10.1245/ASO.2006.05.042
Annals of Surgical Oncology 13:453-464 (2006)
© 2006 Society of Surgical Oncology
Anaplastic Thyroid Carcinoma: Biology, Pathogenesis, Prognostic Factors, and Treatment Approaches
Chandrakanth Are, MD1 and
Ashok R. Shaha, MD2
1 Department of Surgical Oncology, Memorial Sloan-Kettering Cancer Center, 1233 York Avenue, 16 I, New York, New York 10021
2 Department of Head and Neck Surgery, Memorial Sloan-Kettering Cancer Center, 1233 York Avenue, 16 I, New York, New York 10021
Correspondence: Address correspondence and reprint requests to: Chandrakanth Are, MD; E-mail: chandrakanth{at}hotmail.com.
 |
ABSTRACT
|
|---|
Background: Anaplastic thyroid carcinoma (ATC) is one of the most aggressive solid tumors known to affect humans and carries a dismal prognosis. Our primary aim was to review its epidemiology, biology, risk factors, and prognostic indicators. We also reviewed the individual and combined roles of surgery, radiotherapy, chemotherapy, and newer therapeutic options in the management of ATC.
Methods: An extensive literature review was conducted to include all published reports on ATC. The changing trends in the management of anaplastic thyroid cancer were analyzed to summarize the current practice of management of ATC.
Results: Although ATC is rare, there has been a decline in its incidence worldwide. ATC accounts for more than half of the 1200 deaths per year attributed to thyroid cancer. Long-term survivors are rare, with >75% and 50% of patients harboring cervical nodal disease and metastatic disease, respectively, at presentation. ATC can arise de novo or from preexisting well-differentiated thyroid cancer. Surgical management has shifted from tracheostomy only for palliation to curative resection when possible. Tracheostomy is performed for impending obstruction rather than for prophylaxis. Radiotherapy has evolved from postoperative administration only to preoperative treatment, combining preoperative and postoperative treatment and using higher doses, along with hyperfractionating and accelerating dose schedules. Chemotherapy has changed from monotherapy to combination therapy, and newer drugs such as paclitaxel show promise. Similarly, novel angiogenesis-inhibiting agents are currently being used, with early reports of some benefit.
Conclusions: Despite multimodality approaches, ATC still carries a dismal prognosis. This should provoke innovative strategies beyond conventional methods to tackle this uniformly lethal disease.
Key Words: Anaplastic thyroid carcinoma • Biology • Pathogenesis • Prognostic factors • Treatment approaches
|
INTRODUCTION
|
|---|
Anaplastic thyroid carcinoma (ATC) is one of the most aggressive and lethal solid tumors known to affect humans. This is in contrast to the well-differentiated thyroid cancer (WDTC) that accounts for most thyroid malignancies; it has an indolent course and a good prognosis. ATC portends a dismal prognosis, with a median survival of 4 to 12 months from the time of diagnosis.1–23 Long-term survivors are so rare3,4,24 that the diagnosis is questioned in reports that describe more than the anecdotal 5-year survivors.25–28 Cervical lymph nodal involvement is common, and >50% of the patients have metastatic disease at presentation.3,4,12,17,29 Initial treatment options were limited to palliation of asphyxiation by tracheostomy, which was invariably associated with a poor outcome. Although ATC is radiation resistant, radiotherapy (RT) was added to relieve the airway obstruction. Because of ATC’s systemic nature at presentation, chemotherapy (monotherapy and combination therapy) has been increasingly used over the last few decades. New treatment strategies, such as chemotherapy agents (paclitaxel and paclitaxel + manumycin),30–33 bovine serum ribonuclease,34 bone morphogenic protein,35 and p53 gene therapy,36,37 are being developed to alter the course of the disease. Despite this, ATC still harbors a uniformly fatal outcome and begs for a different approach to treat this lethal disease. In this article, we review the epidemiology, biology, risk factors, therapeutic options, and new strategies in the management of ATC.
