| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Plenary Session |
1 From the Departments of Radiation Oncology (R.K.), Diagnostic Radiology (M.H.C.), Thoracic and Cardiovascular Surgery (J.B.P.), Thoracic/Head and Neck Medical Oncology (F.V.F.), and Pathology (J.Y.R.), University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 97, Houston, TX 77030; the Department of Pulmonary and Mediastinal Pathology, Armed Forces Institute of Pathology, Washington, DC (W.D.T.); and the Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee (R.W.B.). From the Oncodiagnosis Panel at the 1999 RSNA scientific assembly. Received March 5, 2001; revision requested April 2 and received May 21; accepted May 31. Address correspondence to R.K. (e-mail: rkomaki@mdanderson.org).
Index Terms: Lung neoplasms, diagnosis, 60.32 • Lung neoplasms, metastases, **.332 • Lung neoplasms, staging, 60.32 • Lung neoplasms, surgery, 60.32 • Lung neoplasms, therapeutic radiology, 60.32 • Lung neoplasms, therapy, 60.32
 |
Introduction |
|---|
 Top Introduction Case 1Case 2Case 3ConclusionsReferences |
|---|
The patient in case 1 had limited small cell lung cancer (SCLC) and a history of heavy smoking. The importance of differentiating the diagnosis of pure SCLC from small cell carcinoma mixed with adenocarcinoma or neuroendocrine carcinoma is discussed. The metastatic work-up for SCLC and associated paraneoplastic syndromes is described. The discussion emphasizes combined chemotherapy and radiation therapy (RT), including the timing and dose of thoracic RT as related to the course of chemotherapy and the importance of prophylactic cranial irradiation (PCI) in improving survival.
The patient in case 2 had squamous cell carcinoma of the left lung. Radiologic imaging showed probable invasion of the tumor into the mediastinum, which influenced the resectability of this lesion. The patient had a cardiovascular problem as well. Treatment decisions include the medical condition of the patient in addition to the extent of the disease.
The patient in case 3 had poorly differentiated adenocarcinoma of the left pulmonary apex that manifested as a superior sulcus tumor. He was found to have extensive tumor involving the left subclavian artery. Despite the extensive disease, this young, physically fit patient underwent surgery with a subclavian artery graft and received postoperative chemotherapy, RT, and PCI. The pre- and postoperative adjuvant treatment for superior sulcus tumor is discussed.
|
Case 1 |
|---|
|
TopIntroductionCase 1Case 2Case 3ConclusionsReferences |
|---|
Chest radiography revealed a right lung mass (Fig 1a). Chest computed tomography (CT) also showed the mass as well as right hilar (Fig 1c), subcarinal (Fig 1b, 1c), and paratracheal (Fig 1d) nodal enlargement. There was no pleural effusion. Biopsy performed during bronchoscopy revealed undifferentiated SCLC demonstrating scanty cytoplasm, ill-defined borders, hyperchromatic nuclei, and absent nucleoli (Fig 2) along with multiple mitoses (Fig 3). Although CT showed enlarged right paratracheal lymph nodes, mediastinoscopic biopsy of the lymph nodes showed no metastases. The remainder of the metastatic work-up, which included magnetic resonance (MR) imaging of the brain, CT of the upper abdomen, bone scanning, and bone marrow biopsy, also showed no metastases.
|
|
|
|
|
|
The patient is alive without any evidence of disease 7 years after completion of treatment. There was no evident neuropsychologic deterioration when results of pre- and posttreatment neuropsychologic testing were compared (2).
Diagnostic Radiologist’s View
Question: How Does SCLC Usually Manifest at Routine Chest Radiography?—
SCLC may manifest in a variety of ways at routine chest radiography. In many cases, this malignancy is centrally located with extensive metastatic involvement of the ipsilateral hilum and mediastinum, but it is not unusual for the involvement to extend across the midline to involve the contralateral mediastinum and even the hilum on that side (extensive disease). However, it may manifest in a more limited form involving only one hemithorax. Occasionally, the disease may even manifest as a solitary lesion within the lung parenchyma (3,4). Note that squamous cell carcinoma may manifest in exactly the same way as SCLC; on rare occasions, adenocarcinoma can manifest as a central mass with a similar metastatic pattern. The presence of a pleural effusion complicates the appearance of the limited form of SCLC because, if malignant, it does affect the staging and can affect patient outcome (see the medical oncologist’s discussion later in this article).
The radiologic appearance of SCLC in this patient does not define its histologic features. Although the lateral chest radiograph showed the mass to be somewhat central in association with the right hilum, its overall characteristics were better defined at CT (Fig 1b, 1c). Histologic confirmation with biopsy or cytologic confirmation with sputum or fine-needle aspiration is necessary for a definitive diagnosis of SCLC.
Question: What Is the Present Role of Positron Emission Tomography in Staging Lung Cancer?— The application of positron emission tomography (PET) to stage lung cancer continues to evolve. Studies have shown its ability to not only demonstrate the metabolic activity within a primary lung cancer but, in many cases, to allow staging of the disease (5–7). General applications of PET in oncology have been explored in survey articles (8). Comparison analyses between PET and CT have indicated greater sensitivity and specificity with PET (5). However, the anatomic presentation of an abnormality at CT (or MR imaging) surpasses that at PET, but it is PET that has the greater capability to allow localization of a focal area of activity within the overall lesion. Therefore, in practical terms, both modalities are necessary to fully define the stage of disease. This combined approach may have its greatest value in evaluation of residual or recurrent lung cancer after various combinations of therapy (ie, surgery, chemotherapy, and radiation) (9).
Specific PET studies for evaluation of solitary pulmonary nodules based on metabolic activity have shown great promise, but false-positive and false-negative results do occur (10,11). PET evaluation of abnormal adrenal morphology noted at CT of patients with lung cancer has allowed accurate verification of the malignant nature of the abnormality and may eventually eliminate other methods of evaluation (eg, MR imaging) (12).
False-negative PET scans have been reported (13,14). Since the positivity of fluorodeoxyglucose (FDG) PET scans is dependent on the glucose uptake rate, FDG PET scans might be negative if the tumor is slow growing. Erasmus and colleagues (13) reported that PET scans of six of seven carcinoid tumors were negative because of low uptake (standardized uptake ratio <2.5). Usually, typical carcinoid tumors are slow growing and demonstrate minimal mitotic activity, resulting in less FDG accumulation than in non–small cell lung cancer (NSCLC). In another study, FDG PET scans of four of seven bronchioloalveolar carcinomas were negative because the tumors were less proliferative and had a longer doubling time than other NSCLCs (14).
Decision-tree analysis of the cost-effectiveness of PET in conjunction with CT in the staging of NSCLC has shown positive results in decreasing the overall costs of this disease based on the potential for eliminating unnecessary additional tests or surgery (15).
