- Email: [email protected]
A randomized trial of intracavitary bleomycin and corynebacterium parvum in the control of malignant pleural effusions
Radiotherapy and Oncology, 14 (1989) 19-26 Elsevier
19
RTO00518
A randomized trial of intracavitary bleomycin and corynebacterium parvum in the control of malignant pleural effusions M.J. Ostrowski t, T.J. Priestman 2, R.F. H o u s t o n 3 and W.M.C. Martin 3 1Norfolk and Norwich Hospital, Norwich, Norfolk, U.K., ZDudley Road Hospital, Birmingham, U.K, and 3Belvoir Park Hospital, Belfast, Northern Ireland (Received 26 May 1987, revision received 30 May 1988, accepted 9 June 1988)
Key words." Malignant pleural effusion; Intracavitary bleomycin; Intracavitary corynebacterium parvum
Summary Fifty-eight patients with malignant pleural effusions were entered into a prospectively randomized clinical trial comparing the efficacy of a local instillation of bleomycin or corynebacterium parvum (C. parvum) in controlling fluid reaccumulation after simple needle aspiration (thoracentesis). The response was assessed at 30 days by chest X-ray and clinical examination. There were 44 evaluable patients; 18 of 25 (72%) of those receiving bleomycin and 9/19 (47%) of those who had C. parvum gained a complete or partial response. This difference in response rate was not statistically significant (p = 0.13). The majority of patients had an effusion from a primary breast carcinoma and the response in this group was almost statistically significant (p = 0.06) with 74% of bleomycin patients and 43% of C. parvum patients responding. Fever following instillation was more common with C. parvum (53% of patients compared with 24% after bleomycin, p = 0.02), whereas nausea was more common after bleomycin (28% vs. 10.5%, p = 0.16). Local chest pain after aspiration occurred in 52% of the bleomycin group and 47% of the C. parvum subjects. There was no significant difference between the groups in age, sex, tumour type, presenting symptoms, volume of aspirate, systemic therapy or number of previous aspirations. Both of these agents appeared to be active in the control of malignant pleural effusions although the response rate was higher with bleomycin and overall, both have acceptable levels of toxicity.
Addressfbr correspondence: Dr. M.J. Ostrowski, Department of Radiotherapy and Oncology, Norfolk and Norwich Hospital, Brunswiek Road, Norwich NRI 3SR, Norfolk, U.K. 0167-8140/89/$03.50 © 1989 Elsevier SciencePublishers B.V. (Biomedical Division)
20 Introduction
Materials and methods
Malignant pleural effusions are a common complication of advanced cancer with two thirds of cases occurring in patients with carcinomas of the breast, bronchus or a lymphoma. Up to 50% of cases of carcinoma of the breast or bronchus are said to develop a pleural effusion at some time during the course of their disease [6,9]. Simple needle aspiration (thoracentesis) alone, although effective for immediate relief of symptoms, is usually inadequate to prevent the reaccumulation of malignant pleural effusions as 97% [1] recur within 30 days and repeated aspirations increase the risk of complications. It is for this reason that a wide variety of agents have been instilled into the pleural cavity after aspiration in an attempt to overcome this problem. A number of studies [4,1214] have demonstrated the efficacy of intracavitary bleomycin which produces responses in 62-85% of cases and its overall effectiveness is probably greater than that of tetracycline [7]. In addition, the incidence of side effects such as pain and fever are low and it does not promote myelosuppression. Although its mode of action is unknown, systemic absorption of intracavitary bleomycin is approximately 45% [2] of the instilled dose. The findings at post mortem of fibrous adhesions [14] suggest that bleomycin stimulates an irritative pleuritis which induces pleuradesis although a local cytotoxic effect cannot be ruled out. Early clinical studies [5,11,15] with a freeze-dried preparation of corynebacterium parvum (C. parvum) demonstrated that it produced a vigorous inflammatory reaction in the pleura and prevented the reaccumulation of about 80% of effusions, with a significant decrease in the number of malignant cells present in the aspirate. It would appear that C. parvum acts by stimulating an acute inflammatory response, but it is doubtful if the immunostimulant effect seen with the systemic administration of this agent plays any part in the prevention of recurrent malignant effusions. In view of these encouraging reports it seemed appropriate to carry out a prospective comparison of the efficacy and toxicity of these two preparations in the control of malignant pleural effusions.
All patients had histologically proven malignancy with the presence of a pleural effusion proven, or presumed to be, secondary to that tumour and the aspiration of that effusion considered to be clinically indicated. Although no age limit was set, it was suggested that patients entered into the study should be in the age range of 18 to 75 years and that they should have a minimum life expectancy of 30 days. Previous aspiration of the effusion did not exclude patients from the study, provided that there had been no previous intrapleural drug instillation. No additional or change in systemic anti-tumour therapy was permitted 30 days before or after entry into the study. Patients accepted for the study were randomly allocated, by means of consecutively numbered sealed envelopes containing the treatment allocation, to receive either bleomycin or C. parvum, after aspiration of the effusion. Effusions were aspirated as completely as possible through a fine needle canula over a period of 30-60 min. On completion of the aspiration either 60 mg of bleomycin dissolved in 100 ml of normal saline or 7 mg of C. parvum dissolved in 20 ml of normal saline, were instilled into the pleural cavity via the aspirating needle. The needle was then removed and the patient's posture changed every 5 rain for 20 min to ensure adequate dispersion of the agent. Response was assessed at 30 days. A complete response (CR) was defined as no reaccumulation of fluid confirmed by chest X-ray. A partial response (PR) indicated minimal fluid reaccumulation not sufficient to produce symptoms and/or the need for a further aspiration. The duration of response was measured from the day of aspiration until the time of documented fluid reaccumulation, or death. The toxicity of treatment was closely monitored, particularly with respect to fever, local pain and nausea and any side effects which occurred were graded at the time as either mild, moderate or severe.
