European Radiology
© Springer-Verlag 2004
10.1007/s00330-004-2240-5
Клиническое исследование, посвященное аспектам использования метода
химиоэмболизации (ХЭ) в качестве паллиативного лечения метастазов в позвоночный
столб и таз. Исследование проводилось во Франции, в нем принимали участие 25
пациентов (средний возраст – 59 лет), у которых не наблюдалось улучшений после
курсов химио- и лучевой терапии. Для химиоэмболизации использовались частицы
поливинилалкоголя с ПИРАРУБИЦИНОМ.
Теги: Интервенционная хирургия – Хемоинфузии – Химиоэмболизация – Метастазы в
кость – Пирарубицин – Франция – 2004.
Musculoskeletal
Selective intra-arterial chemoembolization of
pelvic and spine bone metastases
Jacques Chiras1 , Carmen Adem1, Jean-Noël Vallée1, Laurent Spelle1, Evelyne Cormier1
and Michèle Rose1
(1) Department of Neuroradiology, Groupe Hospitalier Pitié-Salpêtrière, 47, boulevard de l Hôpital,
75651 Paris, France
Jacques Chiras
Email: jacques.chiras@psl.ap-hop-paris.fr
Phone: +33-1-42163602
Fax: +33-1-42163598
Received: 28 January 2002 Revised: 30 December 2002 Accepted: 5 January 2004 Published
online: 19 March 2004
Abstract The purpose of this study was to determine the effect of interventional palliative
therapy by using chemoembolization on metastatic bone pain and tumor bulk in inoperable
metastases where chemotherapy and radiotherapy had failed. Twenty-five patients (mean age:
59 years) underwent chemoembolization of symptomatic lytic lesions involving the spinal
column (n=10), iliac bone and sacrum (n=15). The study design consisted of at least three
procedures based on combined chemoembolization performed under analog-sedation.
Therapeutic agents were carboplatin for selective chemotherapy and pirarubicin mixed with
polyvinyl alcohol foam for chemoembolization. Fifteen of 18 (83%) patients had significant pain
relief, as shown by the decrease of analgesic drug use. Mean clinical response duration was
12 months. Radiologically, ten patients were stable. A partial response was observed in four
patients, while a complete response was seen in two others. Selective intra-arterial
chemoembolization gives longer pain relief than embolization, compared to the literature data,
probably because of partial response with local anti-cancer drugs.
Keywords Interventional procedure-arteries - Chemotherapeutic infusion-arteries Chemotherapeutic embolization-arteries - Bone metastases
Introduction
Local pain secondary to tumor growth in inoperable metastases is a difficult therapeutic problem
after failure of both chemotherapy and radiotherapy. In order to prevent fracture and induce pain
relief, acrylic cementation is a well-documented alternative therapy. However, in some cases
where large or predominant involvement of soft tissue or anatomical sites such as the sacrum are
seen, this technique cannot be performed [1, 2]. Some reports of palliative embolization showed
temporary pain relief, but no effective tumoral response [3–15].
To improve the quality of life in those patients and obtain a tumoral response, some authors have
combined the use of local chemotherapy and embolization [6, 16]. Kato et al. [6] treated 20
patients who had intractable carcinomas, two of whom had painful bone metastases, by intraarterial infusion of mitomycin C microcapsules and observed pain relief. Courtheoux et al. [16]
reported on this technique in 12 cases of thoracic and lumbar spine metastases using mitomycin
C microspheres. Our experience in both fields of selective chemotherapy in brain tumors [17, 18]
and palliative embolization [14] of bone metastases led us to extend the use of the combination
of these techniques as a palliative treatment. Our aim was to evaluate the effect of
chemoembolization on metastatic bone pain and tumor response.
Materials
Twenty-five patients (18 men, 7 women; age range: 35–75 years; mean: 59 years) with solitary
(n=18) or multiple (n=7) metastases involving the iliac bone or spine were referred to our
institution from September 1996 to November 1999 for palliative chemoembolization after
failure of all usual general and/or local treatment, including surgery, chemotherapy and radiation
therapy (Table 1). Bone metastases were defined as lytic lesions with tumor bulk extending to
adjacent soft tissue.