|
EPIDEMIOLOGY
|
|---|
Thyroid tumors account for 1% of all neoplasms, and WDTCs (papillary and follicular) account for 80% to 85% of all thyroid tumors.38–41 The 1997 report of the Surveillance, Epidemiology, and End Results program documented that ATCs accounted for 1.6% of all thyroid cancers,42,43 with an annual age-adjusted incidence of 2 million per year.44 Although ATC accounts for a small fraction of thyroid cancers, it is responsible for more than half of the 1200 deaths per year attributed to thyroid cancer.44,45 Fortunately, the incidence of ATC has been declining over the past few years despite an increase in the number of WDTCs.46 In a report from Mumbai, the incidence of ATC decreased from 7.7% to 4.2% between 1969–1973 and 1989–1993, although WDTC increased by 3.5% during the same period.29 A similar decrease from 11% to 5% was noted in an Italian report over a period from 1979 to 1989.47 However, although the number of WDTCs has increased in Japan over the last 30 years, the incidence of ATC has remained constant.48
There are several reasons for this decline in ATC. Several cases of previously diagnosed ATC have been reclassified as lymphomas or undifferentiated medullary cancer by application of immunochemistry.8,49–51 Some authors have suggested that previous or concurrent thyroid disorder (benign or WDTC) is a risk factor for the development of ATC.1–4,8,10,24,52,53 It is likely that aggressive resection for WDTC has reduced its incidence by eliminating the risk of dedifferentiation of WDTC to ATC.10 ATC is twice as common in areas with endemic goiter, and the decline could also be due to iodine prophylaxis.8,10,45,54 Similarly, improvements in socioeconomic status have been shown to be associated with a reduction in the incidence of ATC.54
|
CLINICAL FEATURES
|
|---|
The mean age at diagnosis is 55 to 65 years, with the peak incidence in the sixth to the seventh decade of life.3,4,8,10,29,42,45,52,55 Hadar et al.17 noted that 90% of the patients were older than 50 years. There is a preponderance of the disease in women, who outnumber men by a ratio of 3.1:1 to 1.2:1.4,10,17,25,29,42
Most patients present with a rapidly enlarging mass.3,4,17 In their study extending over a 50-year period and consisting of 134 patients, McIver et al.52 noted that a rapidly enlarging mass was the most common presenting symptom in 97% of the patients. Hemorrhage into the mass may be associated with a rapid spurt in growth, with increasing pain and dysphagia. It is not unusual for the tumor volume to double in a span of 1 week.44 The mean size of the mass is 8 cm and ranges from 3 to 20 cm.8,45 Symptoms related to mechanical compression, such as dyspnea, stridor, dysphagia, neck pain, and hoarseness, are also present. Involvement of the cervical lymph nodes and recurrent laryngeal nerve is seen in up to 40% and 30% of patients, respectively.3,12,17 The surrounding structures, such as muscle (65%), trachea (46%), esophagus (44%), and larynx (13%), may also be involved in up to 70% of the patients.56 Evidence of metastatic disease is seen in 50% of the patients at presentation, and another 25% develop metastasis during the course of the illness. The lung is the most common site (80%), followed by bone (6%–15%) and brain (5%–13%).3,4,8,45,52,57 Other rare areas, such as cardiac and intra-abdominal metastasis, have also been reported.58,59 McIver et al.52 documented that 46% of their patients had evidence of metastatic disease at presentation and that 68% of the patients developed metastatic disease at some stage of their illness.
|
PATHOLOGY
|
|---|
Microscopically, there are three patterns of ATC—spindle cell (53%), giant cell (50%), and squamoid (19%)—although they carry the same prognosis.25,26,60 In their series of 134 patients, McIver et al.52 reported a distribution of 40, 38, and 22% of patients with spindle cell, giant cell, and mixed pattern, respectively and mixed pattern, respectively. The spindle cell variants are occasionally arranged in fasicles to resemble sarcoma, and at other times the arrangement resembles fibrous histiocytoma. The giant cell variant is characterized by more pleomorphism and numerous tumor giant cells, whereas the squamoid pattern is identified with irregular configuration and abundant cytoplasm. All three variants are seen as tan white fleshy large tumors with large areas of necrosis, hemorrhage, a high mitotic activity, marked invasiveness, a high tumor proliferative index, and a low apoptotic rate.61 Small-cell ATC has been a source of diagnostic confusion because most of these lesions have now been reclassified as either lymphoma or poorly differentiated medullary thyroid carcinoma. This is important because these two diagnoses have a better prognosis than ATC and may have accounted for the better-than-expected survival reported in some series. A reanalysis of the 3 of 82 patients with >5-year survival from the Mayo series revealed that 2 patients had lymphoma and that the third had medullary thyroid carcinoma.3,26
Similarly, Wolf et al.62 thoroughly analyzed 68 patients with ATC by using immunohistochemical methods and found that 65 patients had lymphoma and that the remaining 3 patients had tumors that suggested an epithelial origin. Lymphomas do not tend to display the marked cellular pleomorphism that is characteristic of ATC.49–51,60 Medullary thyroid carcinoma can be differentiated from ATC by immunohistochemical staining for neuron-specific enolase, chromogranin, and calcitonin.63 Other immunohistochemical markers that may be helpful in correctly diagnosing ATC include vimentin, keratin, 
1-chymotrypsin, and desmin.4,25,64–66 ATCs rarely stain positive for thyroglobulin.67 Carcinoembryonic antigen can help to identify the squamoid variant.25,64 The spindle cell variant can be differentiated from sarcoma with immunohistochemical staining to anticytokeratin antibodies.
