Ethmoid sinus carcinomas: Results and prognosis after neoadjuvant chemotherapy and combined surgery—A 10-Year experience

Ethmoid sinus carcinomas: Results and prognosis after neoadjuvant chemotherapy and combined surgery—A 10-Year experience

98 NEOPLASM/SKULL Ethmoid Sinus Carcinomas: after Neoadjuvant Surgery-A F.X. Roux, M.D., D. Brasnu, G. Schwaab, S. Nguyen, M.D., and Combined E...

1MB Sizes 7 Downloads 74 Views

98

NEOPLASM/SKULL Ethmoid

Sinus Carcinomas:

after Neoadjuvant Surgery-A

F.X. Roux, M.D., D. Brasnu, G. Schwaab, S. Nguyen,

M.D.,

and Combined

Experience M.D.,

0. Laccourreye,

M.D., J. Bertrand,

0,

Results and Prognosis

Chemotherapy

lo-Year

BASE TUMORS

B. Devaux,

M.D., E. Chabardes,

M.D., M. Menard,

M.D.,

M.D., and J.F. Meder,

M.D.,

F. Janot,

M.D.,

M.D.

Departments of Neurosurgery and Neuroradiology, C.H. Ste Anne; Departments of Otolaryngology, Head and Neck Surgery and Anesthesiology, C.H. Laennec; Department of Otolaryngology, Head and Neck Surgery, Institut Gustave Roussy, Paris, France; Department of Otolaryngology, Centre Hospitalier General, Beauvais, France

Roux FX, Brasnu D, Devaux B, Chabardes E, Schwaab G, Laccourreye 0, Menard M, Janot F, Nguyen S, Bertrand J, Meder JF. Ethmoid sinus carcinomas: results and prognosis after neoadjuvant chemotherapy and combined surgery-a lo-year experience. Surg Neural 1994;42:98-104.

From June 1982 to June 1992, 144 ethmoido-sphenoidoorbital tumors have been referred to the neurosurgical department of Ste Anne Hospital. One hundred five of them were malignant lesions, among which 83 were included into our therapeutic protocol (1) neo-adjuvant chemotherapy (CDDP + 5-FU), (2) combined surgical procedure (subfrontal and transfacial), (3) postoperative radiotherapy. Fifty nine percent of the patients had no response to chemotherapy; 19% had a partial response (reduction of the tumoral volume >50% and
A&w reprint requests to: F.X. Roux, M.D., Department of Neurosurgery, C.H. Sainte Anne, 1 rue Cabanis, 75014 Paris, France. Received September 9, 1993; accepted November 11, 1993.

portunity of intraorbital exenteration, the indications, and the limits of combined surgery. We emphasize the importance of neo-adjuvant chemotherapy and of combined surgical procedures, even when the patients are complete responders to chemotherapy: complete responders who had only a transfacial approach have a 5-year actuarial survival rate of 80% (instead of 100% when a combined procedure was performed). Those who were not operated primarily recurred within 3 years and then had to be operated. We propose to follow such a combined surgery for all large ethmoidal cancers (T3 and T4) and for small tumors (Tl and T2) developed superiorly and posteriorly. Anterior Tl and T2 tumors should be operated through a single transfacial route. KEY WORDS:

Cranial base tumors; Adenocarcinomas; Chemotherapy; Combined surgery; Epidermoid carcinomas; Esthesioneuroblastomas; Ethmoidal cancers

The

treatment

sinuses

of malignant

improved

anesthesiology niques

during

last

and postoperative

allow large tumor

carcinologic

tumors the

surgical

Reconstruction

resections

paranasal

years.

intensive

removals,

of the skull

of the 25

Modern

care

tech-

and therefore

more

of cranial

basis cancers.

base was simplified

and is

now easier and more efficient than it was a few years ago. Neo-adjuvant chemotherapy is effective on certain cancers

of

this

region

and

has

now

a predominant

influence on the prognosis of such lesions. The efficacy of radiotherapy in such a disease was discussed by a number

of authors

for several years but is now consid-

ered as a useful postoperative treatment by a great majority of them. The aim of the present article is to present our surgical

experience

and to discuss

the indications

and

0090.3019/94/$7.00

Surg Neural 1994;42:98-104

Ethmoid Sinus Carcinomas

Table 2. Classification of Ethmoid Sinu Cancers

the results of combined surgical procedures in the treatment of cancer of the ethmoid sinuses, associated with preoperative chemotherapy.