Pathologist’s View
Question: Is There Any Significance to the Subclass of SCLC as a Prognostic Factor?—
In the past, SCLC was divided into four subgroups: oat cell carcinoma; small cell carcinoma, intermediate cell type; mixed small cell–large cell carcinoma; and combined small cell carcinoma. Problems that were subsequently recognized included difficulties in reproducibility and lack of prognostic significance. Therefore, it was recommended that the terms oat cell carcinoma and small cell carcinoma, intermediate cell type be dropped and that all tumors with pure histologic features simply be termed small cell carcinoma. The category of mixed small cell–large cell carcinoma was defined as a tumor with both small cell and large cell carcinoma components, but subsequent studies did not clearly confirm clinical significance or interobserver reproducibility for this category. The category of combined small cell carcinoma included cases with a mixture of small cell carcinoma and squamous cell carcinoma, large cell undifferentiated carcinoma, or adenocarcinoma. The World Health Organization recently divided small cell carcinoma into two subgroups: small cell carcinoma and combined small cell carcinoma (16). The term small cell carcinoma alone is used for tumors of pure histologic features without a non–small cell component. Combined small cell carcinoma is a mixture of small cell carcinoma and any other non–small cell components, including large cell neuroendocrine carcinoma (Table 1).
|
Medical Oncologist’s View
Question: How Do You Stage Cases of SCLC and What Are the Important Prognostic Factors?—
In each patient, a history should be obtained, physical examination should be performed, and the diagnosis previously obtained with bronchoscopy or transthoracic needle biopsy should be confirmed. Hematologic and blood chemistry tests are performed, including liver function tests with measurement of lactate dehydrogenase and alanine transaminase levels. Radiologic studies should include chest radiography and CT of the chest with the upper abdomen to include the adrenal glands. Further work-up should include CT or MR imaging of the brain and bone scanning. The value of bone marrow aspiration and biopsy is debatable; however, because there is a 10% prevalence of isolated bone marrow involvement, many clinicians would include this test in a patient with otherwise limited disease, particularly if there is an elevated lactate dehydrogenase level or abnormal blood cell count.
The stage of disease at diagnosis has an important bearing on the ultimate outcome of therapy. Patients with limited disease treated with concurrent chemotherapy and RT have a median survival of 14–30 months and a cure rate of about 20%. In contrast, patients with extensive disease have a median survival of only 8–14 months with cures being exceedingly rare (18–23).
Limited disease is defined as disease confined to the hemithorax, although the presence of a malignant pleural effusion will adversely affect the outcome. Whether the presence of negative cytologic features in the pleural effusion affects the outcome is controversial in SCLC. Extensive disease is defined as disease beyond the hemithorax, such as involvement of the lungs bilaterally or extension beyond the chest, often to the brain, bone, bone marrow, or upper abdomen. Patient factors that influence outcome are performance status and gender. Age is not a significant prognostic variable in patients with limited SCLC (24). Continuation of smoking will adversely affect the outcome. Other prognostic factors are elevated lactate dehydrogenase level, the alkaline phosphatase level, hyponatremia, and possibly presence of a paraneoplastic syndrome. The biology of SCLC is an important prognostic factor. Excellent reviews of the biology of SCLC have been published by Carney (25) and Stahel and Weber (26).
The most important negative prognostic factors for patients with SCLC include extensive stage of the disease, poor performance status, weight loss, and elevated lactate dehydrogenase level. In patients with extensive disease, a higher number of metastatic sites and involvement of the liver, central nervous system (CNS), or bone marrow are also associated with a worse prognosis. A recent study found that other prognostic factors such as age and gender (which are controversial) are independent characteristics (27).
Question: What Is the Best Chemotherapeutic Agent at Present for SCLC?— SCLC is generally considered a systemic disease, even in patients with apparent limited disease. Therefore, chemotherapy has been the cornerstone of management of this disease. In patients with limited disease, chemotherapy is generally administered concurrently with RT. In contrast, patients with extensive disease are treated with chemotherapy only, with RT reserved for symptom palliation or consolidation in those patients who achieve a complete response.
In SCLC, many drugs show significant activity. These include cisplatin, etoposide, doxorubicin, cyclophosphamide, docetaxel, paclitaxel, and topotecan. Although many of the newer agents are currently undergoing investigation in various combinations, the standard frontline regimen remains cisplatin plus etoposide.
For limited disease, patients with a good performance status are treated with four cycles of chemotherapy plus RT to the primary tumor within the chest. RT is generally given concurrently with the first one or two cycles of chemotherapy. Randomized trials show a clear survival advantage for patients receiving chemotherapy and RT versus chemotherapy alone. Most patients with limited disease respond well to treatment, both subjectively and objectively; approximately 20%–25% of patients with limited SCLC may be cured with chemotherapy and RT (22,28).
For extensive SCLC, treatment consists of chemotherapy alone, usually six cycles. In these patients, RT is used only for palliation of symptoms. As in limited disease, most patients with extensive SCLC respond to treatment with improvement in symptoms and prolongation of survival. However, cure of extensive SCLC is rare. Some controversial areas for patients with extensive disease are the role of consolidation within the lungs and the use of PCI in patients who achieve a complete response.
Surgical Oncologist’s View
Question: Is There Any Role for Surgery in Treatment of Limited SCLC?—
Negative results at metastatic evaluation may suggest that appropriate treatment would include local control by means of resection. If the patient has a clinical T1 N0 tumor and is physiologically fit to undergo resection (ie, has appropriate and adequate cardiac and pulmonary function), the patient may undergo repeat bronchoscopy performed by the surgeon and mediastinoscopy. Positive mediastinoscopy results represent positive N2 lymph nodes, and the patient should be treated with concurrent chemotherapy and RT. Negative mediastinoscopy results provide an objective basis for resection. Postresection treatment should include chemotherapy (such as cisplatin and etoposide for four cycles) and consideration of PCI for patients with limited disease.
The role of surgery in SCLC is controversial (29–31). Surgery has several potential advantages in terms of decisive local control. Resection will (a) reduce the frequency and extent of local relapses, (b) not impede the intensity of preoperative chemotherapy, (c) not affect the bone marrow, and (d) improve the pathologic staging to enhance prognostic significance. Several smaller retrospective studies have focused on surgery alone or surgery in combination with chemotherapy or RT (1,32–36). Both tumor status and nodal status appear to be significant prognostic factors. Surgery may yet play an important role in SCLC, especially in diagnosis, staging, and optimization of local control for localized SCLC without nodal involvement (T1 N0 M0), which is rare. With new improved noninvasive staging procedures such as PET, resection may lead to cure in selected patients. Prospective randomized studies are needed to settle this issue.
Radiation Oncologist’s View
Question: What Is the Role of RT in Limited SCLC?—
Chemotherapy is considered the primary treatment modality for limited SCLC, but thoracic RT and PCI also play a significant role. The case presented is somewhat unusual in that the patient had limited SCLC at a relatively young age. Fortunately, he had no weight loss and a favorable performance status. These factors allowed him access to aggressive combined modality treatment as part of a national cooperative group trial. The following discussion will focus not on the chemotherapy aspects of his treatment but will outline the recent evolution of both thoracic RT and PCI to their current use in the context of the clinical trials that evaluated their usefulness.
