Home
  Maîttrise Education
   Contact
   French
 
                      Articles      Interviews      Meetings      Links     Medline   
     
     
   

 

S'abonnez à Maîtrise Orthopédique

 

header
     
   
footer
Télécharger le bulletin au format PDF

TREATMENT OPTIONS FOR GIANT-CELL TUMOR
B. Tomeno
Article Summary

Why write an article specifically on GCT (giant-cell tumor) ?
INTRODUCTION
BENIGN GCT OF LIMBS
Reasons for suspecting the presence of a GCT
Besides plain radiogaphs, what other investigations are appropriate ?
Is biopsy a requirement ?
Appropriate method for an effective GCT curettage
What about the disadvantages of cement ?
Apart from cement, what are the other filling options available ?
How to manage local recurrent GCT ?
What else about benign GCT ?
MALIGNANT GCT OF LIMBS
GCT OCCURRING IN THE TRUNK REGION

 

Hôpital Cochin - Paris, France

 

Why write an article specifically on GCT (giant-cell tumor) ?

This decision has been carefully thought through : as a giant-cell tumor (GCT) can be benign or malignant, an in-depth study of GCT provides the essence of what one needs to know about tumors and their variants, as well as the technical tricks regarding curettage & bone-grafting which is the classical treatment for all benign intraosseous tumors. An overview of the malignant forms of GCT gives an idea of the resection-reconstruction techniques and associated therapies (radiation therapy and chemotherapy). GCT is an ideal "model" for educational purposes.

 

INTRODUCTION

Giant cell tumor (GCT) is a frequent occurrence. It accounts for 15 to 20% of all bone tumors. However, its histogenesis remains unclear (it is thought to originate from connective tissue precursor cells). GCT is mostly found in the limbs (90%) and sometimes in the trunk. The giant cell tumors described in this study involve essentially the pelvis, sacrum, and in rare cases the vertebral bodies. In long bones, GCT occurs almost exclusively in the end portion of the bone, next to the joint (epiphysis-metaphysis)+++, whereas most bone tumors occur in the flared portion of the bone (metaphysis). It is most frequently seen around the knee joint (60%), in the upper end of the humerus (15%), and in the lower end of the radius (10%) or the tibia (10%).

It generally occurs in young adults aged 20 to 40 (75%). Clinical signs are unremarkable (pain, sometimes a mass, rarely a pathologic fracture). Importantly, giant cell tumor is benign in 90% of the cases. Malignant giant cell tumors are cancerous immediately (5%), or may be benign tumors that progressed to malignancy (5%).

Histologically, anatomopathologists distinguish three grades :

  • grade 1 : benign, low aggressive behavior
  • grade 2 : benign, highly aggressive behavior
  • grade 3 : malignant

Curettage & bone-grafting is the classical treatment for benign tumors, but it is associated with a high local recurrence rate (20 to 40%).

 

BENIGN GCT OF LIMBS

Reasons for suspecting the presence of a GCT

The presence of a GCT in a young adult should be suspected whenever there is osteolysis in the metaepiphyseal region of a long bone. Few tumors involve the epiphysis, most are found in the metaphysis. Only three types of tumor commonly involve the epiphysis : GCT, clear cell chondrosarcoma, and chondroblastoma. In this study group, GCT is the most prevalent by far (Fig. 1).

 

Figure 1 : The "Big Three" epiphyseal GCTs. GCT (left), chondroblastoma (center), clear cell chondrosarcoma (right).
TCG = GCT
Chondroblastome = Chondroblastoma
Chondrosarcome à .... = clear cell chondrosarcoma

 

Radiographically, GCT can be classified in two main types (Fig. 2) :

 

Figure 2 : Characteristic appearance of a multiloculated lesion (honeycomb-like pattern) (left) ; pure lytic lesion (right).

 

  • Multiloculated lesion with a very characteristic appearance (honeycomb-like pattern) (30% of the cases)
  • Pure lytic lesion (70% of the cases).

Also, it is important to differentiate between :

  • Nonprogressive forms (x-rays being taken a few weeks later and showing little or no changes) 
  • Aggressive forms in which radiographic changes are obvious and even impressive : a real explosive growth of the tumor (Fig. 3).

 

Figure 3 : Aggressive GCT : obvious radiographic changes on the right x-ray taken one month after the left one.
1 mois d'intervalle = One month

 

Nonprogressive, multiloculated giant cell tumors (actually, this apparent multiloculation is caused by endosteal scalloping creating ridges : Fig. 4) are hardly ever malignant or potentially malignant. Therefore, it is in pure lytic lesions and particularly in radiographic aggressive forms that malignancy may be encountered.