Materials One hundred forty-four ethmoido-spheno-orbital tumors were treated in our department between June 1982 and June 1992; 105 were malignant tumors of the ethmoid and/or sphenoid sinuses. They concerned 16 women and 89 men aged 25 to 77 years (mean age: 55 years). Twenty-two patients were not operated because they were over 70 years and/or because the tumor was considered as nonresectable because of important local extension, and 83 underwent a combined surgical procedure. Histologic data are provided in Table 1. No international TNM classification exists for ethmoid tumors. Therefore we proposed several years ago Cl,141 the following classification, which is based on the only tumor extension in the ethmoid sinuses (Table 2). Tl:

the tumor is developed in only region. The oseous walls are intact.

T2: the tumor the ethmoid

one ethmoid

is developed in at least two regions of sinuses, with respect to their walls.

T3: the tumor destroys any ethmoid the cribriform plates.

cell wall, except

T4: the tumor destroys the cribriform plate(s), out objective intracranial extension, neither dural nor epidural. The dura is intact.

withintra-

T4IC: the cribriform plates are destroyed with intracranial extension, whatever its volume.

an

Sixty-five percent (n = 54) had a large tumoral extension (T4); 26.5% (n = 22) presented with a local recurrence when referred to our institution (Tables 3 and 4).

Therapeutic

Management

In all circumstances, the tially confirmed through apeutic schema included: (2) total removal of the

pathologic diagnosis was inian endonasal biopsy. The ther(1) neoadjuvant chemotherapy, tumor and the ethmoid block

Tl: tumor in a single ethmoid region with no wall destruction T2: tumor developed at least in two ethmoid regions, with no wall destruction T3: tumor with destruction of any ethmoid cell wall except the cribriform plate(s) T4: tumor with destruction of the cribriform plate(s), with no intracranial extension extension T4 IC : tumor with intracranial

through a combined radiotherapy. Induction

procedure,

and (3) postoperative

Chemotherapy

Induction chemotherapy was based in almost all cases on a 4-day continuous infusion of cisplatin (CDDP) 125 mg/mz/d] and 5 fluorouracil (5 F.U.) [l gr/m*/dl. Some patients had a slightly different chemotherapy protocol including the association CDPP + 5 FU with Bleomycin (n = 2), Adriamycine (n = 3), or Methotrexate (n = 1). Forty-seven patients had three cycles administered at 3-week intervals. Because of hematologic and/or cardiac intolerance, two patients underwent only one cycle, and three patients underwent only two cycles. When the tumor appeared to respond to chemotherapy, the number of cycles was increased so as to obtain a tumoral reduction as complete as possible: eight patients: four cycles, two patients: five cycles, one patient: six cycles, one patient: seven cycles (Table 5). Combined

Removal

Preoperative evaluation and surgical removal were always performed in the department of neurosurgery. The preoperative radiologic findings were based on computed tomography (CT) scan horizontal and coronal views (for all patients), and magnetic resonance imaging (MRI) views (39 patients). Bilateral carotid angiography was performed 74 times. We have now decided to perform angiography only for unusual tumors but not for adenocarcinomas, esthesioneuroblastomas, or epidermoid carcinomas, because in our opinion, it brings no information of practical value. Table 3. 83 Operated Ethmoid Cancws: Classtjication

Table 1. 83 Operated Ethnoid Sinus Canrevs: Histologic Data Adenocarcinomas Esthesioneuroblastomas Squamous cell carcinomas Undifferentiated Carcinomas Melanomas Sarcomas Plasmocytomas Adenoid Cystic Carcinomas Ameloblastoma