21 TABLE I Characteristics of evaluable patients.
No. of patients Age (mean ± S+D.) Sex (M:F) Primary tumour: Carcinoma of the breast Carcinoma of the bronchus Lymphoma Carcinoma of the pancreas Carcinoma of the stomach Renal cell carcinoma Average duration of effusions (wks) Pain (on presentation) Dyspnoea (on presentation) Vol. aspirate Positive cytology
Bleomycin
C. parvum
25 61 yrs (± 11.8) 5:20
19 59.1 yrs (+ 12.5) 4:15
19 (76%) 3 2 l 0 0 7.6" 36% 100% Mean 1.4 (0.3-3.8 1) 80%
14 (74%) 2 1 0 | 1 7.8 36.5% 100% Mean 1.45 (0.4-3.0 1) 60%
a Two patients had recorded effusions of long duration, 144 and 104 weeks. The mean duration, if included, increases to 16.9 weeks.
Results A total o f 58 p a t i e n t s were entered into this study, o f w h o m 30 were a l l o c a t e d to receive b l e o m y c i n a n d 28 C. p a r v u m . A l t h o u g h the p r o t o c o l stated that p a t i e n t s s h o u l d have a life e x p e c t a n c y o f at least 30 days, four p a t i e n t s in the b l e o m y c i n g r o u p a n d seven given C. p a r v u m , died o f a d v a n c i n g m a l i g n a n t disease before this assessment p e r i o d was r e a c h e d a n d were therefore n o n - e v a l u a b l e . In a d d i t i o n , one p a t i e n t a l l o c a t e d to receive b l e o m y c i n a n d two allocated to C. p a r v u m did n o t receive their drugs, due to difficulties with the a s p i r a t i o n a n d were also considered to be n o n - e v a l u a b l e . O f the r e m a i n i n g 44 p a t i e n t s w h o c o u l d be assessed at 30 days, 25 received b l e o m y c i n a n d 19 C. p a r v u m . T h e c h a r a c t e r istics o f the two t r e a t m e n t g r o u p s d e m o n s t r a t e d in T a b l e I are c o m p a r a b l e except that the average dur a t i o n o f effusion was 16.9 weeks in the b l e o m y c i n group, c o m p a r e d with 7.8 in the C. p a r v u m g r o u p . H o w e v e r , there were two patients in the b l e o m y c i n g r o u p who h a d a s y m p t o m a t i c effusions present for 144 a n d 104 weeks, respectively, before these deter i o r a t e d p r o d u c i n g s y m p t o m s requiring a s p i r a t i o n
for relief. If these two cases are deleted then the m e a n d u r a t i o n o f effusion for the b l e o m y c i n g r o u p falls to 7.6 weeks. T h e only o t h e r m a j o r difference between the two g r o u p s was that a higher percentage o f the b l e o m y c i n g r o u p h a d positve c y t o l o g y , 80% vs. 60%. T h e r e were 16 patients ( 6 4 % ) in the b l e o m y c i n g r o u p a n d 13 patients given C. p a r v u m (68%) w h o
TABLE II Established systemic treatment at entry to study. Treatment None Steroids Hormones Cytotoxics Steroids and hormones Steroids and cytotoxics Cytotoxics and hormones Diuretics Radiation to abdomen Total
Bleomycin C. parvum Total 6 5 4 2 2 3 1 1 1
4 2 4 4 3 2 0 0 0
l0 7 8 6 5 5 1 1 1
25
19
44
22 TABLE III Presenting symptoms. Severity
Dyspnoea
Pain Bleomycin
C. parvum
Total
Bleomycin
C. parvum
0 1 2 3
16 (64%) 7 (28%) 2 (8%) 0
12 (63%) 5 (26%) 0 2 (10.5%)
28 (63.6%) 12 (27.3%) 2 (4.5%) 2 (4.5%)
0 2 (8%) 13 (52%) 10 (40%)
0 1 (5%) 9 (47%) 9 (47%)
Total
25
19
44
25
had no previous aspiration with seven patients (28%) given bleomycin and five patients (26%) in the C. parvum arm, undergoing one previous aspiration. The details of concomitant systemic therapy which did not change for 30 days before or after the drug instillation are given in Table II and are comparable for the two groups. Pain and dyspnoea were the two presenting symptoms indicating a need for aspiration and their severity and distribution is shown in Table III with increasing dyspnoea, the symptom commonly alerting the clinician to the need to perform pleural aspiration. The only other additional recorded pre-treatment symptoms were a cough in one patient in each group, anorexia (one bleomycin patient) and nausea (one C. parvum patient). At aspiration the mean volume of aspirate was 1.4 1 in the bleomycin group (range 0.3 to 3.8 1) and
19
Total 0 3 (6/8%) 22 (50%) 19 (43.2%) 44
1.5 l (range 0.4 to 3.0 !) in the C. parvum patients (Table I). The response to treatment can be assessed in a number of different ways. One can either look at the number of evaluable patients responding at the initial 30-day assessment period or consider those patients who died of advancing disease before the initial 30-day assessment period as treatment failures, and assess the response of the whole treatment group. The former method is compatible with all the published data available on both of these agents and enables a comparison of effect to be made with these published series. Very few, if any, of the published work on the control of malignant effusions evalute results, using an intention to treat analysis method. In Table IV we have illustrated the response rates for evaluable patients at 30 days which indicate an overall response of 72% for the bleomy-
