Ann Thorac Surg 1997;63:1810-1818
© 1997 The Society of Thoracic Surgeons
Current Review
Pancoast (Superior Sulcus) Tumors
Frank C. Detterbeck, MD
Division of Cardiothoracic Surgery, University of North Carolina School of Medicine, Chapel Hill, North Carolina
 |
Abstract
|
|---|
Primary carcinomas arising in the apex of the lung (Pancoast tumors) have attracted attention because of the characteristic syndrome that is produced by local extension into the chest wall and the brachial plexus. This article reviews the history of the treatment of this disease, the natural history of untreated patients, and the diagnosis of Pancoast tumors. The published data on results, prognostic factors, and technical aspects of treatment with combined irradiation and operation are examined, as well as those pertaining to treatment with irradiation alone.
|
Introduction
|
|---|
Primary carcinomas arising in the apex of the lung often invade the lower portion of the brachial plexus, the upper thoracic ribs and vertebral bodies, the stellate ganglion, and the subclavian vessels. They produce a characteristic syndrome of pain down the arm and eventually numbness and weakness, particularly in the T1 and C8 nerve root distribution. A Horner's syndrome is seen in about 20% of patients [1–3], as a result of involvement of the sympathetic chain (Fig 1
). This article presents an up-to-date discussion of aspects of the history, diagnosis, and treatment of Pancoast tumors. Treatment has consisted of either combined irradiation and operation or irradiation alone. Data pertaining to the prognostic factors and technical aspects of each of these treatments are reviewed, as well as the recurrence patterns after treatment.
View larger version (26K):
[in this window]
[in a new window]
|
Fig 1. . Nerve involvement in a typical Pancoast tumor (C8, T1, T2, lower trunk of brachial plexus, sympathetic chain). The dermatomes of C8 and T1 are illustrated, as well as areas of referred pain in the scapular and pectoral regions (mediated through afferent fibers of the sympathetic chain).
|
|
|
History
|
|---|
The first report of a patient with a Pancoast tumor was published by Hare in 1838 [4]. This syndrome gained wider recognition when a radiologist named Henry Pancoast published a study of 7 such patients in 1932 [5]. A literature review in 1946 encompassing 151 cases concluded that the disease had been uniformly fatal, resisting all attempts at treatment [6].
Few data are available on the natural history of untreated Pancoast tumors. In an earlier report of 8 untreated patients, Herbut and Watson [6] found that all were dead by 10 months after diagnosis, and survival had ranged from 3 to 24 months from the onset of symptoms. Later, Paulson [7] observed that the average survival in untreated patients was between 10 and 14 months after diagnosis, with most patients experiencing severe, unrelenting arm pain.
In 1954, Haas and colleagues [8] reported on a patient who was alive 34 months after external-beam radiation therapy for a Pancoast tumor. This patient and 3 others in their report had experienced dramatic relief of their arm pain after radiation therapy. More than 10 years later, several 5-year survivors (for a total of 9 patients) were reported after treatment with irradiation alone [9–11].
The first 5-year survivor, however, was reported by Chardack and MacCallum in 1956 [12]. This patient underwent resection, followed by 65 Gy of irradiation. Such combination treatment, involving irradiation and surgical resection, was popularized by Shaw and colleagues [13] and has come to be viewed as the "standard" treatment. As is often the case, this approach was born out of serendipity, when these authors encountered a patient who had received 30 Gy of irradiation for an "unresectable" Pancoast tumor. Because of a good radiographic response, he then was considered for resection and lived for 27 years after operation [14]. The 1961 article by Shaw and colleagues, which popularized this combination treatment, reported on 18 patients, 12 of whom were alive without disease. Although the survival of these patients extended up to 51 months after resection [13], the follow-up period was less than 1 year in half the patients and less than 2 years in 89%.