99

63 8 1 1 : 2 2 1

First Treatment

RKUtE%W2

Total

Tl T2 T3

5 4 13

1 1 5

6 5 18

T4 T4 IC

27 12

5 10

32 22

n = 61

22

83

100

Surg Neural

Roux et al

1994;42:98-104

Table 4. 63 Operated Adenocavcinomas: Classifiation First Treatment

Recurrence

Table

6. Responseto Neo-adjuvant Chemotherapy

Total TR 100%

Tl T2 T3 T4 T4 IC

4 4 9 21

1 1 3 5

5 5 12 26

9

6

15

n = 47

16

63

The surgical technique was detailed in previous publications [14-161. It includes four steps: a subfrontal approach, a transfacial approach, en bloc removal of the tumor, and the cranial basis reconstruction. Postoperative radiotherapy was administered 21 times (average dose administered: 65 grays). Preoperative radiation therapy was never used for first-operation patients.

Adenocarcinomas Esthesioneuroblastomas Adenoid cystic carcinomas Squamous cell carcinomas Undifferentiated carcinoma Melanoma Ameloblastoma

In this series, we considered only the patients (n = 83) operated upon up to June 1992. The minimal follow-up was 6 months for all the patients.

Chemosensitivity The response to neo-adjuvant chemotherapy was visualized on CT scan and MRI. Three groups were determined (Table 6): no response or response under 50% of the initial tumor volume partial response: the initial volume of the lesion has decreased significantly but incompletely (> 50% and
Table

5. Neo-adjuvant Chemotherapy: Numbs of Cych Patients

Cycles

2 3

1 2

478 2 1 1 N = 64

: 5 6 7

8 1

2

1

1

1

1 1 1

1 1 1

N=64

Total Abbreviations;

14

TR<50% NR

12

34 3 1

12

38

TR, tumor response; NR, nonresponder.

(no visible tumor) and nants on the pathological Surgical

Results

54 4

TR > 50%

histologically smears).

(no tumor

rem-

Results

Mortality. one patient died from a pulmonary embolism 3 weeks postoperatively one died 5 months after surgery from a spontaneous intracerebral hematoma, without any relation to his tumor. one died 4 years after surgery due to a pulmonary embolism that appeared after a cosmetic operation (facial retractile scar); the ENT control was normal. These three patients had adenocarcinoma. Morbidity. one patient had an immediate and transient postoperative rhinorrhea: it lasted 2 days and was followed by a meningitis (acinetobacter) that was cured in 3 days. The rhinorrhea ceased spontaneously. four other patients presented with an immediate postoperative meningitis (pneumococcus: 2, acinetobacter: 1, pyocyanic: l), but had no macroscopic rhinorrhea. These cases with meningitis were considered to be due to a direct intraoperative inoculation. Three were cured without any sequellae; one has severe psychointellectual impairment. ?? one patient had subcutaneous infection; four had an close to the cranial base endonasal suppuration “plasty” (coral block: 2; autograft: 2). These infections were cured after a subcutaneous drainage for the first case, and after the removal of the graft (through the endonasal route) for the four other patients. Oncological Resdts We will consider the results of the only adenocarcinomas because they represent a large enough homogeneous

Ethmoid

Surg Neural

Sinus Carcinomas

101

1994;42:98-104

Adonocarcinomas

N = 63

1.0000

0.6613

Figure 2. Actuarial suwival rates of ethmoid sinus adenocarcinomas presenting with, {rec.} (n = 16) m without a recurrence, {lair) (n = 47).

0.3199

0.0000

Figure

_ -12

>months ‘

12

,*

26

IO

I6

62

66

5,

‘0

66

1. Adenocarcinomas (n = 63): overall actuarial survival rate.

population (n = 63) leading to significant statistical figures. As previously noted three patients died from pathology independent of their tumor. Nineteen patients (31.5%) died from their cancer with a mean survival time of 19 months (1 to 48 months). The duration of survival between the tumor recurrence and/or the metastatic dissemination, and the death ranged from 1 month to 27 months (average: 9 months). Six patients are still alive with a recurrence. Forty patients were free of disease with a follow-up ranging from 6 to 88 months: nine have a follow-up of more than 3 years, five of more than 5 years. The overall actuarial survival rate of adenocarcinomas was 85% at 1 year, 70% at 2 years; 53% at 3 years, and 42.5% at 5 years (Figure 1). Surprisingly, there was no significant difference in the survival rates of the patients who were. included in our therapeutic protocol with (n = ‘16) or without (n = 47) a recurrence (Figure 2).