Table 1 Patient characteristics
Patient no. Gender Age Primary cancer
Metastases location
1
M
70
Prostate
Spinal column T9–T11
2
F
30
Breast
Iliac bone
3
F
60
Kidney
Spinal column L1–L3
4
M
58
Kidney
Sacro-iliac
5
M
64
Kidney
Spinal column L1–L3
6
M
64
Kidney
Iliac bone
7
M
65
Lung
Spinal column
8
M
35
Kidney
Spinal column L2–L4, sacrum, iliac bone
9
M
65
Kidney
Spinal column L2–L4
10
M
70
Kidney
Spinal column T11-L1
11
M
38
Lung
Sacrum
12
M
50
Gastro-intestinal tract Sacrum
13
M
57
Gastro-intestinal tract Iliac bone
Patient no. Gender Age Primary cancer
Metastases location
14
F
66
Breast
Iliac bone
15
M
75
Prostate
Iliac bone
16
M
70
Soft tissue
Iliac bone
17
M
55
Soft tissue
Iliac bone
18
F
56
Unknown
Spinal column L4–L5
19
F
64
Breast
Iliac bone
20
M
70
Gastro-intestinal tract Spinal column L3–L5
21
M
63
Bladder
Spinal column L4–L5
22
F
34
Soft tissue
Spinal column L4–L5
23
F
62
Soft tissue
Iliac bone
24
M
54
Lung
Spinal column T7–T9
25
M
70
Kidney
Iliac bone
Patients had analgesic-resistant pain. All patients had contraindications to percutaneous injection
of acrylic surgical cement because of extensive lytic bone metastases and major tumor bulk.
Metastases had different origins (Table 2): most were from kidney cancer (eight patients) and
sarcomas (four patients), then with equal frequency breast, lung and GI cancers (three patients).
Table 2 Distribution of primary cancer
Metatases
Location
Primary cancer
Number
Kidney
8
Solitary Multiple Spine Iliac/sacrum Other
6
2
3
4a
2
Soft tissue tumor
4
2
2
Breast
3
1
2
Lung
3
3
Gastro-intestinal tract 3
2
Prostate
2
Bladder
Unknown
aIn
1
3
3
2
1
1
2
2
1
1
1
1
1
1
1
1
1
one case both locations were treated.
Symptomatic lesions involved the spinal column in 10 cases and the iliac bone and sacrum in 15
cases. In two of these cases, the lesion was a large recurrence in the area of surgical removal
with secondary iliac bone involvement.
Metastases were multifocal in 7/25 cases and solitary in 18/25 (metastases involving adjacent
vertebrae in the spine were considered as one metastatic site). Twenty-five of 28 metastatic sites
previously had received local treatment by radiation therapy. Among these, six patients
previously had undergone surgery at the same metastatic site. All patients were in local relapse
with increasing pain. The decision concerning the treatment by chemoembolization was made by
members of multidisciplinary hospital staff, including oncologists, orthopedists, radiation
therapists and interventional radiologists.
Methods
The procedure was performed via the femoral artery through a 5-F introducer after
administration of a local anesthetic: 10 ml of lidocaine hydrochloride (Astra, Nanterre, France).
Therapeutic agents were carboplatin (Paraplatine, Bristol-Myers Squibb, 92044 Paris La
Défense, France) for selective chemotherapy and pirarubicine (Theprubicine, Rhône-Poulenc
Rorer, 75013 Paris, France) mixed with polyvinyl alcohol particles (PVA) for
chemoembolization (150–300 and 300–600
m).
Angiography was performed first and arteries feeding the tumor localized. Before
chemoembolization of vertebral metastases, anterior and posterior spinal arteries were identified.
Chemoembolization combining pirarubicine and PVA with a decrease in blood flow was
performed after superselective cathetherism of each arterial feeder, except those arising from the
same arteries as the spinal arteries. It was then followed by selective chemotherapy with
carboplatin (Fig. 1).