Although thoracic RT alone had been used with limited success in limited SCLC prior to the availability of effective chemotherapy, its role became controversial when better systemic agents were developed (37). Both retrospective and prospective trials demonstrated a distinct local control benefit from the addition of thoracic RT to chemotherapy, but it was difficult to demonstrate a survival benefit (38). However, in the early 1990s, a large meta-analysis of these trials showed that chemotherapy plus thoracic RT results in better survival than chemotherapy alone (39). Presumably, one mechanism by which thoracic RT might improve survival is by eradication of chemoresistant clonogens, thus reducing the risk of posttreatment "seeding" of distant sites.
Despite the demonstration that thoracic RT improves both local control and survival, the optimal integration of chemotherapy and thoracic RT remains to be defined. At issue for thoracic RT are questions regarding total dose, volume to be treated, and fractionation schedule, as well as the timing of thoracic RT with chemotherapy. The patient in the presented case was entered into a national clinical cooperative research trial intended to answer some of these questions.
Early studies suggested that rapid regression of SCLC might require relatively moderate total doses of daily standard RT to control limited SCLC in the chest. However, subsequent investigations revealed that doses close to those used for NSCLC were necessary for optimal tumor control, at least 50 Gy. It is difficult to demonstrate a dose-response curve above 50 Gy (40). Some of the difficulty in assessing a dose response may be due to the fact that up to 20% of SCLC cases may have mixed histologic components, a finding not always discerned in the limited sample provided by biopsy, which yields only a cytologic specimen. In such small samples, NSCLC elements can easily be missed.
A variety of RT fractionation schedules have been used. In general, schedules that deliver greater than 2 Gy per fraction have been used in Canada and Europe, whereas 1.8–2 Gy per fraction has been used in the United States. Phase I dose-escalation trials have established the maximum tolerated dose of standard RT (1.8 Gy per fraction) to be at least 70 Gy in 35 fractions over 7 weeks when given with platinum-based chemotherapy (41). Two- and 3-year survival rates were 54% and 35%, respectively. However, further testing needs to be done to more clearly establish the effect of this dose on local control and survival, since the patient cohort was small. On the basis of the available information, there is nothing to suggest that daily fractionation might produce a better local control rate or survival rate than other RT fractionation schedules, including accelerated fractionation, as discussed in the next paragraph.
Multiple-daily-fractionation thoracic RT has also been investigated (42). One such approach has been to use accelerated fractionation, namely, the delivery of the total thoracic RT dose in a shorter overall time than with standard RT. The rationale for this technique is that SCLC is capable of accelerated repopulation in response to the cell loss induced by the initial weeks of standard RT. Accelerated fractionation thoracic RT is thought to overcome this phenomenon. A phase I dose-escalation trial has established the maximum tolerated dose of accelerated fractionation RT (1.5 Gy per fraction twice daily) to be approximately 45 Gy in 3 weeks when given with platinum-based chemotherapy (42). Two- and 3-year survival rates were 52% and 25%, respectively, for this regimen. A phase II trial demonstrated favorable tolerance and survival when accelerated fractionation RT to 45 Gy/3 wk (1.5 Gy twice daily) was used concurrently with chemotherapy (42). This regimen was found to be superior to standard RT (45 Gy/5 wk) given concurrently with the same chemotherapy regimen in a large phase III cooperative group trial (31). A clinical trial is planned to investigate whether standard RT to a higher dose (67 Gy/7 wk) is as effective as accelerated fractionation RT to 45 Gy/3 wk (to avoid the inconvenience of twice-daily fractionation). An ongoing Radiation Therapy Oncology Group (RTOG) trial (study 97-12) is investigating whether higher effective doses of accelerated fractionation RT can be achieved by using different means of accelerating the thoracic RT. This dose-escalation trial uses an "end-loaded" concomitant boost approach. Although twice-daily accelerated fractionation RT may be logistically difficult in some radiation oncology clinics, the treatment duration is only 3–4 weeks versus 6–7 weeks for the higher-dose standard RT schedules. With current approaches producing 2-year survival rates exceeding 40%, enough patients will have long-term survival to be concerned about late toxic effects with both the higher-dose standard RT and accelerated fractionation RT schedules.
The timing of thoracic RT relative to the course of chemotherapy for SCLC is important. Three basic strategies define how chemotherapy and thoracic RT can be combined. Thoracic RT can be given before, during, or after chemotherapy. In most earlier trials, thoracic RT was usually preceded by chemotherapy to maximal tumor response. The underlying strategy was based on the view that SCLC is a systemic disease and that control of occult extrathoracic disease is necessary for overall control. Thus, it was thought necessary to give the "primary" treatment first. Also, sequencing permitted the chemotherapy and RT to be used with little modification from how each would be used alone. Thus, one could avoid having the chemotherapy interfere with delivery of the thoracic RT and vice versa. Although this strategy was certainly safer from a toxicity point of view, it did not fully exploit the potential benefit of giving chemotherapy and thoracic RT concurrently, namely, radiosensitization of the tumor by the chemotherapy. In addition, it did not account for the potential of chemotherapy to induce tumor cell accelerated repopulation. Even after an apparent good response to chemotherapy, the tumor may go into an accelerated growth phase, which can reduce the effectiveness of the thoracic RT (43–45). Any combined strategy that extends the overall time of treatment and fails to take account of this tumor response may be disadvantaged by design. Also, the risk of developing drug resistance increases with increasing length of exposure to a cytotoxic agent, favoring a concurrent or reduced overall time strategy (46).
The median survival in trials that favored early irradiation has been 15–20 months, whereas the median survival has been about 12 months in trials that concluded that timing is not critical (21,47,48). A recent large American intergroup trial of over 400 patients started all irradiation on day 1 of treatment and reported a median survival of approximately 20 months and a 2-year survival rate of more than 40% (49). These results represent the best survival data reported from any large randomized multi-institution trial. The potential reasons why this regimen proved superior are the concurrent use of chemotherapy and thoracic RT, as well as the use of shorter-course accelerated fractionation thoracic RT, which overcame the negative effect of accelerated repopulation. However, the 30% prevalence of severe acute esophagitis observed in the chemotherapy and accelerated fractionation thoracic RT arm of the trial reflects the downside risk of this regimen. When the accelerated fractionation thoracic RT is given with cycle one of chemotherapy, a larger volume of the lung and esophagus must be irradiated. In the presented case, the treatment protocol called for accelerated fractionation to 45 Gy in 3 weeks with concurrent ifosfamide, cisplatin, and etoposide therapy, which yielded a complete response and a 7-year survival free of recurrence for him.
