Inr. 1. Rodiarron Oncology Biol. Phys.. Vol 9, pp. 1029- 1033 036~3016/83/0710294510300/0

Printed in the U.S.A. All rights reserved. CopyrIght 0 1983 Pcrgamon Press Ltd.

0 Original Contribution

INTERSTITIAL PNEUMONITIS FOLLOWING BONE MARROW TRANSPLANTATION AFTER LOW DOSE RATE

TOTAL BODY IRRADIATION

ANN BARRETT, M.B., B.S., F.R.C.R., MICHAEL H. DEPLEDGE, B.SC.(HONS.),

PH.D., M.I.BIoL. AND RAYMOND L. POWLES, M.D., B.Sc., F.R.C.P. Bone Marrow Transplant Team, The Royal Marsden Hospital, Clifton Avenue,

Sutton, Surrey, SM2 5PX England

Idiopathic and infective interstitial pneumonitis (IPn) is a common complication after bone marrow transplantation (BMT) in many centers and carries a high mortality. We report here a series of 107 patients with acute leukemia grafted at the Royal Marsden Hospital in which only 11 (10.3 % ) developed IPn and only 5 died (5 % ). Only one case of idiopathic IPn was seen. Factors which may account for this low incidence are discussed. Sixty of 107 patients were transplanted in first remission of acute myeloid leukemia (AML) and were therefore in good general condition. Lung radiation doses were carefully monitored and doses of 10.5 Cy were not exceeded except in a group of 16 patients in whom a study of escalating doses of TBI (up to 13 Cy) was undertaken. The dose rate used for total body irradiation (TBI) was lower than that used in other centers and as demonstrated elsewhere by ourselves and others, reduction of dose rate to to.05 Gy/min may be expected to lead to substantial reduction in lung damage. Threshold doses of approximately 8 Gy for IPn have been reported, but within the dose range of 8 to 10.5 Gy we suggest that dose rate may significantly affect the incidence. Data so far available suggest a true improvement in therapeutic ratio for low dose rate single fraction TBI compared with high dose rate.

Total body irradiation, Interstitial pneumonitis, Low dose rate.

INTRODUCTION Idiopathic and infective IPn is a common complication after BMT.” Patients present with a clinical syndrome which may include dyspnea, pyrexia, cough and cyanosis. Chest X rays often show bilateral diffuse shadowing and lung function tests reveal a restrictive ventilatory defect with a low diffusing capacity (D,CO).’ Histological examination of lung tissue shows interstitial edema and fibrosis with cellular infiltration and alveolar exudates.

The etiology of IPn is multifactorial and therefore differences in transplantation procedures between the various BMT centers may influence the incidence and time of onset of the condition. We report here the experience of the Royal Marsden Hospital Bone Marrow Transplant Team.

METHODS AND MATERIALS Between 1974 and 1981, 107 patients with acute

leukemia have received bone marrow transplants at the Royal Marsden Hospital. Patients with acute myeloid leukaemia (AML) were usually transplanted in first remission; patients with acute lymphatic leukemia (ALL)

were transplanted in second remission. Since 1980, 23 patients have received marrow from a family donor mismatched at the D locus alone or additionally at the A,

B or C locus, but all other patients have received grafts from HLA identical siblings with mutually non-reactive mixed leucocyte cultures (MLC).

Cyclophosphamide 60 mg Kg ’ per day was given on days -3 and -2 before total body irradiation (TBI).

Full details of the irradiation technique have been described elsewhere.’ Patients were irradiated in a plexi- glas* cot at 4 meters from a cobalt” source at an approximate dose rate of 0.025 Gy min ’ measured at the front of the cot. The dose distribution was calculated from lithium fluoride dose meters placed at various sites on the skin on opposite sides of the body to measure entrance and exit doses. The midplane dose was taken as an average of these two readings. The dose quoted for each patient was the maximum lung dose, which differed from those at other points in the body by up to -r 5%. Total treatment times were between 370 and 420 minutes. Turning and nursing time was minimized and averaged about I5 minutes.

*Trade mark Perspex. Members of the Bone Marrow Transplant Team: H. M.

Clink, P. Dady, B. Jameson, H. E. M. Kay, S. Lawler, D. Lawson, T. J. McElwain, G. Morgenstern, J. G. Watson.

Reprint requests to: Dr. Ann Barrett. Accepted for publication 15 February 1983.

1029

Tabl

e 1.