Fig. 1 A 35-year-old female. Sarcoma metastasis of L4–L5. a Angiography shows a hypervascularized
mass (arrow) involving L4, L5 and adjacent soft tissues. b, c Selective angiography of common trunk for
right and left L4 and median sacral arteries: (b) before chemoembolization, (c) after chemoembolization
with PVA foam mixed with 10 mg of pirarubicine and selective infusion of 100 mg of carboplatin. d, e
Selective angiography of the iliolumbar artery. (d) Before chemoembolization and (e) after
chemoembolization with a mixture of 5 mg of pirarubicine, PVA foam and selective infusion of 50 mg of
carboplatin. f, g MR axial T1-weighted image without contrast. (f) Before treatment a soft tissue mass
involving the posterior arch of the vertebra and adjacent soft tissue (arrow). (g) After two sessions of
chemoembolization: complete disappearance of the tumoral mass
The treatment protocol consisted of three or more procedures of combined chemoembolization
and selective intra-arterial chemotherapy (IAC).
Procedures were separated by a 1-month time period. The number of arteries infused during the
procedure ranged from three to six. For each procedure, the total dose of pirarubicine and
carboplatin ranged from 10 to 20 mg and 200 to 400 mg, respectively. The total dose was
divided into the different arteries feeding the metastasis according to the volume of the tumor
bed.
For small arteries such as intercostal arteries, the dose was 3 mg of pirarubicine and 50 mg of
carboplatin. However, for arteries with a diameter of 4 mm or greater, the dose was 10 mg of
pirarubicine and 200 mg of carboplatin.
The number of procedures ranged from one to three (mean: two). Three patients had twice three
procedures delayed, for each patient, by different period of time (6, 12 and 18 months).
All procedures were performed under analgo-sedation with alfentanil chlorhydrate 1–2 mg/h
(Janssen-Cilag, Boulogne-Billancourt, France), a bolus of clonazepam 0.2 mg (Roche, Neuillysur-Seine, France) or a bolus of midazolam 1–2 mg. There was no need for assisted ventilation
with the administered doses. Based on their analgesic and anti-edema effect, intravenous
administration of steroids (120 mg) associated with granisetron chlorhydrate 3 mg (SmithKline
Beecham, Nanterre, France) was given after each procedure. Clinical and radiological follow-up
was performed at the 1st and 4th weeks.
Results were analyzed according to two criteria:
– Analgesic effect was evaluated in pain relief duration and in its intensity using a visual
scale (1-best to 10-worst)
– Tumoral response was evaluated based on imaging data (CT or MRI)
Eighteen of the 25 patients were evaluated after a minimum of two procedures.
Results
Analgesic response definitions [1, 2]
Pain relief was estimated as follows:
– Total improvement: complete pain relief with no need for analgesic medication
– Clear improvement: enough pain relief to decrease analgesic drugs down to 50% or to
replace a narcotic drug by a non-narcotic one
– Moderate improvement: pain relief with a decrease of less than 50% of the dose of
analgesic drugs
Significant pain relief comprised total and clear improvements. An increase of pain immediately
following the procedure and remaining for 2–10 days was observed in 50% of patients.
Following the postoperative period, 15/18 (83%) had significant pain relief with a decrease of
analgesic drug use. Among these, 10/18 (55%) had complete pain relief. This clinical response
duration ranged from 4 to 36 months (mean: 12 months).
Radiological response definitions
Radiological response was estimated as follows:
– Complete response: decrease of more than 50% in the tumor size or volume or appearance
of calcification within the metastasis
– Stable response: no change in tumor size in follow-up
– Partial response: decrease of less than 50% of the tumor size
Radiological response was available for evaluation in 65% of patients (16/25). Ten patients were
stable. Partial and complete responses were observed in four and two patients, respectively
(Fig. 1).
Because of the sample size, no relationship could be established between primary cancer and
tumoral response, except for the kidney primary. Among eight patients who have metastatic
renal cell carcinoma, a clinical and a radiological response were observed in four and two cases,
respectively (Fig. 2).