Patients with a complete response to combined-modality therapy should be considered for PCI. The long-term sequelae of PCI have been reduced substantially by using conventionally fractionated irradiation and by delaying PCI until after the delivery of chemotherapy. As disease-free survival has improved in SCLC, the risk of CNS-only failure has increased. The cumulative risk of CNS failure may be as high as 50%–60% (50). Irradiation decreases the likelihood of isolated CNS failure to the range of 10%–15%. Prevention of CNS failure has a significant quality of life benefit, even if the patient ultimately succumbs to systemic failure. This benefit by itself would be sufficient to recommend PCI in most patients who achieve a good partial response or a complete response. However, recent meta-analyses have demonstrated that PCI may also marginally affect survival (51). Although much has been written about concerns for PCI-related CNS toxic effects, recent reviews suggest that significant toxicity is relatively infrequent with modern treatment. In clinical trials of the usefulness of PCI that included prospective neurologic and neuropsychometric assessment, no significant detriment could be attributed to PCI at the dose schedule used (52). Some of the neuropathic effects attributed to PCI in the past may in fact be related to the use of PCI concurrently with chemotherapy, especially when agents that are also neurotoxic are used. Recent investigations have also found evidence that CNS neuropathic changes may be a related paraneoplastic syndrome particular to SCLC (53).
Several European studies also suggest that the so-called standard doses of PCI of 25 Gy in 10 fractions or 30 Gy in 15 fractions may not provide the most optimal freedom from subsequent CNS metastases (54). A large international trial is being planned that will explore higher doses of PCI either with standard fractionation (36 Gy in 18 fractions over 3.5 weeks) or accelerated fractionation (36 Gy in 24 fractions over 2 weeks). The patient in the presented case received PCI to 25 Gy in 10 fractions after a complete response to treatment of the thoracic disease. Unlike in most clinical settings in which PCI is used, in the presented case pre- and posttreatment neuropsychologic testing was performed, which demonstrated no significant early or late CNS toxic effects (53).
|
Case 2 |
|---|
|
TopIntroductionCase 1Case 2Case 3ConclusionsReferences |
|---|
|
|
|
|
|
|
The patient experienced no acute severe side effects from the treatment, and the tumor regressed. There was no evidence of local recurrence or distant metastases for 3 years after completion of the treatment. His quality of life was excellent without any late toxic effects.
Diagnostic Radiologist’s View
Question: Can You Describe Specific Imaging Findings of Postobstructive Pneumonia?—
Centrally located carcinomas often result in volume loss within the lung parenchyma distal to the bronchus or bronchi that are obstructed. Specifically, postobstructive pneumonia often, but not always, produces a well-defined pattern of such volume loss, which may be segmental, lobar, or multilobar or involve the entire lung. Most cases of such obstruction are demonstrated at routine chest imaging. In many cases, CT better defines the overall nature of such volume loss, especially if the area of volume loss contains regions of necrosis within a larger exophytic component of the central lesion (Fig 5) (55,56). However, use of iodine contrast material during CT is usually necessary to define regions of exophytic involvement or necrosis within the area of obstruction (56). Observation of wedgelike volume loss, lack of air bronchograms or the presence of fluid-filled bronchi within the region of volume loss, or the presence of normal-appearing pulmonary vasculature within the region are all signs of postobstructive pneumonia (55). However, compressive atelectasis due to pleural fluid or traction phenomena within the region in patients with preexisting lung disease may alter and complicate the appearance of such patterns (56).
Question: What Are the Most Typical Findings of Radiation Fibrosis at Chest Radiography with or without Chemotherapy and What Changes Occur as a Function of Time after Such Therapy?— Changes after chemotherapy often result in reduction of the primary tumor and its metastatic lesions. RT often produces similar results, but it has become apparent that combination therapy (as performed in the presented case) is becoming more common (Fig 5). Therefore, the radiologic appearance after therapy may be more complex than with either modality alone. However, it is usually the radiation sequelae that present difficulty in analysis of such changes on routine chest radiographs and CT images.
Observations of radiation changes on routine chest images are influenced by a number of factors (57). Specifically, the volume of lung and other structures within the field (ie, hilum and mediastinum), the field shape, the total dose and time course of delivery, and other therapy given before or concomitant to the RT all influence what is observed on the chest radiograph. What presents the most difficulty for a radiologist is determining the evolution of the radiation changes as a function of time. Early in the course of such changes, pneumonitis without volume loss may be the only finding and is often seen about 8 weeks after completion of RT. Subsequent early volume loss with some bronchiectasis is usually seen approximately 3–4 months after completion of therapy, but it may take up to 9–12 months for radiation fibrosis and its associated volume loss within the lung to stabilize (57).
Although this question addressed the findings at routine chest radiography, such changes have a parallel appearance at CT (58). In fact, CT is often ordered in conjunction with routine chest radiography to follow the course of radiation changes as a function of time. Libshitz and Shuman (58) described four patterns of radiation-induced pulmonary changes on CT images, and it is important to understand such changes as a function of time (Fig 5e).
The future challenge for radiologists will be the interpretation of radiation changes as more sophisticated fields are used. Increasingly, three-dimensional conformal fields with higher doses are being applied to deliver more effect with less damage to vital structures. Further, studies are in progress to apply such therapy to lower-stage lung cancers within the TNM staging system to improve patient outcome (see the discussion of the radiation oncologist later in this article). Combination therapies are under study not only in the initial treatment of patients with lung cancer but also in cases of tumor recurrence. Therefore, in the future, the radiologist will be required to evaluate radiation effects in light of all these additional variables.
Pathologist’s View
Question: What Are the Characteristic Pathologic and Anatomic Features of Squamous Cell Carcinoma versus Other NSCLCs?—
Squamous cell carcinoma is a malignant epithelial tumor with the differentiating features of squamous epithelium: keratinization (individual or group keratinization with pearl formation), intercellular bridges, or both (Fig 6). Clinically manifested squamous cell carcinomas vary from small obstructive endobronchial tumors to large cavitated masses that replace an entire lung. Squamous cell carcinomas tend to be smaller than other lung carcinomas because they manifest earlier with obstructive symptoms. Squamous cell carcinomas tend to be central rather than peripheral. Because of the frequent involvement of large airways, exfoliated cells are more commonly identified in sputum cytology specimens in squamous cell carcinoma than in other types of lung carcinomas. Human papilloma virus has been found more frequently in squamous cell carcinoma than in other NSCLCs. Most patients with squamous cell carcinoma have a history of cigarette smoking, and most are men. Unlike other NSCLCs, squamous cell carcinoma is believed to progress from precursor conditions, such as squamous metaplasia, dysplasia, and carcinoma in situ.
|
|
Pretreatment staging remains the critical step prior to initiating therapy (59–62). History taking, physical examination, blood chemistry analysis, and radiologic staging—which may include chest radiography, CT of the chest to include the liver and adrenal glands, CT of the brain, and bone scanning—are routinely performed. Physiologic evaluation may include spirometry, ventilation and perfusion lung scanning for determination of split lung function, arterial blood gas studies, measurement of the diffusing capacity, and cardiac evaluation (echocardiography, electrocardiography, various stress tests). In patients with clinical stage I or clinical stage II tumors, CT of the brain and bone scanning are not routinely performed unless symptoms are present. Patients with advanced-stage lung cancer may undergo brain CT and bone scanning to exclude occult metastases even in the absence of symptoms, particularly when surgery or RT is anticipated.