Four other variants of ATC deserve mention: insular carcinoma, pure squamous cell carcinoma, carcinosarcoma, and the paucicellular variant. Insular carcinoma was previously considered a variant of ATC, but Carcangiu et al.25,68 characterized and named this new entity in 1984. The diagnosis lies morphologically between WDTC and ATC, and the histological features include formation of solid clusters (insulae) containing a variable number of follicles, variable but persistent mitotic activity, capsular and vessel invasion, and frequent necrotic foci leading to the formation of periepitheliomatous patterns. The incidence of insular carcinoma is approximately 3%.69 It is postulated that WDTC progresses to ATC by dedifferentiation of insular carcinoma70,71 through overexpression of p53.69 Small-cell carcinoma of the thyroid is extremely rare and carries an even worse prognosis than ATC.72 The paucicellular variant and carcinosarcomas are equally rare and share the same dismal prognosis.
|
PATHOGENESIS
|
|---|
Although the lethality of ATC is not in question, it is not known whether ATC arises de novo or from dedifferentiation of previous WDTCs. It is likely that at least some ATCs arise from the dedifferentiation of previous WDTCs.4,8,10,44,52 Models of thyroid carcinogenesis have shown progression from WDTC to ATC secondary to loss of the p53 tumor-suppressor gene.73 Similarly, a large proportion of ATCs develop in elderly patients with long-standing goiter or previous malignant thyroid disease; this provides evidence for anaplastic transformation. It is also well known that the incidence of ATC is higher in areas endemic for goiter and in patients with previously inadequately treated papillary or follicular thyroid cancer.4,9,10,25,44,74 In their series of ATC, DeMeter et al.10 documented that 76% of the patients had evidence of a previous thyroid disorder (benign or malignant), of which 46% had previous or concurrent WDTC. This has led to the suggestion that increased resection of WDTC may be responsible for the reduction of ATC by eliminating the potential of transformation.10
This is further supported by the fact that coexistence of WDTCs and ATCs is very well documented.1,25,75,76 In fact, some studies have reported transition zones from WDTC to ATC in the same specimen and also findings of tiny foci of WDTC within ATC and vice versa.17,27 The reported rate of WDTC’s being found in association with ATC ranges from 23% to 90%.1,4,6,8–10,16,17,22,25,52,53,69,74,77–79 Although every type of WDTC can be found in association with ATC, papillary cancer is the most common type.4,8,10,25,52,78 Within papillary cancer, the biologically aggressive forms (insular and tall cell types) are found more commonly, thus further strengthening the theory of transformation from WDTC to ATC through the intermediate forms.1,69,80–82 It has been suggested that if enough sections are taken, one will eventually find foci of WDTC in every specimen of ATC.53 In their series of 42 patients, by using entire organ sections, Ibanez et al.83 found foci of WDTC in all specimens. The inability to find foci of WDTC is thought to be due to inadequate sections or overgrowth of anaplastic cells over papillary cells.25
Evidence at the genetic and molecular levels also supports the possibility of anaplastic transformation. Because most of these data are gathered from anaplastic cell lines in vitro, they have to be interpreted with caution. The loss of expression of the tumor-suppressor gene p53 has been shown to be involved in the malignant transformation of colon, lung, and breast carcinoma.84–86 Similarly, the loss of p53 or the presence of abnormal p53 can be responsible for transformation of WDTC cells to ATC.69,87–89 Stoler et al.90 have noted that p53 alterations are rarely found in cell lines of WDTC. Along the same lines, re-expression of p53 in these cell lines has been shown to be associated with reversal of some aspects of the anaplastic transformation.36,87,90,91 These include restoration of chemosensitivity and radiosensitivity, inhibition of cellular proliferation, restoration of response to thyroid-stimulating hormone, and re-expression of thyroid peroxidase. The ploidy status has been studied to understand the process of anaplastic transformation. Whereas some authors92 have shown that WDTC and ATC share aneuploidy status, thus suggesting transformation, others75 have not (therefore supporting the de novo theory). There is enough evidence to suggest that anaplastic transformation of thyroid cancer does occur. What is not clear is whether ATC arises de novo as well, whether any specific types of WDTC are more prone to anaplastic transformation, and what mechanisms are involved. Understanding these pathways may be of help in developing treatment strategies for ATC.