 

Figure 4 : CT scan clearly shows that this apparent multiloculation is caused by endosteal scalloping creating ridges.

 

Besides plain radiogaphs, what other investigations are appropriate ?

CT scan or MRI can be very helpful (both are not really necessary for GCT of limbs), particularly where soft-tissue or intra-articular extension is suspected.

Bone scintigraphy does not provide any valuable information (multifocal lesions are too exceptional to justify routine bone scintigraphy ; furthermore, it fails to distinguish between benign and malignant lesions).

A chest x-ray can be of value since GCT has the potential for metastasis to the lungs :

  • it is obvious in malignant GCT
  • metastases may also be seen in some benign lesions (in this case, the lung lesions will also be histologically benign, which is a curious paradox). This occurs in only 1 to 2% of giant cell tumors, and exclusively in the lytic, aggressive forms, never in the multiloculated, nonprogressive forms.

A phosphocalcic assessment is helpful to confirm the absence of bone lesion of hyperparathyroidism which may simulate (even histologically !) GCT.

Is biopsy a requirement ?

The answer is : definitely, YES. However, the working environment is an important factor that must be taken into account :

  1. If the medical center is close to a good laboratory of anatomic pathology with a staff that is familiar with bone tumor pathology, one may consider performing all the tests under one single anesthesia. An extemporaneous examination is performed at the beginning of the procedure. If it confirms that the lesion is benign, one can carry on with the procedure, creating a window (in most cases) to expose the entire tumor and proceed to curettage & bone-grafting. If it does not positively confirm the benignity of the lesion, it is wiser to close the wound and postpone treatment until after the results of conventional histology are available.
  2. If extemporaneous biopsy cannot be performed in good conditions, a two-stage procedure is recommended : first procedure for the biopsy, second procedure performed two weeks later for the surgical treatment.

Now, regarding biopsy, it is important to point out that :

  1. Biopsy of a GCT does not involve "removing a small tissue sample of the tumor". A large-size biopsy specimen (almost the entire contents of the cavity) is necessary, as it can be considered as the first step of the curettage/bone-grafting procedure. As a matter of fact, malignant transformation of a benign giant-cell tumor is not uniform ; in other words, a GCT may have both benign and malignant areas (Fig. 5), and the latter must not be missed.
  2. As long as benignity of the lesion is not confirmed, a 2 cm incision (if the affected bone is close to the surface of the skin) is sufficient for the biopsy. This small incision can subsequently be extended for curettage ; in case of malignancy, the involved bone segment and the biopsy scar will have to be removed en bloc, which will be all the easier if the prior incision was very small. 
  3. Recurrence of an initially benign GCT must not dispense with a new biopsy prior to treating the recurrent tumor, because one can never be sure that it is also benign.
Figure 5 : A GCT may have both benign and malignant areas. A large-size biopsy specimen is required to avoid missing any malignant areas.
Bénin = Benign
Malin = Malignant
Bénin + Malin = Benign + Malignant
Petite biopsie = PIEGE = Small biopsy specimen = PITFALL
Biopsie large = SECURITE = Large biopsy specimen = SAFETY

 

Appropriate method for an effective GCT curettage

Curettage is a meticulous and rigorous procedure which determines the potential for recurrence (the term "recurrence" is very controversial : if GCT recurs, it is because microscopic disease has been left behind, which means that in fact, it is not a "recurrence" but a "persistence" of the tumor).

A correct curettage is performed with a large curette to remove macroscopically the main part of the tumor. Grossly, the curettage material is soft and dark brown (very characteristic Buff color) (Fig. 6) ; some more yellow areas (xanthelasma lesions, a fatty involution), or much whiter and fibrous areas (therefore suspected of malignancy) may be seen.

 

Figure 6 : Characteristic Buff color.

 

Then, one must progressively decrease the size of the curette to the smallest size (so-called "grain de mil") and patiently scrape the tumor remnants off the bone walls, paying particular attention to crevices, tiny recesses and pockets, fissures ....

Now, there is the option to use local adjuncts. Curettage alone, as described above, is associated with a high local recurrence rate of 30 to 40% ; when combined with local adjuncts, this rate is decreased by 50%. Optional adjuncts include :

  • Liquid nitrogen : efficient, but difficult to manipulate (sight is obscured by the vapors) and hazardous : 30% local complications due to frostbite of the soft tissues (superficial or deep), bone or cartilage which become friable and brittle. Many surgeons have given it up.
  • Phenol : according to French pharmacological regulations, only low phenol concentrations (too low to be efficient) can be used.
  • Bone cement : it is both a very good adjunct (only 15% of local recurrence) and a filling material. However, it also has a few disadvantages (we shall revert to this later). 
  • High-speed burring of the cavity : it allows complete destruction of the cavity walls down to the healthy cancellous bone (but it also carries the risk of tumor seeding). 
  • Carbonization of walls : it is achieved with an electrocautery and a ball-tipped probe, using high current intensity (Fig. 7). The color of the walls turns black, a second curettage is performed, followed by jet lavage. The procedure is repeated 2 or 3 times. It is the method we are currently using.