CT of the chest permits evaluation of the location of the primary tumor, its relationship to the thoracic and mediastinal structures, and the size and extent of mediastinal lymph nodes. Any mediastinal lymph node greater than 1.0 cm in transverse diameter is considered enlarged. Histologic confirmation of metastatic involvement in enlarged lymph nodes is required prior to initiation of therapy. Typically, enlarged lymph nodes at CT of the chest cannot be considered involved unless disease is pathologically proved. A small percentage of small (
1 cm in diameter) lymph nodes may contain metastatic disease. Biopsy of these nodes is performed with mediastinoscopy, thoracoscopy, or fine-needle aspiration in all patients with identified mediastinal lymph nodes greater than 1 cm in greatest diameter (usually transverse) in whom surgery is planned.
In the patient with a right upper lobe cancer, pathologically confirmed metastasis to region 5 (aortopulmonary window) or region 6L (left anterior mediastinal) mediastinal lymph nodes (clinical stage IIIB) in the absence of extensive subcarinal adenopathy is extremely unlikely. However, region 4R (right paratracheal) lymphadenopathy may occur in 10% of patients with left lower lobe cancers. Left upper lobe cancers are unlikely to produce region 4R adenopathy in the absence of extensive subcarinal disease. Asymptomatic bone or brain metastases are also unusual. PET has been helpful in selected patients in determining the nature of suspicious primary lesions (recurrences after RT) or possible metastatic lesions. The use of PET is becoming more frequent; however, the sensitivity and specificity of this technique alone and in conjunction with other modalities (eg, CT of the chest, mediastinoscopy) are being determined.
Mediastinoscopy, extended mediastinoscopy, or video-assisted thoracic surgery may be used to evaluate enlarged mediastinal lymph nodes. Mediastinoscopy allows evaluation of level 2R, 2L, 4R, 4L, and 7 nodal stations. The aortopulmonary window (level 5) or anterior mediastinum (level 6) can be evaluated by using a left parasternal incision, the Chamberlain procedure, or extended mediastinoscopy anterior to the innominate artery. Video-assisted thoracic surgery allows evaluation of enlarged level 5 or 6 lymph nodes as well as level 8 or 9 or low level 7 lymphadenopathy.
Treatment decisions require accurate and complete surgical staging as an integral component of pulmonary resection for lung cancer. For postoperative treatment decisions, mediastinal lymphadenectomy determines the pathologic stage and provides information to the clinician regarding the patient’s prognosis and the need for postresection therapy. For a pulmonary resection, the mediastinal nodal stations listed in Table 2 should be inspected and dissected, and identified lymph nodes should be resected.
|
|
Pulmonary resection after chemotherapy and RT presents a number of technical challenges. Patients who have received in excess of 50 Gy during neoadjuvant treatment have a higher rate of serious complications, such as bronchopleural fistula, prolonged air leak with empyema, and prolonged postoperative ventilation; therefore, the dose should be limited to 45 Gy. The sequelae of chemotherapy and RT include dense fibrous tissue in the hilum and mediastinum. Tissue planes are obliterated (64). Rusch et al (65) reported the results of Southwest Oncology Group trial 8805, in which 50% of patients who received neoadjuvant chemotherapy and RT required pneumonectomy to achieve a complete resection compared with approximately 10% of patients who did not receive neoadjuvant chemotherapy and RT. At the time of resection, blood supply to the bronchus is preserved and all bronchial stump closures are reinforced with intercostal muscle, pericardium, or pericardial fat. Parietal pleura may be another alternative.
Advanced-stage lung cancer, particularly with nodal spread, typically cannot be considered a disease effectively treated with a single modality, whether that modality is surgery, chemotherapy, or RT. Only in rare patients will surgery alone be performed for advanced-stage lung cancer. Surgery alone for stage IIIA (N2), IIIB, or IV lung cancer is infrequently performed because the benefits of surgery frequently do not exceed the risks. The surgeon must balance the value of mechanical extirpation of the local disease (local disease control, pain relief, potential for improved survival) with the risks of a surgical procedure and potential improvement in survival or quality of life. Typically, the risks exceed the benefits and surgery is not considered; however, in some patients, surgery for advanced-stage lung cancer may be beneficial in terms of local tumor control, palliation of symptoms, improved quality of life, and potential for improved survival. Resection of an isolated brain metastasis followed by whole brain RT is warranted for improvement in quality of life and survival (66). The primary tumor is then treated according to the T and N stage.
Medical Oncologist’s View
Question: What Is the Role of Induction Chemotherapy in Early-Stage Resectable NSCLC?—
On the basis of the favorable results of studies of advanced-stage disease, application of chemotherapy in earlier stages of lung cancer may improve survival. Pisters and colleagues (67) conducted a phase II trial to assess the feasibility (response, toxicity, resectability, and morbidity) of perioperative paclitaxel and carboplatin therapy in patients with stage IB, IIA, IIB, or T3 N1 NSCLC. Mediastinoscopy results were negative for N2 disease in all patients. Chemotherapy consisted of paclitaxel (225 mg/m2 in a 3-hour infusion) and carboplatin with area under the concentration - x time curve (AUC) = 6 every 21 days for two cycles prior to surgery. Patients with a complete response at pathologic analysis received an additional three cycles. Patients with postoperative N1 or N2 disease were excluded from the study. Ninety-four patients were entered in the study. A major response (complete or partial) occurred in 50 patients (54%). Eighty-three patients (90%) underwent surgical exploration, and 75 (82%) underwent complete resection. Two postoperative deaths occurred; four patients had complete responses at pathologic analysis. The authors concluded that induction chemotherapy is feasible and has a high response rate (67). A prospective randomized intergroup trial comparing surgery alone with induction chemotherapy and surgery for early-stage lung cancer (stage IB, IIA, IIB, or IIIA [T3 N1]) is ongoing.
Most patients with histologically confirmed N2 disease have a biologically aggressive tumor with probable occult metastatic disease. Although pulmonary resection and mediastinal lymphadenectomy can provide some patients with improved survival and enhanced local control, most patients will not benefit from surgery as the sole modality for treatment of pathologic stage IIIA NSCLC. As Rosell et al (68) and Roth et al (69) have shown, neoadjuvant therapy (platinum based) prior to surgery for pathologic stage IIIA (N2) disease improves survival compared with surgery alone. Preoperative chemotherapy does not appear to increase surgical morbidity in patients who receive neoadjuvant chemotherapy. Recent studies suggest that there is increased morbidity and mortality with chemotherapy and surgery compared with surgery alone (64).
Question: What Is the Role of Induction Chemotherapy in Locally Advanced Unresectable NSCLC?— The rationale behind the addition of chemotherapy to RT is to reduce the risk of distant failure by means of sterilization of occult extrathoracic disease with chemotherapy. If chemotherapy can produce radiosensitization, it may also enhance local control. There are two general strategies for combination chemotherapy and RT: (a) sequential chemotherapy followed by RT with use of high-dose induction chemotherapy or (b) concurrent chemotherapy and RT with use of either low-dose or high-dose induction chemotherapy.