TABLE IV Response at 30 days (evaluable patients). Primary carcinoma of the breast
Whole group Bleomycin Complete responders (CR)
12 (48%) ]
6 (32%) 72%
Partial responders (PR) Non-responders (NR) Totals
6 (24%) 7 (28%) 25
Total
C. parvum
18
Bleomycin 9 (47%) "1
47% 3 (16%) 10 (53%) 19
C. parvum
74% 9 44
5 (26%) 5 (26%) 19
Total
3 (21%) "1 12 I 43% 3 (21%) 8 8 (57%) 13 14
33
23 TABLE V Duration of response. Response of evaluable patients (44)
30 Days
60 Days
90 Days
6 Months
9 Months
1 Year
Bleomycin(25) C. parvum (19)
18/25 (72%) 9/19 (47%)
17/25 (68%) 9/19 (47%)
14/25 (56%) 4/19 (24%)
8/25 (32%) 3/19 (16%)
5/25 (20%) 2/19 (10%)
4/25 (16%) 1/19 (5%)
18/29 (62%) 9/26 (34%)
17/29 (58%) 9/26 (34%)
14/29 (48%) 4/26 (15%)
8/29 (27%) 3/26 (11%)
5/29 (17%) 2/26 (7%)
4/29 (13%) 1/26 (3%)
18/25 (72%) 9/19 (47%)
17/24 (71%) 9/19 (47%)
14/22 (64%) 4/17 (24%)
8/17 (47%) 3/16 (19%)
5/16 (31%) 2/15 (13%)
4/15 (27%) 1/15 (7%)
Response of all patients treated (57) Bleomycin(29) C. parvum (26) No. of live patients recurrence-free Bleomycin(29) C. parvum (26)
cin group, compared with 47% for C. parvum, this difference is not statistically significant (p = 0.13). However, the overall response rates for patients with effusions from primary breast carcinoma (74% bleomycin vs. 43% C. parvum) approaches significance with a p value of 0.06. Corresponding figures for the response at 30 days of all patients treated are obviously lower at 62% for bleomycin and 34% for C. parvum (Table V). The duration of response is shown in Table V where the number of evaluable patients still responding at 60 days, 90 days, 6 months and one year are recorded and compared with the response rate achieved if all patients treated are included and non-evaluable patients considered to be nonresponders. In addition, these results are also compared with the number of live patients still effusion-free. Of the patients in the bleomycin group still responding at one year, two patients developed recurrent effusions at 56 and 61 weeks posttreatment and the other two patients died of advancing disease, but effusion-free at 61 and 73 weeks. The solitary C. parvum patient still alive and effusion-free at one year died at 56 weeks post-treatment of advanced disease. All of these patients with long-term responses had breast carcinoma.
The treatment of malignant pleural effusions is a palliative procedure, designed to give symptomatic relief and improve the patient's quality of life. The responses at 3 and 6 months in the bleomycin group of 64 and 47% (Fig. 1) respectively are especially encouraging as it means that these patients have 100%
90%ti\
8Oo/ot |ii
7o%. i
.....
•
Bleomycin
................. C. Parvum
"N"\.
60% i 5O9 40% 3O% 2O% 10%
N
"\. N. N
i
3/12
6/12
9/12
1 year
Fig. 1. Percentageof live patients responding to treatment.
25
Total
19
9 (47.3%) 3 (15.8%) 2 (10.5%) 5 (26.3%)
C. parvum
44
28 (63.6%) 6 (13.6%) 5 (11.4%) 5 (11.4%)
Total C. parvum 17 (89.5%) 0 2a (10.5%) 0 19
Bleomycin 18 (72%) 4 (16%) 1 (4%) 2 (8%) 25
Nausea
a One patient had continuous nausea present before aspiration. b One patient had continuous pain present before aspiration.
19 (76%) 3 (12%) 3 (12%) 0
Bleomycin
Fever
0 1 2 3
Severity
Side effects of bleomycin/C, parvum.
TABLE VI
44
35 (79.5%) 4 (9.1%) 3 (6.8%) 2 (4.5%)
Total
25
12 (48%) 6 (24%) 6 (24%) 1 (4%)
Bleomycin
Pain
19
9 (47.4%) 3 (15.8%) 5b (26.3%) 2 (10.5%)
C. parvum
44
21 (47.7%) 9 (20.4%) 11 (25%) 3 (6.8%)
Total
to
25 been kept out of hospital and avoided symptoms due to recurrent effusions, which is usually the norm with simple aspiration alone, where almost 100% of patients recur [1] within 30 days and repeated aspiration causes considerable inconvenience to the patient, as well as being tiresome and very debilitating. Fever, nausea and local pain were the side effects most commonly seen and their incidence and severity is shown in Table VI. Fever was more common after C. parvum instillation (52.7% vs. 24%, p = 0.02), whereas nausea was more common after bleomycin administration (28% vs. 10.5%, p = 0.18). Whilst the frequency of pain was similar in the two groups, a slightly higher proportion of the patients in the C. parvum group had most severe pain (37% compared with 28% for bleomycin). Most pain which lasted less than 24 h was associated with the aspiration procedure and easily controlled by mild analgesics. Other reported side effects were one patient receiving bleomycin who developed a mild skin rash and another in this group who collapsed 24 h after the instillation from an unknown cause. In the C. parvum group there was one patient who experienced moderate malaise of 24 h duration and a further patient who had a "flu-like" illness with general malaise for 4-5 days.