In

ters

titia

l pn

eum

oniti

s fo

llow

ing

bone

m

arro

w

trans

plan

tatio

n

Don

or

Rem

issi

on

Type

of

Im

mun

o-

GvH

D

Patie

nt

UPN

* A

ge

Sex

sex

Dis

ease

st

atus

gr

aft

supp

ress

ion

biop

sy

1 8

10

M

M

ALL

2 54

30

M

M

C

GL

3 38

19

M

M

A

LL

4 94

31

F

M

AM

L

5 42

45

M

F

ALL

6 35

28

F

M

AM

L

7 43

14

F

F A

ML

8 11

4 35

F

F A

ML

9 69

18

F

F A

ML

10

112

35

M

F A

ML

11

65

12

M

M

AM

L

Seco

nd

rem

issi

on

Chr

onic

ph

ase

Seco

nd

rela

pse

Seco

nd

rela

pse

Firs

t rem

issi

on

Firs

t rem

issi

on

Seco

nd

rela

pse

Seco

nd

rem

issi

on

Firs

t rem

issi

on

Firs

t rem

issi

on

Firs

t rem

issi

on

Mat

ched

Mat

ched

Mat

ched

Mis

mat

ched

Mat

ched

Mat

ched

Mat

ched

Mis

mat

ched

Mis

mat

ched

Mat

ched

Mis

mat

ched

Met

hotre

xate

Cyc

losp

orin

A

+

Cyc

losp

orin

A

+

Cyc

losp

orin

A

+

Cyc

losp

orin

A

Cyc

losp

orin

A

Cyc

losp

orin

A

Met

hotre

xate

+

+ C

yclo

spor

in

A

Cyc

losp

orin

A

+

Cyc

losp

orin

A

+

Cyc

losp

orin

A

+

Irra

diat

ion

lung

do

se

(GY

) O

nset

(d

ays)

Pneu

mon

itis

Type

O

utco

me

Not

m

easu

red

67

9.4

180

9.5

150

10.2

97

9.5

355

9.1

365

10.0

40

9

13.0

60

10.3

38

0

9.6

120

9.4

136

Idio

path

ic

Pncu

moc

ysti

s

Pneu

moc

ysti

s

Pneu

moc

ysti

s

Pneu

moc

ysti

s

Pneu

moc

ysti

s

Pneu

moc

ysti

s

CM

V

CM

V

CM

V

Her

pes

sim

plex

*Uni

que

patie

nt

num

ber.

Die

d

Die

d

Die

d

Die

d

Res

olve

d

Res

olve

d

Res

olve

d

Die

d

Res

olve

d

Res

olve

d

Res

olve

d

IPn following BMT 0 A. BARRE-IT et al. 1031

Table 2. The influence of granulocyte transfusions on the Table 3. Influence of donor sex on the incidence of incidence of interstitial pneumonitis interstitial pneumonitis

Granulocyte transfusion Incidence of I Pn

Lung complications

None Interstitial pneumonitis

(+) (-)

9 87 0 II

Donor sex Recipient sex ‘% No.

Male Male Female Female

Male Female Male Female

l2.S 4132 7.1 2128 8.3 2124

13.0 3123 Seventeen patients received methotrexate alone for

post graft immunosuppression, 75 received cyclosporin A alone and 15 cyclosporin A and methotrexate.

Lung function testing using spirometric and respira- metric techniques was undertaken before BMT and at intervals of 2 to 4 weeks after TBI. Any patient develop- ing respiratory problems was retested.

GvHD, again, not statistically significant. GvHD did not therefore predispose to the development of IPn in our series.

RESULTS

Clinically significant lung complications occurred in 11 out of 107 patients (10.34) and were fatal in 4.7%. There were IO cases of infective pneumonitis of which 5 were fatal (Table 1). Six patients had Pneumocystis carinii pneumonitis, 3 confirmed histologically and serologically. Three patients died, one within 100 days of grafting, while 3 out of 5 who became infected 100 days or more after grafting recovered. CMV pneumonitis was seen in 3 patients. One died 60 days after grafting; the other 2, who became infected at 120 and 395 days respectively, recov- ered. One patient had a Herpes simplex pneumonitis which resolved following treatment with acyclovir.

None of the patients who developed IPn had had granulocyte transfusions (Table 2).

In matched BMT recipients, comparison of post graft immunosuppression regimens showed a similar incidence of IPn in patients receiving methotrexate (6%) or cyclo- sporin A (6.6%)).

Of the 1 1 patients developing lung complications, 4 had received sex mismatched grafts and 7 sex matched grafts. The various combinations of donor and recipient sex did not appear to influence the incidence of IPn (Table 3). Of the 60 AML. patients transplanted in first remission, IO relapsed subsequently. This is apparently higher than the relapse rate in a comparable group of patients in Seattle (Table 4).9

DISCUSSION Only one case of idiopathic pneumonitis was seen. This

child, transplanted in 1974, before lithium fluoride do- simetry was used had had a Pneumocystis pneumonitis prior to TBI and BMT which resolved. The exact lung dose is unknown, but during TBI he suffered a respiratory arrest and subsequently remained unconscious for 5 days. He died 77 days after grafting. There was no evidence of infection at post mortem.