The survival rate was inversely proportional to the volume of the tumor (Figure 3). The most significant figure results concern intracranial extensions (T4IC): the actuarial survival rate of patients with T4IC tumor was about 19% at 3 years, and no patient survived after 4 years (Figure 4). Lastly, induction chemotherapy influenced the prognosis (Figure 5): the actuarial survival rate of the patients with a complete response stayed at 100% at 3 and 5 years as well as 10 years. Concerning the influence of radiotherapy, it was not possible to determine significant figures because of the small number of patients having followed such a treatment.

Discussion The advantages of combined surgical procedures (subfrontal and transfacial) have been emphasized by many authors dealing with cancers of the ethmoid f2,7,14, 16-191. The surgical goal is to perform a complete tumor resection in order to avoid as far as possible intracranial recurrences. Indeed, none of our patients presented such a recurrence. The attitude concerning intraorbital extension is not yet definitively defined {lo}. Ketcham et al (1973) [7] and Schramm et al (1979) El71 demonstrated that the survival rate was much higher if an orbital exenteration was performed (60% versus 30% at 5 years) when there was evidence of intraorbital extension. In fact, data have been modified recently due to the development of preoperative chemotherapy and postoperative radiotherapy 16,8,9,121. Kraus et al IlO} suggested that lamina papyracea erosion without periorbital invasion permits

Surg Neural 1994;42:98-104

Roux et al

T

+---t

-+-+-+--b--t-+_+_+_+_+_+__

).-.. 9-' -.

‘1

t-

lOOI

“ZB

.sa\n=12 0 w
i. ',' ?

‘, ‘\

.b ‘, ,\ ‘. \

:

‘,

;

‘\ ‘,

‘,

‘.

‘1 . ‘)

‘.

b...,

.-‘I

0.0000

~ -II

Figure 3. Actuarial survival rate of ethmoidal adenocarcinomasdepending on the tumoral extension. Tl: n = 5; T2: n = 5, T3: n = 12; T4: n = 41

(T4 + T4 KY).

preservation of orbital content. In our opinion an orbital exenteration is indicated only for anterior orbital invasion. In case of extension to the posterior orbit, the resection is often impossible because of the invasion of the sphenoid and dura mater, and the resection is microscopically not complete. In our series, only one patient underwent such an exenteration; he was a complete responder to chemotherapy and therefore even if he is still free of disease with a 3-year follow-up, it is not

t

0.0000

,

_

-,I



II

18

II

10

lb

42

48

54

60

months

‘6

Figure 4. Actuarial survival rate for patients presenting with an intracranial extension (T4 IC ti = 1 S) WWJ patie& without any intracranial extension (no IC: Tl + T2 + T3 + T4, n = 48).

_

! 0

6

LI

*II

24

Xl

36

42

48

54

‘0

9

months

‘6

Figure 5. Ethmoidal adenocarcinomas: actuarial survival rate depending on the responseto neo-adjuvant chemotherapy. TR = 100%: n = 8; TR = > 60%: n = 12; TR = < 50%: n = 34; (TR): tumor response).

