis, remission status, ABO and HLA match Group I. All cases included (Febr. 980-Jan. 98, n=0) (May 98-Sept. 98, n=7) Group I. ALL-patients only (January 98, n=l) (Oct. 98-Sept. 98, n=9) Group I (Jan. 98 -April 98) Number of patients Total body irradiation Interruptions Dose rate' lung Dose rate' pelvic midpoint Age (years), median (range) Diagnosis: ANLL/ALL Remission First Second Third Partial3 Unstable4 ABO major mismatch HLA mismatch (43) (3-3) 00 ANLL=acute non-lymphoblastic leukemia. ALL=acute lymphoblastic leukemia. ' Overall mean dose rate (Gyimin), including interruptions. ' Last 9 Gy to lung and pelvic midpoint. marrow. 0-38% blasts in bone marrow. Table 4 7 ' Outcome and causes of death in the patient groups * 9 (6-4) 07 Group Group I Group All patients receiving ALL-patients only Higher dose rate lower dose rate receiving lower dose rate Observation time (months) Number of patients 7 0 Patients with early septicemia 8 (30%) 3 (30%) Death 9 (33%)* 5 (50%) Relapse 3 (70) (0%) GVHD 3 Septicemia (4%) Interstitial pneumonitis (7%) I]%** : lo%* Pulmonary fibrosis 0 0 Significant difference from group I *p<0.0, **p<o.oi (Fisher's exact test). L (7 %) 6 (86%) 0 (4%) 3 (46%) I 7% 7% and 8% blasts in bone marcescens septicemia in combination with gastrointestinal bleeding. A 4-year-old male with ALL in first remission died on the 4th day of septic chock caused by pseudomonas. A 3-year-old male with ALL in second remission developed septicemia and meningitis caused by a-streptococci and died from status epilepticus with subsequent brain damage on the 30th day. Two boys with ALL in second remission, 3 and 8 years of age, respectively, died of pulmonary complications. One of the boys died of progressive pulmonary fibrosis 8 months after transplantation confirmed at autopsy, the other patient died of interstitial pneumonitis that probably was caused by cytomegalovirus. Autopsy was refused by the relatives. A -year-old female with ALL in first remission died of severe GVHD and bronchopneumonia 3 months after transplantation. The only survivor in group I is a 6-year-old boy

4 46 O. RINGDEN, I. BARYD, B. JOHANSSON ET COLL with ALL in his second remission who received a graft from his, HLA-D one-antigen, mismatched sister. He is doing well 5 months after transplantation. In group I there was only one death induced by early septicemia and this occurred in a 40-year-old female with acute myeloid leukemia (AML) in first remission. This patient developed septicemia by a- streptococci and probably also legionairs disease and died of broncopneumonia and generalized candidiasis on the st day after transplantation. Two patients died of interstitial fungal pneumonitis, one of them had previously developed veno-occlusive disease of the liver. Three patients in group I died of severe GVHD and 3 patients relapsed 5, 0 and 3 months after transplantation, respectively. Discussion The increased death rate among the patients receiving the higher dose rate is striking. However, there are many factors that affect patient survival after bone marrow transplantation like patient age, HLA match between recipient and donor, remission status, original disease, etc. Regarding age and matching there is no difference between group I and group. Only /3 of the patients in group I were in their first remission compared with half of the patients in group. Patients who were in a later remission received more cytotoxic drugs before transplantation: this may have increased the incidence of complications. However, the higher patient survival rate seen among patients grafted in their first remission (THOMAS et coll. 979a) is mainly due to a lower incidence of relapse compared with patients transplanted in their second or subsequent remissions (THOMAS et coll. 979b). Since there were no relapses in group, the remission status of these patients most probably did not have any major effect on the death