There are three positive trials of sequential chemotherapy followed by RT: the French, Cancer and Leukemia Group B, and RTOG 88-08 trials (70–72). In these trials, the natural history of the disease was altered, with median survival around 15 months, 2-year survival around 20%, and 5-year survival around 15%. At failure pattern analysis, these improvements appeared to be due to deterred progression of micrometastases. Aggressive hyperfractionated RT alone produced similar long-term survival (72).
There are three positive, large, randomized trials of concurrent chemotherapy and RT, one of which used standard RT and two of which used hyperfractionated RT (71–73). The European Organization for Research and Treatment of Cancer trial showed that, in patients with a favorable performance status, low-dose daily cisplatin therapy and standard RT was superior (2-year survival of 32%) to standard RT and weekly cisplatin therapy and to RT alone (73). Failure pattern analysis showed that this improvement was due to better control of intrathoracic tumor. Several phase II trials of concurrent chemotherapy and hyperfractionated RT have shown further increases in median survival (around 19 months) and 2-year survival (35%) (74–76). However, acute toxicity is significant with these regimens, with grade 3 or higher-grade esophagitis in 35%–45% of patients. A recent Japanese phase III trial that compared concurrent and sequential chemotherapy and RT had results that favored the concurrent regimen (77).
The goal of current phase III studies, including those of the RTOG, is to refine the timing and sequencing of chemotherapy and RT to obtain the benefits of both concurrent and sequential approaches with acceptable toxic effects on normal tissue. The recently completed RTOG 94-10 trial compared sequential and concurrent chemotherapy and RT and was designed to assess the risk-benefit ratios of the two strategies (78). It also compared concurrent chemotherapy with both standard RT and hyperfractionated RT. Although early follow-up confirmed that acute esophageal toxicity is increased with both concurrent regimens compared with sequential chemotherapy and RT, no survival results are available as yet.
Radiation Oncologist’s View
Question: What Are the Most Important Prognostic Factors for Locally Advanced NSCLC?—
The presented case is typical regarding age, smoking history, and presentation for patients with locally advanced NSCLC. Fortunately, the patient had a good performance status and no weight loss, allowing aggressive treatment.
Question: Is There Any Progress in the Outcome of Unresectable but Locally Advanced NSCLC?— The continuing research challenge for unresectable NSCLC has been to optimize the nonoperative treatment strategy. One general approach has been to investigate ways to increase the dose intensity of RT by using both conventional and altered fractionation schedules. A second general approach has been to improve systemic control by the addition of systemic cytotoxic chemotherapy. The observation that response rates to chemotherapy were higher for locally advanced tumor than for metastatic disease led to incorporation of chemotherapy with both standard and altered fractionation RT schedules. More recently, a third approach has been to explore the potential for small-field, high-dose RT by using three-dimensional, conformal RT. Each of these approaches is briefly discussed in the Conclusions section.
|
Case 3 |
|---|
|
TopIntroductionCase 1Case 2Case 3ConclusionsReferences |
|---|
Physical examination revealed his weight to be 59 kg without recent weight loss. His score on the Karnofsky performance scale was 90. He had Horner syndrome on the left side with an ipsilateral narrowing palpable fissure, miosis, and anhidrosis. There was a palpable left supraclavicular lymph node that measured 1 cm in diameter. He had radiating pain from the left shoulder to the ulnar distribution of the left upper extremity. Otherwise, results of physical examination of the chest, abdomen, neuromuscular system, and CNS were unremarkable.
The patient underwent a complete blood cell count, measurement of electrolyte and calcium levels, CT of the chest and upper abdomen, and CT of the brain. Bone scans were negative. CT (Fig 8a) and multiplanar MR imaging (Fig 8b, 8c) of the chest showed an invasive sulcus tumor. Results of a needle aspiration biopsy were negative, but subsequent open biopsy revealed poorly differentiated adenocarcinoma (Fig 9).
|
|
|
|
weeks according to the in-house protocol. The chemotherapy was given every 28 days for four cycles. The patient also received PCI (250 cGy per fraction in 10 fractions), which he tolerated well. During the thoracic RT with concurrent oral etoposide and cisplatin therapy, the patient developed grade 3 esophagitis, which required percutaneous endoscopic gastrostomy. He regained weight back to the normal level in the 6 months after completion of treatment. He is free of disease 6 years after treatment and is working as a computer specialist without any side effects.
Diagnostic Radiologist’s View
Question: What Are the Most Useful Imaging Studies for Evaluation of Superior Sulcus Tumors?—
MR imaging appears to be the most useful imaging study in evaluation and staging of sulcus tumors primarily because of its multiplanar capability and ability to show certain structures to greater advantage than other modalities (Fig 8) (79). Certainly, routine chest radiography remains the initial imaging study within the work-up, although thoracic CT, brain imaging, and bone scanning remain part of the complete evaluation. Specific advantages of MR imaging include (a) evaluation of the brachial plexus, (b) demonstration of invasion of cardiovascular structures (eg, the subclavian artery), and (c) no need for iodine contrast material (80).
Invasion by sulcus tumors into the chest wall and structures in the immediate vicinity can be quite extensive, but in many cases the overall tumor does not metastasize more distantly. If there is no evidence of such spread, extensive surgical resection of the tumor and surrounding involved structures can be performed, including resection of ribs, reconstruction of vascular structures, and, if necessary, vertebrectomy (see the surgical oncologist’s discussion later in this article). Therefore, it is important for the radiologist to fully evaluate the overall anatomy of the area of involvement prior to such surgical procedures; it appears that MR imaging is the best overall imaging modality for such evaluation.
Pathologist’s View
Question: Are There Any Characteristics of Lung Cancer Originating in a Superior Sulcus Location?—
The term superior sulcus tumor (Pancoast tumor) refers to any primary lung cancer that manifests as constant pain in the nerve distribution of the eighth cervical and first and second thoracic nerve roots and as Horner syndrome caused by invasion of the stellate ganglion. Pulmonary symptoms are less frequent because of the peripheral location of the lesion. At morphologic analysis, superior sulcus tumors can demonstrate any histologic type of lung cancer, but squamous cell carcinoma and non–squamous cell carcinoma (adenocarcinoma and undifferentiated large cell carcinoma) are equally distributed. Attar et al (81) reported that squamous cell carcinomas made up 39% of the superior sulcus tumors in their study, and Strojan et al (82) reported a similar result.