|
Diagnosis
|
|---|
It is standard practice in most institutions to obtain a histologic diagnosis before beginning treatment. Fiberoptic bronchoscopy has achieved a diagnosis in only 30% to 40% of patients [15–17]. Fine-needle aspiration is the diagnostic procedure of choice, with a greater than 90% success rate [16, 18, 19]. However, the incidence of conditions other than non–small cell lung cancer causing Pancoast syndrome probably is small [19]. Other conditions producing a Pancoast syndrome have been described, including lymphoma, plasmacytoma, and infections with Staphylococcus, Cryptococcus, Echinococcus, and Actinomyces [20–23]. The incidence of small cell cancer is less than 5% [20]. In current reports, adenocarcinoma accounts for about two thirds of cases, and squamous cell cancers make up most of the remainder, with less than 10% being caused by large-cell cancers [24, 25].
|
Palliation
|
|---|
Good palliation of pain is achieved in most patients with either irradiation alone or irradiation and operation. Van Houtte and associates [26] reported good palliation in 75% of patients with irradiation alone. Others, using a combination of irradiation and operation, have reported excellent palliation in about 80% and good palliation in about 90% of patients [2, 27–31]. The reported duration of palliation usually is either at least 1 year or for the rest of the patient's life [26, 29]. There is some evidence that palliation with irradiation alone may be achieved more readily when higher doses (greater than 50 Gy) are used [26, 32].
|
Curative Treatment With Combined Irradiation and Operation
|
|---|
The results of all studies since 1975 reporting data on at least 15 patients with Pancoast tumors treated with irradiation and operation are shown in Table 1. To be consistent and to avoid the inflated results produced by a "diminishing denominator," data are reported on the basis of "intent to treat." In other words, the values shown in Table 1 pertain to all patients initially selected for treatment with irradiation and operation, even if they did not complete the entire treatment regimen.
In general, the number of patients who did not undergo thoracotomy because of progressive disease was small. The rate of complete resection was about 70%. The remaining 30% of patients either could not undergo resection at all or, more commonly, underwent resection but were found to have a microscopically positive margin at the time of final histologic examination. Staging usually was done radiographically, and about 10% to 20% of patients were found to have N2 disease at the time of thoracotomy. The treatment usually has involved about 30 Gy of preoperative irradiation, although in some studies as many as half the patients received only postoperative irradiation.
Five-year survival rates ranged from 15% to 50%. The stage of disease does not seem to explain the difference in survival rates. It appears that those studies reporting poor results (5-year survival rates of 25% or lower) were relatively nonselective; that is, a high percentage of all patients with Pancoast tumors seen at the institution were begun on this treatment. In addition, in many of the patients, a limited resection (wedge or segmentectomy) was performed. Centers reporting better survival (5-year survival rates of 40% or higher) appear to have been fairly selective, and resections usually involved a lobectomy.
Those patients who underwent complete resection with negative margins had a somewhat better 5-year survival rate of about 40% (Table 2). However, there is still a great deal of variation in the 5-year survival rate and the type of resection performed (lobectomy versus limited resection).
View this table:
[in this window]
[in a new window]
|
Table 2. . Pancoast Tumors Treated With Irradiation and Operation: Survival of Patients With Completely Resected Disease and a Negative Margin
|
|
|
Prognostic Factors in Combined Irradiation and Operation
|
|---|
T STATUS.
Most Pancoast tumors are T3 lesions, because they invade the parietal pleura and brachial plexus. Invasion of the subclavian vessels or vertebral bodies is defined as T4 disease and conveys a poor prognosis in patients undergoing resection. Of 22 patients with vertebral body involvement, Ginsberg and associates [24] found only two 5-year survivors (9%) after resection. Others have reported no 5-year survivors [3, 29]. Komaki and co-workers [25] reported 2 4-year survivors out of 18 patients (11%), although many of these patients may have been treated with irradiation alone. Anderson and associates [16] reported significantly worse survival (p = 0.001) with vertebral body invasion in 7 patients, but no actual survival data were given. Thus, although vertebral body invasion may not preclude long-term survival after resection entirely, it is a clearly negative prognostic factor.