 

Figure 7 : Carbonization of walls : achieved with an electrocautery and a ball-tipped probe, using high current intensity.

 

What about the disadvantages of cement ?

Disadvantages of cement are both theoretical and practical.

1) Theoretical disadvantages : as GCT normally affects young adults, it is not unreasonable to think that these patients may require internal fixation of a fracture or replacement of an arthritic joint at some time or another, and that cement may be a problem and even compromise healing of a fracture.

2) Practical disadvantages : actually, we have had a few cases in which, unfortunately, the tumor extended to the subchondral bone so that the block of bone cement was almost flush with the joint line. Such a thin osteocartilaginous shell may wear out or fracture (Figs 8, 9, 10).

 

Figure 8: Cement disadvantages : a few months after cement packing of this GCT of the distal radius, the thin osteocartilaginous shell had worn out. Pain with mobilization made arthrodesis necessary (however, and we insist on that : free motion of the mediocarpal joint was preserved, which allowed 30-40 degrees of flexion-extension).
Figure 9: Cement disadvantages : a few months after cement packing of this GCT, the thin bony bridge in the intercondylar notch broke and the residual medial condyle moved medially. At reoperation, destruction of the distal femur was such that we had no other option than to implant a knee prosthesis.
Figure 10: Cement disadvantages : C = cement block, A = window created for curettage-filling, B = subchondral bone disruption. The remaining portion of the tibial plateau (between A and B) slides down over the cement block, leading to mechanical failure.

Although we do not reject this technique, we carefully place our indications and use it only in the following cases :

  • in the elderly. 
  • in multiply operated patients who have previously undergone multiple bone graft harvests. 
  • GCT infection secondary to biopsy (advantage of using antibiotic bone cement). 
  • in cases where there is no subchondral extension of the tumor (with at least 1 cm of healthy bone left between the tumor and the joint line+++).

Apart from cement, what are the other filling options available ?

Actually, in terms of functional results, there is no significant difference between GCT filled with autologous bone grafts and with morsellized femoral head bone grafts from the bone bank. Radiographically, allograft bone is generally whiter and qualitatively denser (but it is no more than a radiographic curiosity) (Fig. 11). Of course, in a large GCT, a mix of autograft and allograft can be used (Fig. 12).

 

Figure 11: Allograft bone is dense. Excellent functional outcome.
Figure 12: Combination of several filling materials : iliac autograft fragment positioned horizontally flush with the joint line with the cancellous bone facing up + cancellous bone obtained from the bone bank placed below the iliac graft + supportive internal fixation device.
Greffon iliaque = Iliac graft
Tête de banque broyée = Morsellized femoral head

 

As regards bone substitutes, our experience with this type of filling is rather limited and not very convincing. On the other hand, femoral heads are so readily available that we hardly ever use bone substitutes.

What is important, independent of the nature of the graft, is that the cavity is packed densely with graft material using a graft pusher. Should it not seem to provide adequate primary stability and eliminate the need for plaster immobilization and relief from weight bearing, it can be enhanced (if desired) with : 

  • cortical bone blocks 
  • supportive internal fixation device (Fig. 12). Devices such as the hook plate are particularly well suited to tumors of the proximal humerus that leave a very thin healthy portion of the humeral head which is unable to accommodate fixation screws. The hook has two functions : it contains the bone grafts and supports the residual cancellous bone of the humeral head (Fig. 13). 

 

Figure 13: Molded hook plate. The hook supports the residual cancellous bone of the humeral head ; only the longitudinal part of the plate that rests on the bone surface is fixed with screws.
Tumeur =Tumor
Greffons = Bone grafts
Plaque moulée = Molded fixation plate

 

How to manage local recurrent GCT ?

The most important thing is detection : radiographic follow-up at 6-month intervals for 2 years (80% of local recurrences manifest themselves within 2 years), and then once a year for 10 years (if possible) ; MRI or CT scan is performed only if the x-rays are equivocal ; bone scintigraphy may be of value if it shows increased uptake on a sequence of bone scintigrams. It may be difficult to differentiate between bone graft remodeling and development of recurrence. In this situation, clinical evaluation is interesting : if there is a radiological concern but the patient feels all right, the likelihood of recurrence is very low.