significant. Practically speaking, the few patients to whom we proposed an exenteration refused such an operation. On the other hand, we have the experience of several patients with intraorbital extension in whom the tumor disappeared after chemotherapy, even if it was an esthesioneuroblastoma 112). None of these patients who were responders to chemotherapy presented with a local recurrence. We therefore value such chemotherapy and do not even propose to the patient an orbital exenteration. Lastly, if a total removal was not possible because of a persistent intraorbital extension, postoperative radiotherapy was given. A posterior extension of the tumor can be a limit to the indication of a surgical removal. Indeed we consider that a total cancer removal, as well as the reconstruction of the cranial basis and of the dura mater are not possible if the tumor has extended into the sphenoid sinus and if it invades the dura posterior to the optic canals. Therefore, a tumor extension in the jugum and/or the clivus are contraindications to surgery. On the contrary, if only the anterior wall of the sphenoidal sinus is invaded, the removal must be as large and complete as possible. It is most important to have before surgery a good appreciation of the posterior extension. MRI is the best study for such a purpose: T2 views show the difference between a tumor extension and a reaction with retention of fluid in the sphenoid sinus, which is not demonstrated by CT scan. MRI can also show a dural infiltration of the cranial basis, which is, for us, a contraindication to surgery if it extends too posteriorly (Figures 6 and 7).

Ethmoid

Surg Neural 1994:42:98-104

Sinus Carcinomas

103

Figure 7. Magnetic resonance imaging can demonstrate the infiltration of the dura (arrow).

Figure 6. Magnetic resonance imaging views make the dijJf@encebetweena sphenoid retention (arrow) and a tumoral extension (T).

Once these limits are defined, an intracranial and subdural extension itself is not a contraindication to surgery whatever the volume of the tumor even if it is developed in the subfrontal parenchyma; technically an intracerebral tumor is indeed a usual problem with which neurosurgeon has to deal. It has now been proved that induction chemotherapy is efficient on a significant percentage of epithelial cancers (adenocarcinomas, epidermoid carcinomas) of the paranasal sinuses and esthesioneuroblastomas 14,6,9,12,16}. In our experience, the survival of the responders to chemotherapy is longer than that of the nonresponsive ones {4,8,11]. This point is particularly demonstrated in those who have a complete response, because they have a 5-year and a lo-year actuarial survival rate of 100% 1161. Concerning these patients, a question must be raised: is it worth operating on them through a combined route as, at the time of surgery, no more tumor is visible clinically and radiologically? We consider that the answer is positive because the complete responders who have been operated only through a transfacial procedure have a lower survival rate (80% at 5 years), and those who have not been operated on, have all presented a recurrence within 2 or 3 years 191. Such a recurrence is probably due to the persistence of cell residuals resistant to chemotherapy which are not detectable even on pathologic examination. The only way to take these residuals away is to perform a large removal of all the ethmoid block. We want to emphasize that

there is no report in the literature of a randomized study of neo-adjuvant chemotherapy in tumors of the ethmoid sinuses. Only retrospective studies have been published. The role of induction chemotherapy in term of survival has still to be defined and no firma1 conclusion can therefore be drawn, Postoperative radiotherapy is also an important factor of prognosis, and its indications seem to widen and to be more often proposed by different authors [6,9-j. None of our patients presented a radionecrosis of the cranial base graft.

Conclusion Though the benefit of neo-adjuvant chemotherapy in carcinoma of the paranasal sinuses requires further investigation with randomized trials, we consider that induction chemotherapy should be administered for almost all cancers of the paranasal sinuses and that postoperative radiotherapy is relevant particularly if surgical removal has not been complete. Concerning the surgical procedure, all ethmoid cancers do not necessitate a combined approach. Such a large removal should be indicated for all large tumors, i.e., T3 and T4, and for small tumors, Tl and T2, developed in the posterior and superior ethmoid. Tl and T2 tumors of the anterior ethmoid should be operated through a single transfacial route-combined surgery being performed only in case of a posterior recurrence.

References 1. Brasnu D, Roux FX, Fabre A, Menard M, Manolopoulos L, Chodkiewicz JP, Laccourreye H. La voie d’abord combinCe ORL et Neurochirurgicale dam les adCnocarcinomes de l’ethmoide. Ann Otolaryngol Chir Cervico-Fat 1987;104:347-51.