Subclavian artery involvement also is a negative prognostic factor. One study has reported 1 of 4 patients alive at 2 years or longer [29], and another has reported 1 of 5 alive at 2 years [3]. Only Dartevelle and associates [34], whose anterior approach makes the resection of Pancoast tumors with vascular involvement easier, have reported 5-year survival after subclavian vessel resection (a 30% actuarial 5-year survival rate in 12 patients). Therefore, if a resection is undertaken in a patient with suspected subclavian vessel involvement despite the negative prognostic influence this carries, it may be important to choose an anterior approach.
Controversy exists about the importance of actual rib destruction by tumor. Shahian and co-workers [28] reported no difference in the 5-year survival rate with or without rib involvement (53% versus 56%, respectively). Sartori and associates [3] found the 2-year survival rate with rib involvement to be slightly inferior to that without rib involvement (45% versus 55%, respectively; no p value calculated). Komaki and co-workers [25] found the 2-year survival rate with rib involvement to be significantly inferior to that without rib involvement (25% versus 37%, respectively; p = 0.014), although many of their patients were treated with irradiation only. An example of the negative prognostic implications of vertebral body, subclavian vessel, or rib involvement is shown in Figure 2.
View larger version (16K):
[in this window]
[in a new window]
|
Fig 2. . Survival curves of patients with Pancoast tumors who had only pleural invasion compared with those who had vascular invasion, bone invasion, or both (p = 0.003). (Reprinted with permission from Sartori F, Rea F, Calabrï F, Mazzucco C, Bortolotti L, Tomio L. Carcinoma of the superior pulmonary sulcus: results of irradiation and radical resection. J Thorac Cardiovasc Surg 1992;104:679–83.)
|
|
NODAL STATUS.
Mediastinal lymph node involvement has been found to be a negative prognostic factor in many studies involving the resection of Pancoast tumors [1, 3, 16, 19, 24, 28, 35]. Survival curves similar to those shown in Figure 3 were found by two other large series [1, 3]. Only one study, which involved only 5 patients with N2 disease, suggested that there was no difference in the survival of patients with N2 versus N0 disease [2]. Although the 15% 5-year survival rate for patients with completely resected N2 or N3 disease shown in Figure 3B may seem reasonable, it must be emphasized that this is a select group. Most patients with N2 or N3 disease did not undergo resection, and a complete resection was achieved in only one third of those who did. Out of 75 patients with resected N2 or N3 disease who have been reported, only 6 (8%) were found to be 4-year survivors [1, 3, 19, 24, 28, 35].
View larger version (29K):
[in this window]
[in a new window]
|
Fig 3. . (A) Survival by nodal status for entire patient cohort. (B) Survival by nodal status for patients with completely resected disease. (Reprinted with permission from The Society of Thoracic Surgeons [Ann Thorac Surg 1994;57:1440-5].)
|
|
There is some evidence that patients with ipsilateral supraclavicular node (N3) involvement have a better prognosis than patients with ipsilateral mediastinal node (N2) involvement [1, 24]. Ginsberg and associates [24] found a 5-year survival rate of 14% in patients with N3 disease as opposed to 0% in patients with N2 disease. Similarly, Hilaris and co-workers [1] found a median survival time of 13 months for 23 patients with N3 disease, as compared with 9 months for 19 patients with N2 disease. Other studies have not differentiated between patients with N2 and N3 node involvement. It may be that for Pancoast tumors, as opposed to other lung cancers, ipsilateral supraclavicular node involvement represents limited "local" nodal extension.
It should be pointed out that pretreatment mediastinoscopy was not done routinely in any of the studies of Pancoast tumors, and only Stanford and associates [38] state that pretreatment mediastinoscopy was performed in "many" of the patients. In the other studies, the N2 involvement was found after preoperative irradiation at the time of thoracotomy. Although such N2 to N3 involvement does not absolutely preclude long-term survival, it represents a major negative prognostic factor. This would argue that pretreatment mediastinoscopy should be used in planning the appropriate treatment, and that it may be best to treat patients with N2 to N3 disease with irradiation alone.