Discussion
Thoracentesis alone is inadequate to control malignant pleural effusions as fluid almost always reaccumulates within 30 days and often as rapidly as 3-4 days [1]. A recent review of the literature [7] indicated four approaches which gave average response rates in excess of 80%. These were pleurectomy and the instillation of either bleomycin, quinacrine or talc after aspiration. Recent reports have indicated that an instillation of C. parvum may also control malignant pleural effusions in at least 80% of patients [5,8,10,11,15]. Many patients with malignant effusions have advanced disease and treatment is usually symptomatic, directed at controlling the effusion with the minimum of inconvenience to the patient. For this
reason pleurectomy is not a popular approach to treatment despite the very high control rates achieved. Similarly, many of the studies reporting good results with talc instillations have involved thoracotomy and, even when the agent is introduced by intercostal cannula insufflation under local anaesthetic, there is a significant morbidity and so it should probably be reserved for selected goodrisk patients with the potential for a good life expectancy, when other methods have failed. Quinacrine, whilst being effective, often requires a multiple-dose technique and is associated with fever and pain in up to 80% of patients. These problems with the above therapies leave Neomycin and C. parvum as the remaining options for prevention of fluid reaccumulation. The results of this study support the already published data on the efficacy of bleomycin in controlling malignant pleural effusions by producing comparable response rates to other reported series. The relatively low toxicity associated with this agent, as noted by other studies is also seen. C. parvum is also shown to be effective, but the results obtained in this study are not as good as those from the other reported series using this agent. One reason for this may be that the other studies using C. parvum contained patients with a predominance of effusions from lung cancer, whereas in this study there was a high proportion of patients with a carcinoma of the breast. Whether this difference in underlying tumour type is relevant to response rates is impossib!e to determine at the present time but obviously requires further investigation. It is unfortunate that this study was closed prematurely due to the fall off in recruitment, as the number of patients treated was too small to draw any definitive conclusions about whether one agent was superior or not to the other. In order to detect a treatment difference of 25% (72% vs. 47%) with a power of 80% one would need at least 57 patients in each treatment arm. With the number of evaluable patients in this study the power to detect the 25% difference is just over 50%, assuming a significance level of 5%. Both of these agents appeared to be active in the control of malignant pleural effusions and there was some evidence of a higher response rate with bleo-
26 mycin. Overall, both agents a p p e a r to have acceptable levels of toxicity.
Acknowledgements The a u t h o r s acknowledge with grateful t h a n k s the c o n t r i b u t i o n to this study m a d e by Mrs. E. T o n g u e a n d Mr. P. Cooksey o f L u n d b e c k U . K . Ltd. who collated the data a n d provided the statistical analysis a n d Mrs. S. P a l g r a v e - M o o r e a n d Mrs. J. F a r r o w who typed the manuscript.
References 1 Anderson, C.B., Philpott, G.W. and Fergusion, T.B. The treatment of malignant pleural effusion.Cancer 33: 916-921, 1974. 2 Alberts, D.S., Chen, H.S.G., Mayersohn, M. et al. Bleomycin pharmacokinetics in man; 11, intracavitary administration. Cancer Chemother. Pharmacol. 2: 127-132, 1979. 3 Bayley, T.C., Kisner, D.L., Sybert, A. et al. Tetracycline and Quinacrine in the control of malignant effusions.Cancer 41: 1188-1192, 1978. 4 Bitrin, J.D., Brown, C., Desser, R.K. et al. Intracavitary Bleomycin for the control of malignant effusions. J. Surg. Oncol. 16: 273-277, 1981.
5 Felletti, R. and Rajazzoni, C. lntrapleural Corynebacterium parvum for malignant effusions.Thorax 38: 22-24, 1983. 6 Fracchia, A.A., Knapper, W.H., Carey, J.J. et al Intra pleural chemotherapy for effusion from metastatic breast carcinoma. Cancer 26: 626-629, 1970. 7 Hausheer, F.H. and Yarbro, J.W. Diagnosis and treatment of malignant pleural effusion. Semin. Oncol. 12: 54-75, 1985. 8 Leahy, B.C., Honeybourne, D., Brear, S.G. et al. Treatment of malignant pleural effusions with intrapleural Corynebacterium parvum or tetracycline.Eur. J. Respir. Dis. 66: 50-54, 1985. 9 Left, A., Hopewell, P.C. and Costello, J. Pleural effusionfrom malignancy. Ann. Intern. Med. 88: 532-537, 1978. 10 McLeod, D.T. and Calverley, P.M.A., Millar, J.W. et al. Further experience with Corynebacterium parvum in malignant pleural effusions.Thorax 40: 515-518, 1985. 11 Miller, J.W., Hunter, A.M. and Horne, N.W. Intrapleural immunotherapy with Corynebacterium parvum in recurrent malignant pleural effusions. Thorax 35: 856-858, 1980. 12 Ostrowski, M.J. An assessment of the long-term results of controlling the reaccumulation of malignant effusions using intracavitary Bleomycin.Cancer 57: 721-727, 1986. 13 Ostrowski, M.J., Halsall, G.M. lntracavitary Bleomycinin the management of malignant effusions:a multicentre study. Cancer Treat. Rep. 66: 1903-1907, 1982. 14 Paladine, W., Cunningham, T.J., Sponzo, R. et al. Introducing Bleomycinin the management of malignant effusions. Cancer 38: 1903-1908, 1976. 15 Webb, H.E., Oaten, S.N. and Pike, C.P. Treatment of malignant ascites and pleural effusions with Corynebacterium parvum. Br. Med. J. 1: 338-340, 1978.