|
DIAGNOSIS
|
|---|
The diagnosis of ATC is usually made on clinical examination and fine-needle aspiration (FNA). The clinical course of a rapidly enlarging mass that is firm and fixed to surrounding structures in an elderly patient is quite suggestive of ATC. The mass is usually >5 cm, presents as a single nodule or multiple nodules, and is associated with pressure symptoms and hoarseness. This diagnosis is confirmed with the help of FNA, which has been shown to be accurate in 90% of patients.44,45,93 An inability to obtain a diagnosis on FNA may necessitate an open biopsy. The reasons for failure to obtain a diagnosis on FNA may be sampling in the areas that consist of foci of WDTC or in areas burdened with necrosis, fibrosis, or hemorrhage. The diagnosis of ATC can also be incidentally found in patients who undergo operation with a known diagnosis of WDTC.
|
PROGNOSTIC FACTORS AND CLINICAL COURSE
|
|---|
Several risk factors have been shown to help prognostication for patients with ATC. Age, sex, size of the tumor, resectability, and extent of disease have been shown to affect the course of the disease.52 In a published report of 21 cases over 24 years, Tan et al.8 noted that female sex and a tumor size <6 cm conferred a better prognosis. Giuffrida and Gharib56 noted that age and extent of disease were the most important prognostic factors: patients <60 years presented with less extensive disease. In one of the largest single-institution series, Venkatesh et al.4 noted a median survival of 8 months in patients with localized disease, versus 3 months for patients with metastatic disease. Gilliland et al.42 showed that evidence of metastatic disease at the time of presentation conferred a mortality risk ratio of 3.2 (95% confidence limits, 2.0–5.1) compared with patients with localized disease only. Sugitani et al.77 reviewed their series of 47 patients over 33 years and developed a novel prognostic index. This prognostic index was based on a combination of the most important risk factors: the presence of acute symptoms, tumor >5 cm, distant metastasis, and white cell count 
10,000/µL. Patients with a prognostic index 
1 had a 62% survival rate at 6 months, whereas all patients with prognostic index of 3 and 4 died within 6 and 3 months, respectively. It is not clear whether patients with incidentally detected ATC fare better or worse compared with those with typical ATC.94,95 The favorable prognostic factors seem to include younger age, female sex, smaller lesions, small foci of ATC, and no evidence of metastatic disease.4,10,13
|
SURGERY
|
|---|
The role of surgery in patients with ATC is controversial and depends on the extent of disease at presentation. The vast majority of these patients, unfortunately, have disease beyond the bounds of any meaningful resection. It is well known that surgery alone does not have the potential to alter the course of the disease. All the same, some studies suggest that in a select subset of patients with localized disease, achieving complete resection of all gross disease without sacrificing vital structures can lead to some prolonged survivals.5,6,8,9,16 In their series of 91 patients, Junor et al.9 noted that patients who underwent total or partial thyroidectomies followed by external beam RT had prolonged survival when compared with patients who had only a biopsy. In another series, Haigh et al.5 reported an improved median survival of 43 months in patients who underwent potentially curative resection (followed by postoperative chemotherapy and RT), compared with 3 months for patients with palliative resection. In addition, they noted that the completeness of resection did not have a bearing on survival. Venkatesh et al.4 documented survival periods >24 months in 12 patients who underwent complete macroscopic resection of the tumor. Similarly, Koboyashi et al.16 reported a increase in survival from 2 to 6 months with complete macroscopic resection of tumor. It should be noted that there seems to be a selection bias among these studies: patients with less extensive disease underwent more complete resections. Conversely, McIver et al.52 from the Mayo clinic noted that the neither the extent nor the completeness of resection had any bearing on survival. Notwithstanding these conflicting results, the recently published consensus on the treatment of ATC recommends complete surgical resection whenever possible in selected patients.96 This should avoid resection of vital structures (larynx, pharynx, and esophagus) and should be attempted only if all gross cervical and mediastinal disease can be resected without excessive morbidity.96 A neck dissection should be performed only in the setting of complete macroscopic resection.