HORNER'S SYNDROME.
The presence of a Horner's syndrome appears to be a negative prognostic factor in patients with resected disease [3, 16, 24]. Sartori and associates [3] found a 5-year survival rate of 8% in 15 patients with a Horner's syndrome, as opposed to 35% in 27 patients without the syndrome (p = 0.05). Ginsberg and co-workers [24] found a 5-year survival rate of 13% in 30 patients with a Horner's syndrome, as opposed to 26% in the overall population of 124 patients (no p value calculated).
HISTOLOGY.
It is not clear that the histologic type of lung cancer influences the survival. Hilaris and associates [1] found better survival in adenocarcinomas versus squamous cell cancers (median survival time, 25 months versus 14 months, p = 0.005). However, Shahian and co-workers [28] found the opposite trend in a much smaller group of patients (p = not significant).
|
Preoperative Irradiation
|
|---|
Ever since the original report by Shaw and colleagues in 1961 [13], 30 to 40 Gy of preoperative irradiation has been commonly used. However, the sequencing and dosage of combined modality therapy have never been examined in a randomized fashion, or even in a nonrandomized but controlled fashion. Ginsberg and associates [24] state that in their experience with 124 patients, preoperative irradiation was no better than postoperative irradiation, but no data are provided. A recent review also suggests that the results of preoperative irradiation are similar to postoperative irradiation, but without corroborating data [20]. The most controlled data available examining the value of preoperative irradiation are shown in Figure 4. The data in this figure are stratified by nodal involvement, which appears to be the most important prognostic factor. This figure shows that preoperative irradiation appears to be of benefit for each nodal stage. The use of preoperative irradiation was a significant prognostic factor in both univariate analysis (median survival time, 23 months versus 14 months; p = 0.05) and multivariate analysis [1]. The only other significant factor in the multivariate analysis was the mediastinal node status. Although this is a retrospective, nonrandomized study, it provides the only controlled data available analyzing the influence of preoperative irradiation.
View larger version (24K):
[in this window]
[in a new window]
|
Fig 4. . Survival according to mediastinal node status in 128 patients with superior sulcus tumors who received or did not receive preoperative radiation therapy (preop RT). Squares indicate preoperative radiation therapy, nodes negative, n = 70; "x"'s indicate no preoperative radiation therapy, nodes negative, n = 39; triangles indicate preoperative radiation therapy, nodes positive, n = 12; and underlined "x"'s indicate no preoperative radiation therapy, nodes positive, n = 7. (neg = negative; pos = positive.) (Reprinted with permission from Hilaris BS, Martini N, Wong GY, Nori D. Treatment of superior sulcus tumor (Pancoast tumor) . Surg Clin North Am 1987;67:965-77.)
|
|
|
Postoperative Irradiation
|
|---|
Postoperative irradiation has been used in many institutions, but its use has been inconsistent, and this hampers interpretation of the results. In 69 patients with completely resected disease from the Memorial Sloan-Kettering Cancer Center who received postoperative brachytherapy, a trend toward worse survival with brachytherapy was seen, but the difference was not statistically significant [24]. This suggests a selection bias in this retrospective series, as well as little benefit to the brachytherapy. In the earlier analysis of patients from Memorial Sloan-Kettering Cancer Center, Hilaris and associates [1] also found, by both univariate and multivariate analysis, that the use of brachytherapy and postoperative irradiation was not of significant benefit.
|
Irradiation Dose
|
|---|
Higher doses (greater than 45 Gy) of preoperative irradiation have been used only sporadically. Attar and associates [15] found increased morbidity with the use of 55 Gy preoperatively in 13 patients (23% with bronchopleural fistula; 15% mortality rate). However, Fuller and Chambers [36] used 55 Gy preoperatively in 21 patients and found no major complications and only one death (8%), caused by a cerebrovascular accident. Higher doses of irradiation do not appear to enhance resectability or survival in these reports [15, 16].