19
RTO00518
A randomized trial of intracavitary bleomycin and corynebacterium parvum in the control of malignant pleural effusions M.J. Ostrowski t, T.J. Priestman 2, R.F. H o u s t o n 3 and W.M.C. Martin 3 1Norfolk and Norwich Hospital, Norwich, Norfolk, U.K., ZDudley Road Hospital, Birmingham, U.K, and 3Belvoir Park Hospital, Belfast, Northern Ireland (Received 26 May 1987, revision received 30 May 1988, accepted 9 June 1988)
Key words." Malignant pleural effusion; Intracavitary bleomycin; Intracavitary corynebacterium parvum
Summary Fifty-eight patients with malignant pleural effusions were entered into a prospectively randomized clinical trial comparing the efficacy of a local instillation of bleomycin or corynebacterium parvum (C. parvum) in controlling fluid reaccumulation after simple needle aspiration (thoracentesis). The response was assessed at 30 days by chest X-ray and clinical examination. There were 44 evaluable patients; 18 of 25 (72%) of those receiving bleomycin and 9/19 (47%) of those who had C. parvum gained a complete or partial response. This difference in response rate was not statistically significant (p = 0.13). The majority of patients had an effusion from a primary breast carcinoma and the response in this group was almost statistically significant (p = 0.06) with 74% of bleomycin patients and 43% of C. parvum patients responding. Fever following instillation was more common with C. parvum (53% of patients compared with 24% after bleomycin, p = 0.02), whereas nausea was more common after bleomycin (28% vs. 10.5%, p = 0.16). Local chest pain after aspiration occurred in 52% of the bleomycin group and 47% of the C. parvum subjects. There was no significant difference between the groups in age, sex, tumour type, presenting symptoms, volume of aspirate, systemic therapy or number of previous aspirations. Both of these agents appeared to be active in the control of malignant pleural effusions although the response rate was higher with bleomycin and overall, both have acceptable levels of toxicity.
Addressfbr correspondence: Dr. M.J. Ostrowski, Department of Radiotherapy and Oncology, Norfolk and Norwich Hospital, Brunswiek Road, Norwich NRI 3SR, Norfolk, U.K. 0167-8140/89/$03.50 © 1989 Elsevier SciencePublishers B.V. (Biomedical Division)
20 Introduction
Materials and methods
Malignant pleural effusions are a common complication of advanced cancer with two thirds of cases occurring in patients with carcinomas of the breast, bronchus or a lymphoma. Up to 50% of cases of carcinoma of the breast or bronchus are said to develop a pleural effusion at some time during the course of their disease [6,9]. Simple needle aspiration (thoracentesis) alone, although effective for immediate relief of symptoms, is usually inadequate to prevent the reaccumulation of malignant pleural effusions as 97% [1] recur within 30 days and repeated aspirations increase the risk of complications. It is for this reason that a wide variety of agents have been instilled into the pleural cavity after aspiration in an attempt to overcome this problem. A number of studies [4,1214] have demonstrated the efficacy of intracavitary bleomycin which produces responses in 62-85% of cases and its overall effectiveness is probably greater than that of tetracycline [7]. In addition, the incidence of side effects such as pain and fever are low and it does not promote myelosuppression. Although its mode of action is unknown, systemic absorption of intracavitary bleomycin is approximately 45% [2] of the instilled dose. The findings at post mortem of fibrous adhesions [14] suggest that bleomycin stimulates an irritative pleuritis which induces pleuradesis although a local cytotoxic effect cannot be ruled out. Early clinical studies [5,11,15] with a freeze-dried preparation of corynebacterium parvum (C. parvum) demonstrated that it produced a vigorous inflammatory reaction in the pleura and prevented the reaccumulation of about 80% of effusions, with a significant decrease in the number of malignant cells present in the aspirate. It would appear that C. parvum acts by stimulating an acute inflammatory response, but it is doubtful if the immunostimulant effect seen with the systemic administration of this agent plays any part in the prevention of recurrent malignant effusions. In view of these encouraging reports it seemed appropriate to carry out a prospective comparison of the efficacy and toxicity of these two preparations in the control of malignant pleural effusions.
All patients had histologically proven malignancy with the presence of a pleural effusion proven, or presumed to be, secondary to that tumour and the aspiration of that effusion considered to be clinically indicated. Although no age limit was set, it was suggested that patients entered into the study should be in the age range of 18 to 75 years and that they should have a minimum life expectancy of 30 days. Previous aspiration of the effusion did not exclude patients from the study, provided that there had been no previous intrapleural drug instillation. No additional or change in systemic anti-tumour therapy was permitted 30 days before or after entry into the study. Patients accepted for the study were randomly allocated, by means of consecutively numbered sealed envelopes containing the treatment allocation, to receive either bleomycin or C. parvum, after aspiration of the effusion. Effusions were aspirated as completely as possible through a fine needle canula over a period of 30-60 min. On completion of the aspiration either 60 mg of bleomycin dissolved in 100 ml of normal saline or 7 mg of C. parvum dissolved in 20 ml of normal saline, were instilled into the pleural cavity via the aspirating needle. The needle was then removed and the patient's posture changed every 5 rain for 20 min to ensure adequate dispersion of the agent. Response was assessed at 30 days. A complete response (CR) was defined as no reaccumulation of fluid confirmed by chest X-ray. A partial response (PR) indicated minimal fluid reaccumulation not sufficient to produce symptoms and/or the need for a further aspiration. The duration of response was measured from the day of aspiration until the time of documented fluid reaccumulation, or death. The toxicity of treatment was closely monitored, particularly with respect to fever, local pain and nausea and any side effects which occurred were graded at the time as either mild, moderate or severe.