Fig. 2 A 56-year-old male with recurrence of local pain 1 year after surgery and radiation therapy for
lumbar metastases of renal carcinoma. a CT-scan showing osteolysis of the vertebral body (short arrow),
bone graft (long arrow) and soft tissue involvement. b Selective angiography of third left lumbar artery
(arrow) showing intense vascularization of tumoral vertebrae. c Selective angiography of the third right
lumbar artery. Intense hypervascularization of the mass (arrow). d Three years after two sessions of
chemoembolization with pirarubicine and carboplatin, the scan shows a good recalcification of the
vertebra (arrow) and a disappearance of soft tissue involvement. The patient has no pain with a 100-m
walking area.
No major side effects reported. However, one case of rectal hemorrhage, spontaneously
regressive without treatment, was observed following chemoembolisation of the iliolumbar
artery.
Discussion
A curative approach of metastatic tumors of the spinal column is now well established. On one
hand, preoperative embolization of highly vascular lesions reduces blood loss, tumoral volume
and allows a safer, faster and more complete procedure [5, 6, 8, 14, 15, 19, 20]. On the other
hand, tumor resectability is not possible due to its centricity (unique or multiple), tumor size or
the patient s general status. Palliative therapy is proposed when treatments such as general
chemotherapy, irradiation, hormonal therapy or surgery are not possible or not expected to
succeed. By inhibiting and reducing tumor growth, which is responsible for spinal or nerve root
compression, pain relief is the result of this therapy [5, 6, 8–10, 12, 13, 15–19, 21–23]
Local palliative treatments may include percutaneous injection of acrylic surgical cement,
embolization and chemoembolization.
Percutaneous injection of acrylic surgical cement in bone metastases gives notable clinical pain
relief. However, it requires partial preservation of bone limits and cannot be performed in case of
a large mass or after orthopedic treatment with bone grafts or prosthesis [1, 2]
In our review of the literature (Table 3), selective intra-arterial embolization of primary and
metastatic bone tumors was first reported by Feldman et al. in 1975 [3] to reduce local pain.
Table 3 Review of the literature of bone tumor embolization
Palliative bone
embolization
Duration of pain Mean survival
relief
time
1975 2
1
2.5 months
2.5 months
Chuang [5] 1979 9
6
1–6 months
3.5 months–
1 year
Wallace
[4]
1979 9
9
2 months–
5.5 yearsa
Soo [8]
1982 15
15
3 weeks–4 yearsa
Braedel
[11]
1984 4
2
15 months
Treves [9] 1984 4
4
13 months
Varma [10] 1984 5
5
6–18 months
O Reilly
[12]
1989 4
4
3–10.5 months
Barton
[15]
1996 51
29
2–8 months
Chiras [14] 1996 53
22
1–6 monthsb
Author
Year
Feldman
[3]
aGiant
bMore
Total number
of cases
1 month–
4 yearsa
17 months
cell tumor.
than 3 years for the treatment of two vertebral hemangiomas.
A large variety of agents have been used [24, 25]: gelatin sponge (Gelfoam, Upjohn, Kalamazoo,
Mich) [3, 5, 8, 10], PVA foam (Ivalon, Nycomed Medical Systems, Paris) [8, 12, 14, 15],
stainless steel coils [5, 8, 10, 15, 25], synthetic tissue adhesive (cyanoacrylate mixed with
lipiodol) [9, 26] and microfibrillar collagen [12]. Only a few studies have reported the use of
microencapsulated mitomycin as a local anti-neoplastic agent [6, 16].
According to this review, a satisfactory decrease of pain was observed in most cases, but
duration of pain relief did not exceed 6 months in most cases in the two largest series [14, 15],
but a longer response was observed by Wallace and Soo in primary tumors [4, 8].
In 1978, Kato et al. introduced the chemoembolization concept [6]: chemoembolization
combines the benefits of local chemotherapy and embolization.