|
Extent of Pulmonary Resection
|
|---|
Controversy exists about whether a wedge resection or formal lobectomy should be performed in Pancoast tumors, which often have been viewed as exhibiting early local invasion with a very late propensity to spread along lymph node chains [19]. Two studies, both involving only small numbers of patients, have shown a small, nonsignificant trend toward better survival with limited resection [28, 34]. The only data addressing this question in a larger number of patients were reported by Ginsberg and co-workers [24] (Fig 5). Among patients who underwent a complete en bloc resection, a survival benefit was seen in those who underwent a formal lobectomy (p = 0.039). Further, the local recurrence rate was reduced after complete resection involving a lobectomy (23% versus 38%; no p value calculated). Although this was a retrospective, nonrandomized review, the findings are strikingly similar to the prospective, randomized trial of lobectomy versus limited resection carried out by the Lung Cancer Study Group [39].
View larger version (21K):
[in this window]
[in a new window]
|
Fig 5. . Survival rates for patients with completely resected disease according to type of pulmonary resection. (Reprinted with permission from The Society of Thoracic Surgeons [Ann Thorac Surg 1994;57:1440–5].)
|
|
|
Incomplete Resections
|
|---|
Given the location of Pancoast tumors, it is not surprising that there is difficulty achieving a complete resection in many cases. The most common sites of residual tumor are along the brachial plexus, neural foramina, vertebral bodies, and subclavian vessels [2]. The completeness of the resection has a significant effect on the 5-year survival rate [2, 24, 28, 40].
From the data of Ginsberg and colleagues [24] (Fig 6), it appears that there is little difference between an incomplete resection and no resection at all, which is similar to the results of incomplete resection of other non–small-cell lung cancers [41]. Two studies [2, 40] have reported statistically significant poorer 2-year survival rates for patients with incompletely resected disease (0% and 11% versus 60% and 43%). However, the 5-year survival rate of 50% in nine patients with positive margins reported by Shahian and co-workers [28] is at odds with this. Certainly, it appears that every attempt should be made intraoperatively to achieve a complete resection.
View larger version (26K):
[in this window]
[in a new window]
|
Fig 6. . Survival of 124 patients with superior sulcus tumors. The asterisk indicates an incomplete resection curve estimated from reported data of the overall group, and those with complete resection or no resection. (Adapted with permission from The Society of Thoracic Surgeons [Ann Thorac Surg 1994;57:1440–5].)
|
|
Few data are available regarding whether postoperative irradiation is of benefit in patients with incompletely resected disease. Of the patients who had incomplete resections at Memorial Sloan-Kettering Cancer Center, postoperative brachytherapy was used in practically all cases, but it appears to have been of little benefit (see Fig 6
). In the series, reported by Maggi and colleagues [2], the patients with incomplete resections also had poor survival. Although not stated explicitly, it appears that only about half these patients received postoperative external-beam radiation therapy. Similarly, the use of brachytherapy in the series reported by Hilaris and associates [1] made no difference in survival time (median survival time, 19 months versus 16 months; p = 0.58). This series contained 129 patients, all of whom had thoracotomy, but only 63% of whom had a resection. On the other hand, Shahian and colleagues [28] reported a 5-year survival rate of 50% in 9 patients who had positive margins, all of whom were treated with postoperative irradiation. Thus, the preponderance of data suggests no benefit to postoperative irradiation in patients with incompletely resected disease.
|
Curative Treatment With Irradiation Alone
|
|---|
Several series of patients with Pancoast tumors treated with irradiation alone have been published, and the results of all series since 1975 involving at least 15 such patients are summarized in Table 3. Unfortunately, it is unclear in most of these studies whether the irradiation was given in a curative attempt or simply to palliate symptoms. The doses of irradiation given were usually at least 50 to 60 Gy, which is higher than the doses generally used for palliation alone. However, in many of these reports, the disease appears to have been quite advanced, and the long-term survival rates are low [15, 16, 25, 37, 38]. The studies that appear to have selected patients who had a reasonable chance of cure report 5-year survival rates of about 20% [26, 27, 31].