21 TABLE I Characteristics of evaluable patients.
No. of patients Age (mean ± S+D.) Sex (M:F) Primary tumour: Carcinoma of the breast Carcinoma of the bronchus Lymphoma Carcinoma of the pancreas Carcinoma of the stomach Renal cell carcinoma Average duration of effusions (wks) Pain (on presentation) Dyspnoea (on presentation) Vol. aspirate Positive cytology
Bleomycin
C. parvum
25 61 yrs (± 11.8) 5:20
19 59.1 yrs (+ 12.5) 4:15
19 (76%) 3 2 l 0 0 7.6" 36% 100% Mean 1.4 (0.3-3.8 1) 80%
14 (74%) 2 1 0 | 1 7.8 36.5% 100% Mean 1.45 (0.4-3.0 1) 60%
a Two patients had recorded effusions of long duration, 144 and 104 weeks. The mean duration, if included, increases to 16.9 weeks.
Results A total o f 58 p a t i e n t s were entered into this study, o f w h o m 30 were a l l o c a t e d to receive b l e o m y c i n a n d 28 C. p a r v u m . A l t h o u g h the p r o t o c o l stated that p a t i e n t s s h o u l d have a life e x p e c t a n c y o f at least 30 days, four p a t i e n t s in the b l e o m y c i n g r o u p a n d seven given C. p a r v u m , died o f a d v a n c i n g m a l i g n a n t disease before this assessment p e r i o d was r e a c h e d a n d were therefore n o n - e v a l u a b l e . In a d d i t i o n , one p a t i e n t a l l o c a t e d to receive b l e o m y c i n a n d two allocated to C. p a r v u m did n o t receive their drugs, due to difficulties with the a s p i r a t i o n a n d were also considered to be n o n - e v a l u a b l e . O f the r e m a i n i n g 44 p a t i e n t s w h o c o u l d be assessed at 30 days, 25 received b l e o m y c i n a n d 19 C. p a r v u m . T h e c h a r a c t e r istics o f the two t r e a t m e n t g r o u p s d e m o n s t r a t e d in T a b l e I are c o m p a r a b l e except that the average dur a t i o n o f effusion was 16.9 weeks in the b l e o m y c i n group, c o m p a r e d with 7.8 in the C. p a r v u m g r o u p . H o w e v e r , there were two patients in the b l e o m y c i n g r o u p who h a d a s y m p t o m a t i c effusions present for 144 a n d 104 weeks, respectively, before these deter i o r a t e d p r o d u c i n g s y m p t o m s requiring a s p i r a t i o n
for relief. If these two cases are deleted then the m e a n d u r a t i o n o f effusion for the b l e o m y c i n g r o u p falls to 7.6 weeks. T h e only o t h e r m a j o r difference between the two g r o u p s was that a higher percentage o f the b l e o m y c i n g r o u p h a d positve c y t o l o g y , 80% vs. 60%. T h e r e were 16 patients ( 6 4 % ) in the b l e o m y c i n g r o u p a n d 13 patients given C. p a r v u m (68%) w h o
TABLE II Established systemic treatment at entry to study. Treatment None Steroids Hormones Cytotoxics Steroids and hormones Steroids and cytotoxics Cytotoxics and hormones Diuretics Radiation to abdomen Total
Bleomycin C. parvum Total 6 5 4 2 2 3 1 1 1
4 2 4 4 3 2 0 0 0
l0 7 8 6 5 5 1 1 1
25
19
44
22 TABLE III Presenting symptoms. Severity
Dyspnoea
Pain Bleomycin
C. parvum
Total
Bleomycin
C. parvum
0 1 2 3
16 (64%) 7 (28%) 2 (8%) 0
12 (63%) 5 (26%) 0 2 (10.5%)
28 (63.6%) 12 (27.3%) 2 (4.5%) 2 (4.5%)
0 2 (8%) 13 (52%) 10 (40%)
0 1 (5%) 9 (47%) 9 (47%)
Total
25
19
44
25
had no previous aspiration with seven patients (28%) given bleomycin and five patients (26%) in the C. parvum arm, undergoing one previous aspiration. The details of concomitant systemic therapy which did not change for 30 days before or after the drug instillation are given in Table II and are comparable for the two groups. Pain and dyspnoea were the two presenting symptoms indicating a need for aspiration and their severity and distribution is shown in Table III with increasing dyspnoea, the symptom commonly alerting the clinician to the need to perform pleural aspiration. The only other additional recorded pre-treatment symptoms were a cough in one patient in each group, anorexia (one bleomycin patient) and nausea (one C. parvum patient). At aspiration the mean volume of aspirate was 1.4 1 in the bleomycin group (range 0.3 to 3.8 1) and
19
Total 0 3 (6/8%) 22 (50%) 19 (43.2%) 44
1.5 l (range 0.4 to 3.0 !) in the C. parvum patients (Table I). The response to treatment can be assessed in a number of different ways. One can either look at the number of evaluable patients responding at the initial 30-day assessment period or consider those patients who died of advancing disease before the initial 30-day assessment period as treatment failures, and assess the response of the whole treatment group. The former method is compatible with all the published data available on both of these agents and enables a comparison of effect to be made with these published series. Very few, if any, of the published work on the control of malignant effusions evalute results, using an intention to treat analysis method. In Table IV we have illustrated the response rates for evaluable patients at 30 days which indicate an overall response of 72% for the bleomy-