Local chemotherapy has the advantages of releasing the anti-neoplastic agent selectively at the
tumor location. This means of administering drugs increases its active concentration locally
without the adverse effects of the systemic toxicity [24]. Concomitant arterial embolization of
hypervascular lesions slows blood flow. Thus, rapid clearance of the infused drug is diminished
[24]. An increase in the time of anti-cancer drug local action is obtained before its elimination [6,
16].
A first report by Kato [6] showed a significant increase of pain relief duration (Table 4) up to
6 months compared to palliative embolization where pain relief rarely exceeded 3 months.
Similar results were observed by Courtheoux [16]. However, a major bias is due to the fact that
most patients received chemoembolization and radiotherapy concomitantly.
Table 4 Review of the literature of chemoembolization of bone tumors
Duration
Number Patients
Number
Age
of followLiterature of
with bone
of
(years) up
patients metastases
infusions
(months)
Duration
of pain
relief
(months)
Kato (1981) 20
2
42–75 5–9
2–4
5–9
Courtheoux
12
(1985)
12
33–78 2–18
1–3
3–18
Chiras
(1998)
25
30–75 4–36
1–7
4–36
25
Radiological
response
(evaluable)
5/16 (30%)
In our series, metastases were unresectable and resistant to all other local or systemic therapies.
Chemoembolization was proposed in a palliative way to relieve symptoms. As previously
thought [12], we believe that chemoembolization not only reduced tumor bulk, but also killed
tumor cells.
In our study, teprubicin combined with PVA foam has been injected and carboplatin has been
infused during chemoembolization. To our knowledge, this association of anti-cancer drugs
currently has not been used in the treatment of bone metastases, but was chosen for the large
range of cancers with response expectancy to these drugs. Pain relief was obtained in 83% of
cases in our study.
Local pain increase during a 310-day period was observed in 30% of cases. However, we did not
encounter any post-embolization syndrome as has been reported previously by others [5, 8,
10, 15]. As a palliative treatment, it should not be expected to bring major morbid complications:
no general discomfort during or after the procedure was observed. Only one case of rectal
hemorrhage was observed. Nausea or vomiting because of chemotherapy drugs was prevented by
medical treatment during and after the procedures.
Conclusion
This study shows that transcatheter chemoembolization does not cause major discomfort and
appears to be more effective and reliable than embolization to provide pain relief. In some cases,
it showed objective tumoral reduction for bone metastases that were usually resistant to therapy
and could not be treated by local injection of cement.
References
1. Weill A, Chiras J, Simon JM, Rose M, Sola-Martinez T, Enkaoua E (1996) Spinal metastases:
indications for and results of percutaneous injection of acrylic surgical cement. Radiology 199:241–
247
2. Weill A, Kobeiter H, Chiras J (1998) Acetabulum malignancies: technique and impact on pain of
percutaneous injection of acrylic surgical cement. Eur Radiol 8:123–129
3. Feldman F, Casarella WJ, Dick HM, Hollander BA (1975) Selective intra-arterial embolization of bone
tumors. Am J Roentgenol 123:130–139
4. Wallace S, Granmayeh M, DeSantos LA, Murray JA, Romsdahl MM, Bracken RB, Jonsson K (1979)
Arterial occlusion of pelvic bone tumors. Cancer 43:322–328
5. Chuang VP, Wallace S, Swanson D, Zornoza J, Handel SF, Schwarten DA, Murray J (1979) Arterial
occlusion in the management of pain from metastatic renal carcinoma. Radiology 133:611–614