rate. By chance, all 7 patients receiving the higher dose rate had ALL. Among the ALL-patients receiving the lower dose rate the main causes of death were relapse and GVHD. Among the patients treated with the higher dose rate of TBI there were no relapses and 5 of 6 deaths were caused by pulmonary conditions or septicemia which may be related to radiation damage. Radiation damage to the gastrointestinal tract may pave the way for early septicemias as noted in dogs (DEEG et coll. 98). The higher toxicity seen in group I may be due to Patient survival Years Actuarial patient survival (in per cent) was significantly lower (p<o.oi, t-test) among 7 patients with acute leukemia and treated with bone marrow transplantation and receiving a mean dose rate of 0.07 Gyhin of TBI (---) than among 7 patients receiving 0.04 Gyhin (-). several factors. First, the mean dose rate was higher during the latter phase of the irradiation and secondly the TBI was not interrupted so frequently to change the position of the patient. According to the original protocol the lung shielding resulted in a long mid-time break of the irradiation. Not only low dose rate but also fractionated irradiation has been suggested to cause less tissue damage (PETERS et coll. 979) and experiments in mice, using the LDso for respiratory death, indicate that dose fractionation is more important for survival than dose rate (WARA et coll. 973, FIELD et coll. 976). Furthermore, bone marrow transplant recipients receiving fractionated TBI have also been reported to have a reduced incidence of interstitial pneumonitis, although the total dose given was greater (SHANK et coll. 98). In another investigation, among patients with acute non-lymphoblastic leukemia in first remission there was no difference in the frequency of interstitial pneumonitis between patients receiving Gy on each of 6 days compared with patients treated with 0 Gy in one setting (THOMAS et coll. 98). However, the patients who received the fractionated TBI had a better (p<0.05) survival. With regard to acute toxicity and early mortality, these were found to be dependent upon the total dose and dose rate in dogs (DEEG et coll.) In mice, FEOLA et coll. (974) showed that the LDSO30 for

5 INCREASED MORTALITY IN MARROW RECIPIENTS BY A HIGHER DOSE RATE OF TBI 47 TBI was reduced from 0 to 7.5 Gy when dose rates were increased from 0.06 to 0.35 Gy/min. There are wide differences in TBI techniques at various institutions (KIM et coll. 980). As long as dose specifications are not systematized it is difficult to compare the biologic effects of irradiation at different centers and the optimum TBI schedule which gives acceptable toxic side effects and lowest possible relapse frequency has to be determined at each center. In conclusion, the TBI schedule of dose rate and fractionation is most important for survival during the first risk-filled months after bone marrow transplantation. What initially looked like a minor change in the TBI schedule most probably had fatal consequences for one group of the patients. SUMMARY Seven bone marrow transplant recipients with acute lymphoblastic leukemia receiving a mean dose rate of 0.07 Gy/min of total body irradiation towards the pelvic midpoint and the lungs had an increased (p<0.0) overall death rate of 86 per cent compared with 33 per cent among 7 patients with acute non-lymphoblastic leukemia or acute lymphoblastic leukemia treated with a mean dose rate of 0.04 Gyimin. Among the patients receiving the higher dose rate there was an increased mortality in causes related to radiation toxicity like early septicemia, interstitial pneumonitis and pulmonary fibrosis, compared with all patients receiving the lower dose rate (p<o.0) and also with 0 patients from this group with acute lymphoblastic leukemia (p<0.0). ACKNOWLEDGEMENTS This investigation was supported by grants from the Swedish Medical Research Council, Jennyfonden, Karolinska Institutet and the Swedish Society of Medical Sciences. The authors wish to thank Mr Ulf Brodin for statistical advice. Reqtrest for reprints: Dr Olle Ringden, Department of Transplantation Surgery, Huddinge Sjukhus, S-4 86 Huddinge, Sweden. REFERENCES BARRETT A,: Total body irradiation (TBI) before bone marrow transplantation in leukaemia. A co-operative study from the European Group for Bone Marrow Transplantation. J. Radiol. 