The other aspect of surgical intervention is palliation in patients with disease that is not localized. It should be clearly understood that in these patients, surgery followed by chemoradiation only prevents death from asphyxiation and may not have any effect on the distant disease.5,6 This may improve survival by a few months by preventing asphyxiation. Survival may also be improved in patients in whom the primary tumor resembles ATC but the metastasis resembles WDTC rather than ATC. These patients should be selected very carefully and should always receive chemoradiation in addition to surgery to achieve local control. Tracheostomies are performed in patients with impending airway obstruction that cannot undergo local resection. Prophylactic tracheostomies are difficult to perform in the presence of a larger firm mass and are associated with significant immediate morbidity and a high incidence of postoperative healing problems that can delay RT. Hotling et al.97 noted that patients who underwent prophylactic tracheostomies had a lower survival rate of 2 months, compared with 5 months for patients who did not receive a tracheostomy. Nilsson et al.6 showed that the number of patients requiring tracheostomy has declined dramatically over the last four decades with proper application of RT. In their last series of 26 patients, only 1 required tracheostomy, thus suggesting that judiciously used RT can help avoid prophylactic tracheostomy with its attendant complications.6
|
RADIOTHERAPY
|
|---|
Regardless of the extent of the disease, most patients die from uncontrolled local symptoms. It is also true that even in the presence of distant disease, achieving local control can improve short-term survival rates.98 Levendag et al.98 reported their series of 51 patients with ATC who were treated with RT over a 25-year period. In these patients, achieving local control offered a median survival of 7.5 months, versus 1.6 months without local control, even in the presence of distant disease in both groups. Therefore, local control of the disease is paramount, and it is here that RT can play an adjunctive role to surgery. The role of RT in the management of ATC has evolved similarly to that of surgery. RT was initially used as a local measure to prevent asphyxiation. The indications of RT have now evolved into a spectrum that ranges from palliation on the one hand to pre-operative or/and postoperative therapy to prolong survival on the other hand. The issues relating to RT include the timing of RT, the dose administered, and the pattern of delivery. Initially RT was mainly used after surgery to achieve local control. Several studies now show that preoperative RT may help to increase the resectability rate.14,99 Initial protocols that used doses <30 Gy were shown to be associated with a shorter survival period than those with doses >30 Gy.16 Current protocols use doses between 30 and 60 Gy, with the more successful studies using 46 Gy.9,16,99
Another modification of RT has been to follow a hyperfractionation protocol to keep up with the rapid doubling volumes of ATC.5,14,99 Initial attempts at hyperfractionated local RT combined with doxorubicin as a radiosensitizer by Simpson20 and Wong et al.100 proved disappointing: all patients died within 9 months and experienced high toxicity. To reduce toxicity, Kim and Leeper11,101 modified the protocol to administer 160 Gy per treatment twice daily for 3 days per week to a total of 5760 cGy in 40 days combined with doxorubicin 10 mg/m2 given 1.5 hours before RT. Although all patients died from distant disease, the median survival of 12 months was in improvement. Tenvall et al.14 modified this even further by administering preoperative (30 Gy) and postoperative (16 Gy) RT to a total of 46 Gy in 33 patients. The two groups were divided on the basis of the individual dose of RT, which was 1.0 and 1.3 Gy in the first and second group, respectively. They noted a marginal improvement in local control in the second group, with 4 patients surviving >2 years. This was followed by their more recent series of 55 patients with ATC who were treated similarly with hyper-fractionated RT, doxorubicin, and, wherever possible, surgery.99 The dose of a daily fraction of 1.0 Gy (group A), 1.3 Gy (group B), and 1.6 Gy (group C) divided the patients into three groups over three chronological time periods. Groups A and B received 30 Gy before surgery and 16 Gy after surgery, whereas group C received the entire dose before surgery. Overall, 5 (9%) patients survived >2 years, and there were no signs of local recurrence in 33 (60%) patients. Groupwise, there were no signs of local recurrence in 5 of 16 patients (group A), 11 of 17 patients (group B), and 17 of 22 patients (group C). More specifically, in patients who underwent operation, there were no signs of local recurrence in 5 of 9 patients (group A), 11 of 14 patients (group B), and 17 of 17 patients (group C). All these findings reached statistical significance, thus suggesting that RT can play a role in the management of select patients with ATC. There was also a significant correlation between accelerated RT and local tumor control, because none of the patients in group C (the most accelerated RT and subsequent operation) had local remnant tumor or local recurrence.