Question: What Is the Subclassification of Adenocarcinoma?— Adenocarcinoma is a malignant epithelial tumor with gland formation or mucin production by the tumor cells (Fig 10). Adenocarcinomas are classified as acinar, papillary, bronchioloalveolar, solid adenocarcinoma with mucin, and adenocarcinoma with mixed histologic features. The mixed subtype is the most frequent adenocarcinoma encountered in routine practice. There is usually a mixture of the acinar, papillary, bronchioloalveolar, and solid with mucin patterns, but any of these may be seen in a pure form. Rare variants of adenocarcinoma include well-differentiated fetal adenocarcinoma, mucinous adenocarcinoma, mucinous cystadenocarcinoma, signet-ring adenocarcinoma, and clear cell carcinoma.
|
Surgical Oncologist’s View
Question: What Is the Role of Surgery in Pancoast Tumors and Other Superior Sulcus Tumors?—
Pancoast syndrome consists of shoulder and arm pain, Horner syndrome, and weakness of the hand. This group of symptoms is typically caused by an NSCLC that enlarges and compresses the structures of the ipsilateral thoracic inlet. The cupola of the lung extends above the thoracic inlet. Superior sulcus tumors compress the lower trunks of the brachial plexus (C8 and T1 nerve roots). They account for fewer than 5% of all bronchogenic carcinomas.
Most patients present with shoulder or arm pain and are treated unsuccessfully for several months for nonspecific musculoskeletal problems prior to the correct diagnosis being made with plain radiography of the shoulder (which includes the apex of the lung) or chest radiography. Horner syndrome does not consistently occur in patients with superior sulcus tumors. These tumors are apical and therefore peripheral. Obstructive symptoms such as cough, sputum production, or hemoptysis are infrequent. A diagnosis is infrequently obtained with bronchoscopy because of the peripheral location of the lesion; percutaneous biopsy, performed under fluoroscopic or CT guidance, is the most sensitive and specific diagnostic test.
The results of therapy are often unclear because of incomplete information. In the past, a histologic diagnosis was not consistently obtained prior to treatment (83–85). Other diseases may mimic a superior sulcus tumor from non–small cell lung carcinoma, and obtaining a diagnosis prior to treatment is necessary.
Superior sulcus tumors are staged on the basis of standard criteria (86). CT is performed for evaluation of the lung parenchyma (for other lesions), and MR imaging is performed for evaluation of the mediastinum, thoracic inlet, subclavian vessels, and brachial plexus.
Additional surgical staging such as mediastinoscopy for mediastinal lymph node sampling prior to the definitive procedure may allow identification of obvious or occult N2 nodal metastases. Although bone scanning and CT or MR imaging of the brain are not typically performed in patients with resectable tumors in the absence of symptoms, distant metastases in these areas should be excluded given the higher prevalence of brain metastases from superior sulcus tumors.
Tumors that involve only the ribs are considered stage IIB (T3 N0 M0) (19). Resectable tumors that involve the ribs and vertebral bodies are considered stage IIIB (T4 N0 M0). Tumors that involve mediastinal lymph nodes (any T N2 M0) are considered stage IIIA and may be treated with definitive chemotherapy and RT. These patients may also benefit from preoperative chemotherapy and RT followed by surgery.
RT (30 Gy) and surgery have become one (of several) standards of care in the United States. One of the first studies of combined RT and surgery was performed in 18 patients, and this strategy may be the most common treatment of superior sulcus tumors (83,84). RT may shrink the tumor somewhat and provide a surgical margin of dead tissue. In theory, preoperative RT limits the number of viable cells "spilled" at resection and improves local control for the resection attempt itself. RT may have no significant benefit for patients with superior sulcus tumors in terms of survival on the basis of a stage-specific comparison with patients with other non–small cell carcinomas (87).
However, no prospective randomized studies have been performed to compare this treatment with other treatment methods (86). Local control may be compromised with this technique. A major criticism of this approach is that the local control modality of RT is used in an inadequate and incomplete manner. Preoperative RT given to the maximal extent (54–65 Gy) may optimize the radiation effects but may compromise the resection of the tumor. At the M.D. Anderson Cancer Center, surgical resection is the initial modality used and includes rib resection or vertebral body resection with reconstruction. RT is then used to maximal effect after resection of the tumor.
Anatomic resection remains the treatment of choice for patients physiologically capable of undergoing this procedure (88). A posterior thoracotomy incision is made and extended to the base of the neck. The trapezius and rhomboid muscles are divided, and the scapula is elevated to expose the upper ribs. Resection of these ribs and the tumor en bloc is performed (89). Resection of ribs 1, 2, and 3 typically can be accomplished with straightforward thoracic surgical techniques; chest wall reconstruction is not typically needed. Resection of the vertebrae has been successfully accomplished in selected patients (90,91). Resection of a portion of the vertebral body with neurosurgical techniques and rigid reconstruction or stabilization allow enhanced local control with negative vertebral (bone) margins.
For superior sulcus tumors involving the anterior portion of the thoracic inlet, the subclavian vessels, or the brachial plexus, an anterior approach may be used. Successful resection and reconstruction of the subclavian vessels have been performed. The anterior approach removes or elevates the clavicle and first rib to enhance exposure of the tumor and the structures of the thoracic inlet (92,93).
Medical Oncologist’s View
Question: Is There a Role for Induction Chemotherapy and RT Given Concurrently If the Superior Sulcus Tumor Is Marginally Resectable?—
Rates of complete resection (50%) and 5-year survival (30%) for superior sulcus tumors have not changed for 30 years. However, combined modality therapy has improved outcome in other stage III NSCLCs. A prospective intergroup trial tested the feasibility of concurrent induction chemotherapy and RT and surgical resection in mediastinoscopy-negative superior sulcus tumors with the ultimate objective of improving resectability and overall survival (94). Patients with pathologically proved T3–4 N0–1 disease received two cycles of cisplatin and etoposide therapy concurrently with 45 Gy of RT. In patients with no evidence of disease progression, thoracotomy was performed 3–5 weeks later. Two cycles of chemotherapy were given postoperatively. There were 116 patients entered in the study. The analysis included 101 eligible patients, 71 men and 30 women with a median age of 56 years. Induction therapy was completed as planned in 93% of patients, with one case of grade 5 toxic effects and 15 cases of grade 4 toxic effects (predominantly cytopenia). To date, 81 patients have undergone thoracotomy, with the most common procedure being lobectomy plus chest wall resection. One patient died postoperatively. A complete response at pathologic analysis occurred in 57.5% of the patients, and 63% of the tumors were downstaged. At 1 year, the overall survival rate was 77% for T3 tumors and 80% for T4 tumors; at 3 years, the overall survival rate was 50% for both T3 and T4 tumors. The most common site of relapse was the brain. The investigators concluded that (a) this combined modality approach is highly feasible in a multi-institutional setting; (b) the rate of complete responses at pathologic analysis was unexpectedly high; (c) resectability and overall survival were improved compared with those of historical controls; and (d) improved outcome was especially notable for patients with T4 tumors, which usually have a grim prognosis.