55 (98), 56. BORTIN M. M., KAY H. E. M., GALE R. P. and RIMM A. A.: Factors associated with interstitial pneumonitis after bone-marrow transplantation for acute leukaemia. Lancet I (98), 437. DEEG H. J., STORB R., WEIDEN P. L., SCHUMACHER D.. SHULMAN H., GRAHAM T. and THOMAS E. D.: Highdose total-body irradiation and autologous marrow reconstitution in dogs. Dose-rate-related acute toxicity and fractionation-dependent long-term survival. Radiat. Res. 88 (98), 385. DEPLEDCE M. H. and BARRETT A,: Dose-rate dependence of lung damage after total body irradiation in mice. Int. J. Radiat. Biol. 4 (l98), 35. FEOLA J. M., SONG C. W., KHAN F. M. and LEVITT S. H.: Lethal response of CS7BL mice to 0 MeV X-rays and to boco gamma-rays. Int. J. Radiat. Biol. 6 (974). 6. FIELD S. B., HORNSEY S. and KUTSUTANI Y.: Effects of fractionated irradiation on mouse lung and a phenomenon of slow repair. Brit. J. Radiol. 49 (976) KEANE T. J., VAN DYK J. and RIDER W. D.: Idiopathic interstitial pneumonia following bone marrow transplantation. The relationship with total body irradia-. tion. Int. J. Radiat. Oncol. Biol. Phys. 7 (98) KIMT. H., KHAN F. M. and GALVIN J. M.: A report ofthe work party. Comparison of total body irradiation techniques for bone marrow transplantation. Int. J. Radiat. Oncol. Biol. Phys. 6 (980), 779. PETERS L. J., WITHERS H. R., CUNDIFF J. H. and DICKE K. A,: Radiobiological considerations in the use of total-body irradiation for bone-marrow transplantation. Radiology 3 (979), 43. RINGDEN O., LONNQWST B., LUNDGREN G., GAHRTON G., GROTH C. G., M~LLER E., BARYD I., JOHANSSON B., PIHLSTEDT P. and GULLBRING B.: Experience with a cooperative bone marrow transplantation program in Stockholm. Transplantation 33 ( 98), 500. SHANK B., HOPFAN S., KIM J. H., CHU F. C. H., GROSS- BARD E., KAPOOR N., KIRKPATRICK D., DINSMORI: R.. SIMPSON L., REID A., CHUI C., MOHAN R., FINEGAN D. and O REILLY R.: Hyperfracionated total body irradiation for bone marrow transplantation. I. Early results in leukemia patients. Int. J. Radiat. Oncol. Bid. Phys. 7 (98), 09. STORB R., EPSTEIN R. B., GRAHAM T. C. and THOMAS E. D.: Methotrexate regimens for control of graft-versushost disease in dogs with allogeneic marrow grqs. Transplantation 9 (970), 40. THOMAS E. D., BUCKNER C. D., CLIFT R. A., FEFER A.. JOHNSON F. L., NEIMAN P. E., SALE G. E., SANDERS J. E., SINGER J. W., SHULMAN H., STORB R. and WEIDEN P. L. (a): Marrow transplantation for acute nonlymphoblastic leukemia in first remission. New Engl. J. Med. 30 (979), SANDERS J. E., FLOURNOY N., JOHNSON F. L., BUCKNER C. D., CLIFT R. A., FEFER A,, GOODELL B. W., ST~RB R. and WEIDEN P. L. (b): Marrow transplantation for patients with acute lymphoblastic leukemia in remission. Blood 54 (979) CLIFT R. A,, HERSMAN J.,SANDERS E., STEWART P.. BUCKNER C. D., FEFER A,, MCGUFFIN R.. SMITH J. W. and STORB R.: Marrow transplantation for- acute nonlymphoblastic leukemia in first remission using

6 48 0. RINGDEN, I. BARYD, B. JOHANSSON ET COLI,. fractionated or single dose irradiation. Int. J. Radiat. Oncol. Biol. Phys. 8 (98, 87. VAN DYK J., KEANE T. J., KAN S., RIDER W. D. and FRYER C. J. H.: Radiation pneurnonitis following large single dose irradiation. A re-evaluation based on abso- lute dose to lung. Int. J. Radiat. Oncol. Biol. Phys. 7 (98), 46. WARA W. M., PHILLIPS T. L., MARGOLIS L. and SMITH W.: Radiation pneurnonitis. A new approach to the derivation of time-dose factors. Cancer 3 (973), 547.