|
Prognostic Factors for Treatment With Irradiation Alone
|
|---|
Because the limitations of operation and irradiation are different, the prognostic factors for treatment with each of these modalities should be analyzed separately. Unfortunately, little information is available on prognostic factors in patients with Pancoast tumors treated with irradiation alone. The survival has not been correlated with T, N, and M stage. Van Houtte and associates [26] reported that scalene lymph node involvement and radiographic evidence of rib destruction were poor prognostic factors, but no statistical significance was calculated. Ahmad and co-workers [31] also suggested decreased survival when rib destruction was present, but analysis is difficult because 25% of the patients underwent operation in addition to irradiation.
Komaki and associates [25] reported on numerous prognostic factors in a group of 85 patients with Pancoast tumors, two thirds of whom were treated with irradiation alone. Patients with a Karnofsky performance status of 80 or greater clearly exhibited longer survival than patients with a poor performance status (p < 0.0001). Patients with less than 5% weight loss also had a better prognosis (2-year survival rate, 47% versus 17%; p < 0.0003). Patients with vertebral involvement had worse survival (2-year disease-free survival rate, 15% versus 40%; p < 0.0006). Similarly, patients with stage IIIA disease did better than those with stage IIIB disease (2-year disease-free survival rate, 47% versus 21%; p = 0.0042) [25]. However, these patients received a variety of treatments, and each of these prognostic factors affected their selection for type of treatment. Treatment involving operation was clearly associated with longer survival (2-year survival rate, 52% versus 22%; 5-year survival rate, 40% versus 8%; p < 0.0035). Thus, the limited data on prognostic factors for patients treated with irradiation alone are not entirely clear. It seems likely, though, that vertebral body involvement, rib destruction, poor performance status, and weight loss are poor prognostic factors for treatment with irradiation alone, similar to the prognostic factors for treatment with combined therapy.
|
Technical Aspects of Treatment With Irradiation Alone
|
|---|
No information is available about the effect of irradiation dose, size of treatment fields, or fractionation on survival. There is some evidence, however, that local control is improved with higher doses of irradiation. In general, local control with irradiation alone has been achieved in about half of patients [20]. Van Houtte and associates [26] found that local control was better with doses greater than 50 Gy. Similar results were reported by Komaki and co-workers [25] for those patients treated with irradiation alone (local control rate for patients receiving greater than or less than 65 Gy: 69% versus 38%, respectively; p < 0.05). In an earlier study, Komaki and co-workers [27] found that the local control rate was improved with larger (greater than 12 cm2) field sizes (52% versus 36%; no p value calculated). Better control rates also have been found in patients who received continuous irradiation as opposed to split-course irradiation (50% versus 18%; p = 0.05) [25].
|
Recurrence Patterns After Curative Treatment
|
|---|
A summary of recurrence patterns, both for combined therapy and for irradiation alone, is shown in Table 4. However, these data must be viewed with caution because in most studies, recurrence data are reported inconsistently and ambiguously. For example, it seldom is stated clearly whether "recurrence" means the first site of recurrence, any recurrence at any time, or recurrence as a cause of death.
Approximately two thirds of patients experience a recurrence. When first site of recurrence is analyzed, the recurrence also is locoregional in about two thirds of patients. Not surprisingly, when any recurrence at any time is reported, the recurrences are distributed more evenly among the categories. However, locoregional recurrence still remains the most common single category. Distant recurrences appear to be more important as a cause of death.
It is difficult to draw further conclusions. The two studies that have shown surprisingly low overall recurrence rates have used both preoperative and postoperative radiation therapy as well as resection [28, 36]. In one of these studies, the local recurrence rate appears to be low [28], but this was not the case in the other study [36]. Further, postoperative brachytherapy was used in most patients at Memorial Sloan-Kettering Cancer Center, but without an apparent benefit in either the overall recurrence rate or the percentage of local recurrences [24].