The efficacy of RT must be balanced with its toxicity. Kim and Leeper11 reported complications such as pharyngoesophagitis and tracheitis in their original series. Wong et al.100 noted skin changes, esophageal toxicity, and radiation myelopathy (two patients). Hyperfractionation permits the administration of higher doses over a shorter time with less toxicity.14,99 However, daily doses of >3 Gy can increase the incidence of myelopathy, and, therefore, caution should be applied. It seems that although RT does not alter the course of ATC in most patients, in combination with surgery and chemotherapy, it can prolong the short-term survival in a select subset of patients.
|
CHEMOTHERAPY
|
|---|
Because >50% of ATC patients have metastatic disease at presentation, the importance of chemotherapy in the management of ATC cannot be understated. However, outcomes with chemotherapy have been disappointing. Chemotherapy for ATC has evolved from monotherapy with doxorubicin102 to combination therapy (cisplatin, bleomycin, melphalan, and so on)103–107 to newer agents such as paclitaxel.45 Unfortunately, none of these has shown any promise. Initial monotherapy was based entirely on doxorubicin, with disappointing results in several studies.102,103,108,109 Most of these studies reported only a few patients with partial responses and almost none with a complete response. In an exhaustive review of the literature, Ahuja and Ernst110 found the response rate of doxorubicin to be approximately 22.1%. Similar disappointing results were obtained with monotherapy consisting of bleomycin, etoposide, cisplatin, and methotrexate.13,102,107 Combination therapy based on doxorubicin showed only a marginal increase in response rates. Shimaoka et al.103 reported 3 complete and 3 partial responders out of 19 patients treated with doxorubicin and cisplatin. The addition of bleomycin to this regimen resulted in only a slight increase in the mean survival up to 16 months.106 Other combinations with vincristine and melphalan have not produced any improvements in the rate of responders.107,111 It has been shown in vitro that anaplastic cell lines express less mdr1 mRNA and P glycoprotein and at the same time express more multidrug resistance–associated protein (MRP).112,113 These proteins have been shown to expel chemotherapy agents out of cells, and this may explain the resistance of ATC to almost all known chemotherapeutic agents. Chemotherapy remains the weak link in the multimodality management of ATC.
|
COMBINATION THERAPY
|
|---|
The individual failure of all modalities in the management of ATC has prompted the application of multimodality regimens. RT combined with surgery can improve local control, and chemotherapy combined with RT can increase the radiosensitivity of ATC. In one of the largest series from a single institution, Venkatesh et al.4 in 1990 reported survival periods of 13 months by combining chemotherapy, RT, and surgery. By adding RT in the postoperative period, Junor et al.9 in 1992 showed total or partial thyroidectomy to be associated with increased survival. In 1993, Mancusi et al.114 noted that the combination of chemotherapy and RT had a better survival than each alone, regardless of the extent of surgery. Tan et al.8 published their experience with the treatment of 21 patients with ATC in 1995. The estimated overall 5-year survival was 10%, and two patients survived >10 years. More specifically, for the patients who underwent total thyroidectomy, the estimated survival was 60% (median, 131 months). Tenvall et al.14 in 1994 combined preoperative and postoperative RT with surgery in 33 patients. There were no signs of local recurrence in 16 (48%) patients, and 4 patients survived >2 years. These initial studies revealed that combination therapy is indeed well tolerated in this age group, and it also improved local control. In their report of 81 patients with various combinations of chemotherapy, RT, and surgery whenever possible, Nilsson et al.6 in 1998 noted that 8 (10%) patients survived >2 years. Haigh et al.5 reported their series of 33 assessable patients with ATC in 2001: 26 of these 33 patients underwent neck exploration, of which 8 patients underwent potentially curative resection and were given adjuvant RT and chemotherapy. Four of these eight patients survived >2 years, with a 5-year survival estimate of 50%. The investigators concluded that complete resection of ATC with adjuvant chemotherapy and RT resulted in a longer survival period even in the presence of microscopic disease. In the follow-up of their 1994 study, Tenvall et al.99 reported improved local control rates in three groups treated with chemotherapy, increasing individual doses of hyperfractionated RT, and surgery whenever possible. These studies document that combination therapy produces better results, and the various combinations may hold promise in the future.