Are there factors influencing the local recurrence of benign GCT ?

Histological grade 1 or 2 has no influence. In contrast, the recurrence rate in lytic lesions with an aggressive radiographic appearance or extending into soft tissues is twice that in other forms of GCT. Lastly, recurrences seem to be more frequent in the lower end of the radius, sacrum and spine.

As previously said, if GCT recurs, biopsy is mandatory to make sure this recurrence is benign. Should this be the case, one is faced with a big dilemma : perform an iterative curettage & bone-grafting or a resection-reconstruction.

As far as we are concerned, we are clearly in favour of iterative curettage & bone-grafting. As a matter of fact, in 1986, we took part in a European study involving 600 cases which demonstrated that 8 or 9 out of 10 patients could be successfully managed using this method, even if curettage & bone-grafting had to be repeated 2, 3, or even 4 times.

We reserve resection for cases of malignant transformation of a benign GCT, or where local destruction is such that curettage is not mechanically feasible (Figs 14 & 15). 

 

Figure 14: Huge, multiply recurrent GCT of the knee ; extensor mechanism is completely destroyed. Reconstruction with Juvara type arthrodesis.
Figure 15: Multiply recurrent GCT of the femoral neck. Although recurrences were benign, mechanically, there was no other option than to implant a bulky prosthesis.

 

Histological grade 1 or 2 has no influence. In contrast, the recurrence rate in lytic lesions with an aggressive radiographic appearance or extending into soft tissues is twice that in other forms of GCT. Lastly, recurrences seem to be more frequent in the lower end of the radius, sacrum and spine.

As previously said, if GCT recurs, biopsy is mandatory to make sure this recurrence is benign. Should this be the case, one is faced with a big dilemma : perform an iterative curettage & bone-grafting or a resection-reconstruction.

As far as we are concerned, we are clearly in favour of iterative curettage & bone-grafting. As a matter of fact, in 1986, we took part in a European study involving 600 cases which demonstrated that 8 or 9 out of 10 patients could be successfully managed using this method, even if curettage & bone-grafting had to be repeated 2, 3, or even 4 times.

We reserve resection for cases of malignant transformation of a benign GCT, or where local destruction is such that curettage is not mechanically feasible (Figs 14 & 15).

 

Figure 16: a) Primary aggressive GCT of the tibia. Management with curettage & bone-grafting does not seem a reasonable option.
b) Implantation of a sliding prosthesis over a massive allograft (bone bank).
Figure 17: Huge GCT of the wrist that has been progressing (and has remained untreated) for 9 years. Resection-arthrodesis was the only option, but free motion of the mediocarpal joint was preserved, which allowed some degree of flexion-extension.

 

What else about benign GCT ?

Multifocal GCTs of bone are rare and account for 1 to 2% of all cases of GCT of bone. Multiple lesions may be discovered synchronously or metachronously (Fig. 18). Interpretation is difficult : is it a true multifocal lesion or a metastasizing GCT ?

 

Figure 18: Multifocal GCT. a) Primary GCT of the proximal tibia in 1995, successfully managed with curettage & bone-grafting.
b) GCT of the carpus detected in 1996-1998 (several recurrences).
c) Sacral GCT detected in 2000.

 

ABC (aneurysmal bone cyst) components in GCT are not uncommon. We have even had patients with recurrent lesions being sometimes a GCT and sometimes an ABC. There is likely some sort of relationship between GCT and ABC (Fig. 19).

 

Figure 19: Combined GCT and ABC (aneurysmal bone cyst).
TCG = GCT
KA = ABC

 

Histological grade 1 or 2 has no influence whatsoever on therapeutic indications and recurrences. Grade 3 only (malignant) needs to be managed with utmost care.

Radiation therapy should be avoided, be it only for the potential risk of malignant transformation of a benign GCT (30% of the cases). Furthermore, there are no indications for the limbs. 

 Our Tunisian colleagues treated GCT with curettage and calcitonin (no bone-grafting), and demonstrated that calcitonin had the potential to promote new bone formation in the tumor cavity. Patients received intratumoral injections of calcitonin for 2 to 3 weeks, and then subcutaneous injections for a few months. The main drawback of this method is that the patient is kept off weight bearing for several months, until the defect is healed. The promoters of this method suggest that calcitonin further has the ability to reduce the potential for recurrence. However, during the last SOFCOT meeting, they reported on a series of 25 patients, 8 of whom had recurrences ..... These results are not significantly different from those in other series.

GCT is exceptionally seen in children under the age of 15. Contrary to GCT in adults, GCT in children involves exclusively the metaphysis.