There is no obvious difference in the local recurrence rate among series that have used irradiation alone as the primary modality and those that have used both radiation therapy and operative treatment. Further, local recurrences have remained frequent, even with what might be termed maximal efforts at applying local therapies. Ginsberg and co-workers [24] found that even in 69 patients with completely resected disease and negative margins after preoperative irradiation, the first site of recurrence was locoregional in two thirds of the cases. Thus, it is not clear that good locoregional control can be achieved even with preoperative irradiation and a complete resection. On the other hand, Maggi and associates [2] found that local recurrence was the cause of death in only 3% of 36 patients with completely resected disease.
The most common site of distant metastases appears to be the brain, accounting for 40% to 80% of distant recurrences [1, 2, 24, 25, 27, 35]. In one carefully analyzed study of 68 patients treated with irradiation alone, brain metastases developed in 53% of the patients by 3 years [30]. Bone metastases appear to be the second most common site of distant metastasis [1, 25, 36]. New primary lung tumors have been described sporadically [2, 3, 26, 36].
|
Chemotherapy
|
|---|
The existing data concerning chemotherapy in Pancoast tumors are too limited to permit interpretation. Only 33 patients with varying stages of disease, who received chemotherapy as part of a variety of treatment regimens, have been described [25].
However, an extensive body of literature exists regarding chemotherapy with irradiation, operation, or both in patients with stage IIIA or IIIB (non-Pancoast) disease. This includes several randomized phase III trials, most of which have shown a survival benefit with induction chemotherapy in both a surgical [44, 45] and a nonsurgical setting [46–49]. Concurrent chemotherapy and irradiation seeks to exploit the irradiation-sensitizing effects of chemotherapy. This approach seems to improve local control rates [49–51]. Such an approach followed by resection has yielded exciting results (2-year survival rate, 40%) in patients with stage IIIA or IIIB (non-Pancoast) disease with pathologically proven mediastinal node involvement [52].
Thus, it makes sense to explore the use of concurrent chemotherapy and irradiation followed by resection in Pancoast tumors, where local recurrence rates have been about 70%. Such an approach has been developed by the Southwest Oncology Group and is currently open through all the major oncology groups as an Intergroup Study (INT-0160).
|
Conclusion
|
|---|
Pancoast tumors are rare lung cancers that produce a characteristic syndrome resulting from involvement of the brachial plexus. Fine-needle aspiration is reliable in establishing a diagnosis. Palliation of pain is achieved in most patients with irradiation alone as well as with a combination of irradiation and operation.
Curative treatment has involved either irradiation alone or preoperative irradiation followed by resection. The 5-year survival rate of good-risk patients treated by irradiation alone is about 20%. Few data are available concerning prognostic factors for patients treated with irradiation alone. It appears that patients treated with higher doses may be more likely to have local control. However, long-term local control is achieved in only about one third of patients.
The 5-year survival rate after preoperative irradiation followed by resection varies from 15% to 50%. This percentage may be related to how carefully the patients were selected and how extensive a resection was performed. Survival is better when lobectomy is performed as opposed to wedge resection or segmentectomy. The presence of a Horner's syndrome, mediastinal nodal metastases, and involvement of the subclavian vessels, vertebral bodies, or ribs are negative prognostic factors. Preoperative irradiation does appear to be of benefit. The preponderance of data does not support the use of postoperative irradiation in patients with either completely resected or incompletely resected disease. A complete resection with negative margins is achieved in about two thirds of patients. Survival after incomplete resection is poor, similar to that of patients who did not undergo resection. The use of concurrent irradiation and chemotherapy before surgical resection is being investigated.
|
Footnotes
|
|---|
Address reprint requests to Dr Detterbeck, 108 Burnett-Womack Bldg, CB 7065, Chapel Hill, NC 27599-7065.
|
References
|
|---|
- Hilaris BS, Martini N, Wong GY, Nori D. Treatment of superior sulcus tumor (Pancoast tumor). Surg Clin Nort
Top
Abstract
Introduction