Although multimodality treatment has been shown to produce better results, controversy still persists regarding the timing of administering chemoradiation in relation to surgery. Besic et al.57 analyzed their work that included 79 patients with ATC and divided them into 2 groups: group I (n = 26) underwent operation only, and group II (n = 53) received chemoradiation. Of the 53 patients in group II, 12 patients underwent surgical exploration as well. There was no difference in survival between group I (25%) and group II (21%), but the best results (50% survival at 1 year) were obtained in the patients from group II who received chemoradiation and then underwent operation. This led to the suggestion that the appropriate time of chemoradiation is before surgery. Conversely, Sugino et al.95 reported their experience with 40 patients with ATC (29 typical ATC and 11 incidental ATC). Although some patients with incidental ATC had a better prognosis than the typical ATC patients, the remaining patients shared the same poor prognosis. RT did not improve the outcome in these patients with incidental ATC, and three of four deaths in this group were from local failure. The authors concluded that surgery should be the initial mode of treatment, followed by adjuvant therapy. Regardless of the disagreements about the sequence of treatment, multimodality therapy probably holds the best hope for future treatment strategies.
|
REVIEW OF STUDIES
|
|---|
A review of the major published reports in the literature is shown in Table 1
. The uniform theme of all these studies lies in the consistently reported disappointing results. This emphasizes the importance of the lethality of the disease and also the lack of any successful method of treating ATC. Certain trends seem to have evolved over the years in the management of ATC. Multimodality treatments are increasingly used and seem to be the norm. The surgical approach seems to have evolved from only tracheostomy to complete resection whenever possible without sacrificing major structures. Tracheostomy is performed for impending airway obstruction rather than as a prophylactic measure. RT is now increasingly applied before surgery with the hope of increasing resectability rates. Preoperative RT is combined with postoperative RT if operative intervention is possible. The total dose of RT has increased from <30 Gy to between 45 and 60 Gy. The delivery of radiation has been altered to hyperfractionated and accelerated dose schedules to keep up with the rapid doubling volumes of ATC. Doxorubicin monotherapy has been replaced by polytherapy based on either doxorubicin or newer agents such as paclitaxel. It is becoming clear that ATC cells seem to harbor chemoresistance to all known agents. This could be due to molecular mechanisms that permit expelling chemotherapy agents from ATC cells.
The management of ATC has evolved significantly over the decades. Surgical management has changed from only tracheostomy for palliation to curative resection whenever possible. This should be attempted only when complete cervical and mediastinal disease can be excised without sacrificing major structures and causing excessive morbidity. Lymph node dissection should be performed only in the setting of complete curative resection. Patients operated on for a differentiated thyroid cancer and found incidentally to have foci of ATC should undergo complete curative resection with lymph node dissection. Tracheostomy should be performed for impending airway obstruction rather than on a prophylactic basis. Prophylactic tracheostomy has not been shown to prolong survival, is difficult to perform in the presence of a hard mass, is associated with wound-healing problems, and can delay RT, which may be the only modality to help prevent asphyxiation. RT has evolved from postoperative treatment to preoperative treatment, combining preoperative and postoperative treatments, administering higher doses, and using hyperfractionating and accelerating dose schedules. Chemotherapy has changed from monotherapy with doxorubicin to polytherapy based on doxorubicin or newer agents such as paclitaxel. Despite this, ATC still carries a dismal prognosis. This should provoke the development of innovative strategies beyond the conventional methods to tackle this lethal condition.
Received for publication June 15, 2005.
Accepted for publication October 21, 2005.
TOP
ABSTRACT
INTRODUCTION