The incidence of 1Pn reported here is lower than that described by other transplant groups. In patients trans- planted in Seattle an incidence of 54% with a mortality rate of 34% is reported9 and from UCLA 39% mortality.’

The figures given in these papers relate, however, to heterogeneous groups of patients and make meaningful comparisons difficult. The substantially lower incidence in our group may be related to many factors other than

For the range of 991 I Gy the dose of radiation deliv- ered to the lungs did not influence the incidence of IPn. In patients receiving lung doses of < IO Gy, 7 of 69 developed IPn, compared with 4 of 38 patients receiving > 10 Gy, which was not statistically significant.

Four of 23 (17.4%) patients receiving mismatched grafts developed IPn compared with 7 of 84 (3.8%) with matched grafts. (p = 0.2)

Table 4. Relapse rates after TBI/BMT in 1st remission AML patients

Relapse rates

IPn occurred in 7 of 74 patients with GvHD (transient or prolonged) and in 4 of 33 patients with no evidence of

Center

RMH (this study)

Seattle”

No. of patients %

I O/60 16.7 6175 8

Table 5. incidence of IPn after TBI and BMT

Center Lung dose

No. of patients treated Dose rate

Incidence %

Mortality %

Seattle 9.20-I 1 Gy 100 0.0548 or 0.08 54 34 UCLA I I-l2.4Gy 112 0.0770. I Gy/ 39 RMH 9.55105 Gy 107 0.025 Gy/min II 4.5 Toronto 11.4-12.6 Gy I50 0.058 Gy/min 100 100

1032 Radiation Oncology 0 Biology 0 Physics July 1983, Volume 9, Number 7

the radiation given. Fifty-six percent of patients had AML in first remission and were therefore in good general condition having received relatively low doses of chemotherapy. Most were under 35 years of age, where a more favorable outcome has been reported by the Seattle group. The post-graft immunosuppression regimen does not appear to affect the incidence of IPn in spite of a probable direct toxic effect of cyclosporin A on pulmo- nary endothelium and a possibly increased susceptibility to viral infections.

ranged from 9.2 to 11 Gy (similar to those in our patients) although higher doses (11 to 12.4 Gy) may have been given by the UCLA group.

Estimation of lung dose is problematical. Measurement of entrance and exit doses using lithium fluoride dose meters overlying the lung should give an estimate of doses corrected for increased transmission through the lung. Variation will occur, however, throughout the lung fields because of variations in soft tissue thickness due to the heart, arms etc.

Drugs such as busulphan, which independently pro- duce lung damage, were not used. There may be a geographical difference in the distribution and frequency of cytomegalio virus infection since the incidence seems to be lower in Europe than in the United States of America; although Winston et al.,13 have suggested that multiple granulocyte transfusions may be a source of infection and thus related to IPn, we have found no such association.

Data presented here is the first to suggest that patients receiving mis-matched grafts have greater impairment of lung function’ and a higher incidence of IPn than patients receiving matched grafts. The reason for this is not clear but may be related to more intense immunosuppression or greater donor-host immuno-reactivity.

CT scan data have to be interpreted with caution. Scans should be performed in the treatment position which is difficult in some centers because of limitation of ring size. Our preliminary studies show that positioning of the arms may produce considerable variations in dose distribution, and more detailed calculations are needed to determine the significance of this. Dose meters are placed opposite each other at six sites on the chest wall and back (upper, middle and lower lung, right and left) and the doses quoted are the mean of 4 dose meters at the level giving the highest reading.

The radiation factor most likely to affect incidence of IPn is the total dose to the lungs. The Toronto group have demonstrated a threshold for overt lung damage at 8 Gy in patients receiving upper hemibody irradiation.” For patients without previous treatment, lung disease or tumor in the lung, there is a steep rise in incidence of IPn from 5% at 8.20 Gy to 50% at 9.30 Gy and 100% at 10.60 Gy, using dose rates of 50-400 rad/min.

Keane et al.’ have calculated absorbed doses to the lung for patients treated in various centers using CT data on lung density and tissue air ratio measurements. They suggest that doses to the lungs in the Seattle technique

In spite of the problems of accurate estimation of lung dose, it seems clear that the incidence of IPn in this series is lower than for the same doses given in other centers. A dose rate effect may therefore be important and has been demonstrated in animal systems for gut and skin.3*6,‘2 We have shown from measurements of diffusing capacity in mice that for TBI with dose rates below 0.05 Gy min’ no lung damage occurs, but increasing impairment occurs as the dose rate is increased above 0.05 up to 0.5 Gy/min.’ There is a possibility that at very low dose rates the potential of lung tissue to accumulate and repair sublethal damage is not exceeded and we postulate that lowering the dose rate of TBI to below 0.05 Gy/min will shift the threshold for significant IPn to 11-I 2 Gy.