104

Surg Neural 1994;42:98-104

Roux et al

2. Chessman AD, Lund VJ, Howard DJ. Craniofacial tumors of the nasal cavity and paranasal sinuses. Surg 1986;8:429-35.

resection of Head Neck

11. Norris CM, Clark RJ, Frei E, Erwin TY. et al. Pathology of surgery after induction chemotherapy. An analysis of resectability and loco-regional control. Laryngoscope 1986;96:292-302.

Possibilites d’exerese 3. Derome P. Les tumeurs spheno-orbitaires. et de reparations chirurgicales. Neurochirurgie 1972;18 (suppl 1):164.

12. Polonowski JM, Brasnu D, Roux FX, Bassot V. Esthesioneuroblastoma. Complete response after induction chemotherapy Ear Nose Throat J 1990;69:743-6.

4. Fabre A, Saliba N, Menard M, Basset V, De Stebenrath A, Manolopoulos L, Brasnu D, Laccourreye H. Chimiotherapie d’induction dans les adenocarcinomes de l’ethmoide. In: Actualit& de carcinologie cervico-faciale. Paris: Masson, 1986: 17682.

of the 13. Robins PE, Powell DJ, Stansbis JM. Review: carcinoma nasal cavity and paranasal sinuses. Incidence and presentation of different histological types. Clin Otolaryngol 1979;4:431-56.

RT, Fitz-Hugh GS, Constable 5. Jackson plasms of the nasal cavities and paranasal ive study. Laryngoscope 1977;87:726-36.

WC. Malignant neosinuses. A retrospect-

of parana6. Johns ME, Kaplan MJ. Advances in the management sal sinus tumors. In: Gregory T, ed., Head and neck oncology. Boston: Martinus Nijhoff Publishers, 1984:25-52. 7

Ketcham AS, Chretien PB, Van Buren JP. The ethmoid sinuses: a re-evaluation of surgical resection. Am J Surg 1973;126:46976.

8

Kish JA, Ensley JF, Jacobs J, Weaver A, Cummings G, Alsarraf M. A randomized trial of cisplatin and 5-Fluorouracil infusion and CACP + .5 FU bolus for recurrent and advanced cell carcinoma of the head and neck. Cancer 1985;56:2740-4.

9

10

Londero A, Laccourreye 0, Chabardes E, Roux FX, Brasnu D, Laccourreye H. Tumeurs malignes de I’ethmoide. A propos d’une serie de 140 cas. Actualites carcinologie cervico-faciale, # 15, 43-53, Paris: Masson, 1991. Kraus DH, Sterman BM, Levine HL, Wood BG, Tucker HM, Lavertu P. Factors influencing survival in ethmoid sinus cancers. Arch Otolaryngol Head Neck Surg 1992;118:367-72.

14. Roux FX, Brasnu D, Menard M, Bassot V, Nguyen S, Donadieu S, Chodkiewicz JP. Adenocarcinomas of the ethmoide sinuses. Results of a new protocol based on inductive chemotherapy combined with surgery. Acta Neurochir 1989;98:129-34. M, Janot F, Laccourreye H. 15. Roux FX, Brasnu D, Menard Adenocarcinomas of the ethmoidal sinuses treated by induction chemotherapy ad combined surgical approach. Head and neck cancer. Vol 2. Toronto. Philadelnhia. B.C. Decker Inc.. 1990:384-7. 16. Roux FX, Brasnu D, Menard M, Schwaab G, Janot F, Lacau Saint Guily J, Trotoux J, Laccourreye 0. Les abords combines des tumeurs malignes de I’ethmdide et autres sinus paranasaux. Principes et resultats. Ann Otolaryngol Chir Cervico-Fat 1991;108:292-7. 17. Schramm VL, Myers EN, Maroon JC. Anterior gery for benign and malignant disease. 1979;89:1077-91.

skull base surLaryngoscope

du traitement chirurgi18. Schwaab G, Marandas P. Les probltmes cal des tumeurs malignes de I’ethmoide par abord mixte endocranien et facial. Ann Otolaryngol 1983;100:159-61. HH. Craniofacial resection for malignant 19. Shah JP, Galicich tumors of the ethmoid and anterior skull base. Arch Otolaryngo1 1977;103:514-7.