TABLE IV Response at 30 days (evaluable patients). Primary carcinoma of the breast
Whole group Bleomycin Complete responders (CR)
12 (48%) ]
6 (32%) 72%
Partial responders (PR) Non-responders (NR) Totals
6 (24%) 7 (28%) 25
Total
C. parvum
18
Bleomycin 9 (47%) "1
47% 3 (16%) 10 (53%) 19
C. parvum
74% 9 44
5 (26%) 5 (26%) 19
Total
3 (21%) "1 12 I 43% 3 (21%) 8 8 (57%) 13 14
33
23 TABLE V Duration of response. Response of evaluable patients (44)
30 Days
60 Days
90 Days
6 Months
9 Months
1 Year
Bleomycin(25) C. parvum (19)
18/25 (72%) 9/19 (47%)
17/25 (68%) 9/19 (47%)
14/25 (56%) 4/19 (24%)
8/25 (32%) 3/19 (16%)
5/25 (20%) 2/19 (10%)
4/25 (16%) 1/19 (5%)
18/29 (62%) 9/26 (34%)
17/29 (58%) 9/26 (34%)
14/29 (48%) 4/26 (15%)
8/29 (27%) 3/26 (11%)
5/29 (17%) 2/26 (7%)
4/29 (13%) 1/26 (3%)
18/25 (72%) 9/19 (47%)
17/24 (71%) 9/19 (47%)
14/22 (64%) 4/17 (24%)
8/17 (47%) 3/16 (19%)
5/16 (31%) 2/15 (13%)
4/15 (27%) 1/15 (7%)
Response of all patients treated (57) Bleomycin(29) C. parvum (26) No. of live patients recurrence-free Bleomycin(29) C. parvum (26)
cin group, compared with 47% for C. parvum, this difference is not statistically significant (p = 0.13). However, the overall response rates for patients with effusions from primary breast carcinoma (74% bleomycin vs. 43% C. parvum) approaches significance with a p value of 0.06. Corresponding figures for the response at 30 days of all patients treated are obviously lower at 62% for bleomycin and 34% for C. parvum (Table V). The duration of response is shown in Table V where the number of evaluable patients still responding at 60 days, 90 days, 6 months and one year are recorded and compared with the response rate achieved if all patients treated are included and non-evaluable patients considered to be nonresponders. In addition, these results are also compared with the number of live patients still effusion-free. Of the patients in the bleomycin group still responding at one year, two patients developed recurrent effusions at 56 and 61 weeks posttreatment and the other two patients died of advancing disease, but effusion-free at 61 and 73 weeks. The solitary C. parvum patient still alive and effusion-free at one year died at 56 weeks post-treatment of advanced disease. All of these patients with long-term responses had breast carcinoma.
The treatment of malignant pleural effusions is a palliative procedure, designed to give symptomatic relief and improve the patient's quality of life. The responses at 3 and 6 months in the bleomycin group of 64 and 47% (Fig. 1) respectively are especially encouraging as it means that these patients have 100%
90%ti\
8Oo/ot |ii
7o%. i
.....
•
Bleomycin
................. C. Parvum
"N"\.
60% i 5O9 40% 3O% 2O% 10%
N
"\. N. N
i
3/12
6/12
9/12
1 year
Fig. 1. Percentageof live patients responding to treatment.
25
Total
19
9 (47.3%) 3 (15.8%) 2 (10.5%) 5 (26.3%)
C. parvum
44
28 (63.6%) 6 (13.6%) 5 (11.4%) 5 (11.4%)
Total C. parvum 17 (89.5%) 0 2a (10.5%) 0 19
Bleomycin 18 (72%) 4 (16%) 1 (4%) 2 (8%) 25
Nausea
a One patient had continuous nausea present before aspiration. b One patient had continuous pain present before aspiration.
19 (76%) 3 (12%) 3 (12%) 0
Bleomycin
Fever
0 1 2 3
Severity
Side effects of bleomycin/C, parvum.
TABLE VI
44
35 (79.5%) 4 (9.1%) 3 (6.8%) 2 (4.5%)
Total
25
12 (48%) 6 (24%) 6 (24%) 1 (4%)
Bleomycin
Pain
19
9 (47.4%) 3 (15.8%) 5b (26.3%) 2 (10.5%)
C. parvum
44
21 (47.7%) 9 (20.4%) 11 (25%) 3 (6.8%)
Total
to
25 been kept out of hospital and avoided symptoms due to recurrent effusions, which is usually the norm with simple aspiration alone, where almost 100% of patients recur [1] within 30 days and repeated aspiration causes considerable inconvenience to the patient, as well as being tiresome and very debilitating. Fever, nausea and local pain were the side effects most commonly seen and their incidence and severity is shown in Table VI. Fever was more common after C. parvum instillation (52.7% vs. 24%, p = 0.02), whereas nausea was more common after bleomycin administration (28% vs. 10.5%, p = 0.18). Whilst the frequency of pain was similar in the two groups, a slightly higher proportion of the patients in the C. parvum group had most severe pain (37% compared with 28% for bleomycin). Most pain which lasted less than 24 h was associated with the aspiration procedure and easily controlled by mild analgesics. Other reported side effects were one patient receiving bleomycin who developed a mild skin rash and another in this group who collapsed 24 h after the instillation from an unknown cause. In the C. parvum group there was one patient who experienced moderate malaise of 24 h duration and a further patient who had a "flu-like" illness with general malaise for 4-5 days.