There are conflicting reports of experimental data on

I I J 1

0 1 2 3 4

YEAFB SINCE UXPIZTE REMISSION

Fig. 1. Probability of survival of patients transplanted in first remission of AML. ---- Seattle; - RMH.

IPn following BMT 0 A. BARRET et al. 1033

the repair capacity of leukemic cells,‘~’ although because of relatively smaller shoulder size and extrapolation num- bers to the cell survival curve, use of low dose rate or fractionated radiation should improve the therapeutic ratio. Human patient data for comparable groups is scanty. The apparently higher relapse rate in first remis- sion AML patients at the Royal Marsden Hospital com- pared with Seattle would suggest this is not so. However, it must be remembered that lung deaths occur earlier than leukemic relapse so with a higher incidence of IPn,

less patients in Seattle are at risk for relapse. Survival curves for comparable patients show a similar pattern, although the causes of death are different (Fig. 1). More work is needed to determine whether ch?nges in log cell kill can be obtained by variations in dose rate or dose of irradiation. Differences in leukemic cell burden or indi- vidual tumor sensitivity may prove more important for leukemia control than cell kill by radiation and sparing of normal tissue damage may lead to an improved thera- peutic ratio even if relapse rates are unchanged.

REFERENCES I. Depledge, M.H., Barrett, A.: Dose rate dependence of lung

damage after total body irradiation in mice. Int. J. Radiat. Biol. 41: 3255334, 1982.

2. Depledge, M.H., Barrett, A., Powles, R.L.: Lung function after bone marrow grafting. Int. .I. Radiat. Oncol. Biol. Phys. (In press).

3. Hornsey, S., Alper, T.: An unexpected dose-rate effect in the killing of mice by radiation. Nature 20: 212-213, 1966.

4. Keane, T.J., Van Dyk, J., Rider, W.B.: Idiopathic intersti- tial pneumona following bone marrow transplantation: the relationship with total body irradiation. Int. J. Radiat. Biol. Phy. 7(10): 1365-1370, 1981.

5. Lawrence, G., Rosenbloom, M.E., Hickling, P.: A tech- nique for total body irradiation in the treatment of patients with acute leukaemia. Brit. J. Radiol. 53: 8944897, 1980.

6. Morrison, R.A., Zelimer, D.L., Dean, R.D.: Low vs high dose-rate effects on the acute reaction of pig skin to cobalt-60 gamma rays. Int. J. Radiat. 0~01. Biol. Phys. 7(3): 359-364. I98 I.

7. Peters, L.J., Withers, H.R., CunditT, J.H., Dicke, K.A.: Radiobiological Considerations in the use of total body irradiation for bone marrow transplantation. Radiology 131: 2433247, 1979.

8. Song, C.W., Kim, T.H., Khan, F.M., Kersey, J.H., Levitt, S.H.: Radiobiological basis of total body irradiation with different dose rate and fractionation: Repair capacity of

haemopoietic cells. Int. J. Radial. Biol. Phy. 7( 12): I695 1701, 1981.

9. Thomas, E.D., Buckner, C.D., Banaji, M., Clift, R.A., Fefer, A., Flournoy, N., Goodell, B.W.. Hickman, R.O., Lerner, K.G., Neiman, P.E., Sale, G.E., Sanders, J.E., Singer, J., Stevens, M., Storb, R., Weiden, P.L.: One hundred patients with acute leukaemia treated by chemo- therapy, total body irradiation and allogeneic marrow transplantation. Blood 49(4): 5 I I-533, 1977.

IO. Thomas, E.D., Clift, R.A., Buckner, C.D.: Marrow trans- plantation for patients with acute non-lymphoblastic leu- kaemia who achieve a first remission. Cancer Treat. Rep. 66(7): 1463-1466, 1982.

I I. Van Dyk, J., Keane, T.J., Kan, S., Rider, W.D., Fryer, C.J.H.: Radiation pneumonitis following large single dose irradiation: A re-evaluation based on absolute dose to lung. Int. J. Radiat. Oncol. Biol. Phys. 7: 46 I -~-467, I98 I.

12. Wambersie, A., Stienon-Smoes, M.-R., Octave-Prignot, M.. Dutreix, J.: Effect of dose rate on intestinal tolerance in mice. Implications in radiotherapy. Br. J. Radio/. 52: 1533155.1979.

13. Winston, D.J., Bryson, Y.J., Ho, W.G., Territo, M.C., Golde, D.W., Gale, R.P.: Interstitial pneumonia and cyto- megalovirus infection after bone marrow transplantation. In Biology of Bone Marrow Transplantation. London, Academic Press. 1980, pp. 83-95.