6. Kato T, Nemoto R, Mori H, Takahashi M, Harada M (1981) Arterial chemoembolization with mitomycin
C microcapsules in the treatment of primary or secondary carcinoma of the kidney, liver, bone and
intrapelvic organs. Cancer 48:674–680
7. Kato T, Nemoto R, Mori H, Takahashi M, Tamakawa Y, Harada M (1981) Arterial chemoembolization
with microencapsulated anticancer drug. JAMA 245:1123–1127
8. Soo CS, Wallace S, Chuang VP, Carrasco CH, Phillies G (1982) Lumbar artery embolization in
cancer patients. Radiology 145:655–659
9. Trèves R, Legoff JJ, Doyon D, Chasle G, Arnaud M, Jacob P, Burki F, Desproges-Gotteron R (1984)
L embolisation thérapeutique ou embolisation palliative à visée antalgique des métastases osseuses
d origine rénale . Rev Rhum 51:1–5
10. Varma J, Huben RP, Wajsman Z, Pontes JE (1984) Therapeutic embolization of pelvic metastasis of
renal cell carcinoma. J Urol 131:647–649
11. Braedel HU, Zwergel U, Knopp W (1984) Embolization of pelvic bone metastases from renal cell
carcinoma. Eur Urol 10:380–384
12. O Reilly GV, Kleefield J, Klein LA, Blume HW, Dubuisson D, Cosgrove GR (1989) Embolization of
solitary spinal metastases from renal cell carcinoma: alternative therapy for spinal cord or nerve root
compression. Surg Neurol 31:268–271
13. Coldwell DM (1989) Embolization of paraspinal masses. Cardiovasc Intervent Radiol 12:252–254
14. Chiras J, Cormier E, Meder JF, Rocher L, Weill A (1996) Artériographie et embolisation des tumeurs
du sacrum: technique, indications, résultats. Rachis 8:101–109
15. Barton PP, Waneck RE, Karnel FJ, Ritschl P, Kramer J, Lechner GL (1996) Embolization of bone
metastases. JVIR 7:81–88
16. Courtheoux P, Alachkar F, Casasco A, Adam Y, Derlon JM, Courtheoux F, L Hirondel JL (1985)
Chemoembolization of lumbar spine metastases. A preliminary study (French). J Neuroradiol 12:151–
162
17. Follezou JY, Fauchon F, Chiras J (1989) Intra-arterial infusion of carboplatin in the treatment of
malignant gliomas. A phase II study. Neoplasma 36:3
18. Vega J, Davila L, Chatellier G, Chiras J, Fauchon F, Cornu P, Capelle L, Poisson M, Delattre J (1992)
Treatment of malignant gliomas with surgery, intra-arterial chemotherapy with ACNU and radiation
therapy. J Neuro-oncol 13:131–135
19. Meder JF, Fredy D (1991) Embolisation des métastases du rachis mobile. Rev Rhum 58:5–10S
20. Sun S, Lang EV (1998) Bone metastases from renal cell carcinoma: preoperative embolization. J
Vasc Interv Radiol 9:263–269
21. Rossi C, Ricci S, Boriani S, Biagini R, Ruggieri P, De Cristofaro R, Roversi RA, Khalkhali I (1990)
Percutaneous transcatheter arterial embolization of bone and soft tissue tumors. Skel Radiol 19:555–
560
22. Nagata Y, Fujiwara K, Okajima K, Mitsumori M, Mizowaki T, Ohya N, Ohura K, Wataya S (1998)
Transcatheter arterial embolization for malignant osseous and soft tissue sarcomas. I. A rabbit
experimental model. Cardiovasc Intervent Radiol 21:205–207
23. Nagata Y, Mitsumori M, Okajima K, Mizowaki T, Fujiwara K, Sasai K, Nishimura Y, Hiroaka M, Abe M,
Shimizu K, Kohoura Y (1998) Transcatheter arterial embolization for malignant osseous and soft
tissue sarcomas. II. Clinical results. Cardiovasc Intervent Radiol 21:208–213
24. Coldwell DM, Stokes KR, Yakes WF (1994) Embolotherapy: agents, clinical applications and
techniques. RadioGraphics 14:623–643
25. Uemura A, Fujimoto H, Yasuda S, Osaka I, Goto N, Shinozaki M, Ito H (2001) Transcatheter arterial
embolization for bone metastases from hepatocellular carcinoma. Eur Radiol 11:1457–1462
26. Keller FS, Rösch J, Bird CB (1983) Percutaneous embolization with tissue adhesive. Radiology
147:21–27