Discussion
Thoracentesis alone is inadequate to control malignant pleural effusions as fluid almost always reaccumulates within 30 days and often as rapidly as 3-4 days [1]. A recent review of the literature [7] indicated four approaches which gave average response rates in excess of 80%. These were pleurectomy and the instillation of either bleomycin, quinacrine or talc after aspiration. Recent reports have indicated that an instillation of C. parvum may also control malignant pleural effusions in at least 80% of patients [5,8,10,11,15]. Many patients with malignant effusions have advanced disease and treatment is usually symptomatic, directed at controlling the effusion with the minimum of inconvenience to the patient. For this
reason pleurectomy is not a popular approach to treatment despite the very high control rates achieved. Similarly, many of the studies reporting good results with talc instillations have involved thoracotomy and, even when the agent is introduced by intercostal cannula insufflation under local anaesthetic, there is a significant morbidity and so it should probably be reserved for selected goodrisk patients with the potential for a good life expectancy, when other methods have failed. Quinacrine, whilst being effective, often requires a multiple-dose technique and is associated with fever and pain in up to 80% of patients. These problems with the above therapies leave Neomycin and C. parvum as the remaining options for prevention of fluid reaccumulation. The results of this study support the already published data on the efficacy of bleomycin in controlling malignant pleural effusions by producing comparable response rates to other reported series. The relatively low toxicity associated with this agent, as noted by other studies is also seen. C. parvum is also shown to be effective, but the results obtained in this study are not as good as those from the other reported series using this agent. One reason for this may be that the other studies using C. parvum contained patients with a predominance of effusions from lung cancer, whereas in this study there was a high proportion of patients with a carcinoma of the breast. Whether this difference in underlying tumour type is relevant to response rates is impossib!e to determine at the present time but obviously requires further investigation. It is unfortunate that this study was closed prematurely due to the fall off in recruitment, as the number of patients treated was too small to draw any definitive conclusions about whether one agent was superior or not to the other. In order to detect a treatment difference of 25% (72% vs. 47%) with a power of 80% one would need at least 57 patients in each treatment arm. With the number of evaluable patients in this study the power to detect the 25% difference is just over 50%, assuming a significance level of 5%. Both of these agents appeared to be active in the control of malignant pleural effusions and there was some evidence of a higher response rate with bleo-
26 mycin. Overall, both agents a p p e a r to have acceptable levels of toxicity.
Acknowledgements The a u t h o r s acknowledge with grateful t h a n k s the c o n t r i b u t i o n to this study m a d e by Mrs. E. T o n g u e a n d Mr. P. Cooksey o f L u n d b e c k U . K . Ltd. who collated the data a n d provided the statistical analysis a n d Mrs. S. P a l g r a v e - M o o r e a n d Mrs. J. F a r r o w who typed the manuscript.
References 1 Anderson, C.B., Philpott, G.W. and Fergusion, T.B. The treatment of malignant pleural effusion.Cancer 33: 916-921, 1974. 2 Alberts, D.S., Chen, H.S.G., Mayersohn, M. et al. Bleomycin pharmacokinetics in man; 11, intracavitary administration. Cancer Chemother. Pharmacol. 2: 127-132, 1979. 3 Bayley, T.C., Kisner, D.L., Sybert, A. et al. Tetracycline and Quinacrine in the control of malignant effusions.Cancer 41: 1188-1192, 1978. 4 Bitrin, J.D., Brown, C., Desser, R.K. et al. Intracavitary Bleomycin for the control of malignant effusions. J. Surg. Oncol. 16: 273-277, 1981.
5 Felletti, R. and Rajazzoni, C. lntrapleural Corynebacterium parvum for malignant effusions.Thorax 38: 22-24, 1983. 6 Fracchia, A.A., Knapper, W.H., Carey, J.J. et al Intra pleural chemotherapy for effusion from metastatic breast carcinoma. Cancer 26: 626-629, 1970. 7 Hausheer, F.H. and Yarbro, J.W. Diagnosis and treatment of malignant pleural effusion. Semin. Oncol. 12: 54-75, 1985. 8 Leahy, B.C., Honeybourne, D., Brear, S.G. et al. Treatment of malignant pleural effusions with intrapleural Corynebacterium parvum or tetracycline.Eur. J. Respir. Dis. 66: 50-54, 1985. 9 Left, A., Hopewell, P.C. and Costello, J. Pleural effusionfrom malignancy. Ann. Intern. Med. 88: 532-537, 1978. 10 McLeod, D.T. and Calverley, P.M.A., Millar, J.W. et al. Further experience with Corynebacterium parvum in malignant pleural effusions.Thorax 40: 515-518, 1985. 11 Miller, J.W., Hunter, A.M. and Horne, N.W. Intrapleural immunotherapy with Corynebacterium parvum in recurrent malignant pleural effusions. Thorax 35: 856-858, 1980. 12 Ostrowski, M.J. An assessment of the long-term results of controlling the reaccumulation of malignant effusions using intracavitary Bleomycin.Cancer 57: 721-727, 1986. 13 Ostrowski, M.J., Halsall, G.M. lntracavitary Bleomycinin the management of malignant effusions:a multicentre study. Cancer Treat. Rep. 66: 1903-1907, 1982. 14 Paladine, W., Cunningham, T.J., Sponzo, R. et al. Introducing Bleomycinin the management of malignant effusions. Cancer 38: 1903-1908, 1976. 15 Webb, H.E., Oaten, S.N. and Pike, C.P. Treatment of malignant ascites and pleural effusions with Corynebacterium parvum. Br. Med. J. 1: 338-340, 1978.