 

MALIGNANT GCT OF LIMBS

We have previously had a quick skim of the subject (frequency, primary and secondary forms of GCT, predisposing or influential factors, etc ...).

Histologically, malignancy (Grade 3 GCT) is generally an osteosarcoma or a fibrosarcoma ; the so-called "giant cell sarcoma" is not a synonym for "malignant giant cell tumor of bone", it is just one of the numerous histological subgroups, a specific form of malignant GCT. Actually, these nuances have little influence on the treatment strategy.

Once malignancy has been recognized, the malignant GCT is usually treated as an osteosarcoma, which means :

  • biopsy
  • chemotherapy (for osteosarcoma) for 3 to 4 months
  • wide resection + reconstruction with a bulky prosthesis (Fig. 20) in preference to resection + arthrodesis (rarely used) (Fig. 21). Exceptionally, amputation is mandatory in huge or infected GCT
  • second chemotherapy for a few months. 

 

Figure 20: Bulky Guepar knee prosthesis for treatment of malignant recurrence of an initially benign GCT.
Figure 21 : Malignant GCT of the proximal tibia with suspected intra-articular extension. Resection + Juvara type arthrodesis. Good result at 20-year follow-up.

 

Chemotherapy has improved the formerly poor prognosis of malignant GCT from a 30% to a 60-70% success rate. Patient death is generally related to progressive pulmonary metastases.

 

GCT OCCURRING IN THE TRUNK REGION

GCT occurs more rarely in the trunk region (10-15%) than in the limbs, and generally involves the sacrum (5-8%), pelvis (4-5%) and more especially the iliac wing and ischium (Fig. 22), more rarely the vertebral bodies (1-2%).

 

Figure 22: Small GCT of the ischium with a few ABC areas.

 

Recurrence and malignancy are much more frequent than in the limbs : 40 to 45% in each of the above !

Thorough radiological investigations are mandatory and should include : CT scan and MRI with three-dimensional reconstruction. Arteriography can be very helpful, particularly when combined with preoperative embolization for GCTs involving the spine which can be dramatically hemorrhagic (especially sacral tumors+++) ; prior to the procedure, one must make sure that a Jouvelet roller pump and a sufficient quantity of stored blood are available !

Treatment of GCT involving the trunk region is much less standardized and also much more extensive and complex than treatment of GCT of limbs :

 

Figure 23:
a) Benign GCT of L5.

b) Vertebrectomy using a double approach (L5 vertebral body has been reconstructed with a femoral head graft).

 

- vertebrectomy, where feasible, is the best solution, but it often requires a double approach (anterior and posterior) that is performed simultaneously or successively (Fig. 23).

- curettage of benign sacral tumors is never complete and fully satisfactory. Surgery must be combined either with calcitonin injections, or sometimes and as a last resort, with radiation therapy (Figs 24 & 25).

- pelvic tumors are generally too aggressive to be amenable to curettage & bone-grafting, and they are usually managed with different resection-reconstruction techniques (Fig. 26).

- for malignant GCT, in addition to the above, chemotherapy is of course mandatory. However, as for any malignant GCT involving the trunk region, the success rate is dramatically low.

 

Figure 24: Benign sacral GCT managed with curettage and radiation therapy. Good result at 12-year follow-up.
Figure 25: Huge histologically benign sacral GCT in a 17-year-old girl with left lower-extremity monoplegia and sphincter disturbances. a) Initial x-ray.
b) Lateral MRI and arteriography revealing severe tumor blush ; in 1993, several operations performed for GCT recurrences + calcitonin + radiation therapy.
c) In 1994, huge recurrent GCT extending to the abdominal wall, anteriorly, that resulted in bilateral hydronephrosis from compression : again, resection from both an anterior and a posterior approach during which a segment of the small intestine, uterus, one ovary, one kidney (nonfunctional) had to be sacrificed ; an iliac anus was necessary.
d) Image at 10-year follow-up : no recurrence, monoplegia has disappeared, but sphincter disturbances still persist.
Figure 26: Enormous benign GCT of the hemipelvis. Resection + Saddle type prosthesis (the fork supports the remaining portion of the iliac wing).
Maîtrise Orthopédique n° 136 - August 2004
 
 
 
 
 
 
  WARNING: The Site contains information relating to health, surgery, the medical domain and various kinds of medical treatment reserved exclusively for healthcare Professionals. This is for information purposes only and is not meant to be a substitute for the advice provided by your suregon or physician. You should not use the information contained herein for diagnosing any illness or physical problem or in order to prescribe or use any pharmaceutical specialties presented on the Site.
  Hosted by XPERT-MEDECINE