Radiation Oncologist’s View
Question: How Do You Approach a Case of Marginally Resectable Superior Sulcus Tumor?—
Superior sulcus tumor is essentially a subset of NSCLC based on the location of origin. It is a relatively infrequent form of NSCLC, making up only 3% of all lung cancers. The two most common treatment approaches for potentially resectable disease are preoperative RT followed by surgery and surgery followed by postoperative RT with or without chemotherapy for patients with positive nodes or positive surgical margins. These two treatment options have not been compared in phase III trials. Various investigators have championed one approach or the other, despite the absence of comparison of the two approaches. It is useful to examine the background behind these two strategies.
Prior to the development of CT and MR imaging, patients with apparently resectable disease at initial staging were often found to have unresectable disease at surgical exploration. For these patients, RT alone, usually to 60 Gy in 6 weeks, yielded an unsatisfactory 5-year survival rate (5%–7%). In the early 1980s, use of preoperative RT in superior sulcus tumors was investigated. It was thought that preoperative RT could potentially reduce the bulk of the tumor, sterilize microextensions of the primary tumor, and render more tumors resectable. To avoid increasing surgical morbidity, the RT dose was moderate (30–50 Gy in 2–5 weeks).
The larger studies used moderate-dose preoperative RT and surgery and reported 5-year survival rates of 25%–35% (95). Unfortunately, these series often excluded from analysis patients who did not undergo the planned surgery or who died of preoperative complications. Reported survival rates were lower when all patients were included. Before surgery, the true pathologic stage can be estimated only from available imaging results. However, ultimately, the only patients who can undergo complete resection after preoperative RT are those with negative nodes and limited local extension, whether that status was afforded by the RT or existed at presentation. Thus, T2–3 N0 would be the most likely postoperative pathologic stage for these patients. When the location of the primary tumor is not considered, published 5-year survival rates after surgery alone for these stages are 38%–57% (19,96). There is little reason to expect that tumors in the superior sulcus would have a surgical outcome different from those of lung tumors in other locations with equivalent pathologic stages, if it is assumed that complete resection can be done. Similarly, there is no evidence that preoperative RT benefits patients with clearly resectable superior sulcus tumors (97). Preoperative chemotherapy has been tested in non–superior sulcus resectable NSCLCs with the results suggesting a survival improvement, but this approach has not been tested in superior sulcus tumors (69).
The recent improvement in long-term survival noted for inoperable NSCLC by the addition of chemotherapy to RT led to trials of preoperative chemotherapy/RT for NSCLC (65). An intergroup phase II trial of preoperative cisplatin/etoposide concurrent with RT for potentially resectable superior sulcus tumors is nearing completion. The additional tumor cell kill provided by the added chemotherapy may improve local-regional response and overcome accelerated repopulation sufficiently to both facilitate surgery and control occult extrathoracic disease. It is hoped that the study outcome will confirm improved survival.
Patients with clearly unresectable disease at imaging should be treated similarly to patients with NSCLCs in non–superior sulcus sites, namely, with RT plus chemotherapy. The results of treating unresectable superior sulcus tumors with RT alone are reasonably good, with the predominant failure mode being local, although brain metastases seem to occur with greater frequency when the primary tumor is in the superior sulcus than when the primary tumor is elsewhere in the lung (98). When standard doses of RT alone are used, an overall 5-year survival rate of 28% has been reported; in those patients in whom local control was achieved, the 5-year survival rate was 52% (99). Higher RT doses or more intense fractionation schedules might improve these survival figures, especially if RT were combined with chemotherapy (71,75,100). Biologically more effective doses than 60 Gy in 6 weeks can be given safely to patients with superior sulcus tumors by using modern treatment planning computer systems and three-dimensional conformal approaches. The use of these techniques plus altered or standard fractionation can potentially deliver doses up to 90 Gy to a superior sulcus tumor with a higher likelihood of local tumor control. This approach needs to be further tested to determine if it can be performed with acceptable toxicity. Of greatest concern are brachial plexus and pulmonary toxic effects.
Treatment of superior sulcus tumors is an example of a treatment strategy based on accurate staging, an understanding of tumor failure patterns, and the biologic response of the tumor to RT and chemotherapy.
Question: Do You Have Any Concern regarding Brachial Plexopathy When High-Dose RT Is Given to a Superior Sulcus Tumor?— Peripheral nerves have a very high tolerance for ionizing radiations; they are infrequently damaged even with high total doses. There are no recognized acute or subacute effects. Late effects result from marked inhomogeneities of dose with very high areas or are seen after altered fractionation with large fraction sizes.
Stoll and Andrews (101) reported neurologic symptoms and signs referable to the brachial plexus in a high proportion of patients who underwent irradiation of the supraclavicular region with 4-MV photons. However, patients with involvement of the brachial plexus by regional metastasis were not consistently excluded. When stricter criteria for radiation-related damage to the brachial plexus were used, Thomas and Colby (102) found injury in only 14 of 1,202 patients treated (1.2%). As is the case with late effects in other parts of the nervous system, there is no recognized treatment for radiation-related injury.
Use of CT to avoid an excessive volume of brachial plexus within the irradiation field will help minimize the risk of plexopathy. Fractionated RT (2 Gy per fraction) can be applied by homogeneous distribution of the total dose up to 72 Gy without damage to the brachial plexus (99).
|
Conclusions |
|---|
|
TopIntroductionCase 1Case 2Case 3ConclusionsReferences |
|---|
Questions based on these and other issues will define clinical research on limited SCLC until more effective systemic agents are developed. Can the RT in a chemotherapy-RT program be thought of as a nonsystemic, local-regional cytotoxic agent and be given in cycled intermittent courses during chemotherapy without the reduction in tumor control rates associated with split-course RT? If effective chemotherapy is given, does the RT need to prophylactically treat nodal sites that were uninvolved at clinical evaluation? Does concurrent use of chemotherapy and either standard or hyperfractionated RT allow a reduction in total dose for equivalent local-regional control? Recently completed and ongoing phase III trials will clarify the answers to these and other questions.
The discussion of case 2—a patient with marginally resectable NSCLC—presented the pros and cons of the role of surgery in patients with medical problems as well as the uncertainty of complete surgical resection and complete response to chemotherapy and RT. The combination of surgery and RT provides benefit as a local control modality. However, most patients succumb to local or regional disease and, even more frequently, to systemic disease. Today, the most effective treatment for advanced-stage lung cancer is multimodality therapy. Surgery provides good local control that may be enhanced by neoadjuvant therapy in early- and advanced-stage disease. Improved survival awaits more effective systemic control of micrometastases. Significant improvements in survival for unresectable NSCLC have been realized by the addition of chemotherapy to RT, with recent results favoring concurrent chemotherapy and RT. Improvements in RT dose delivery and the use of altered fractionation RT have been attended with more limited success. However, there appears to be a greater potential for improving local control and survival by using three-dimensional conformal RT techniques to substantially higher doses than can be given with standard RT techniques, with relative sparing of normal tissue.
The patient in case 3—a superior sulcus tumor that caused Horner syndrome (Pancoast tumor) with regional nodal metastases—was treated successfully with a multidisciplinary approach. Knowledge of patterns of failure will bring up prevention, such as removal of gross tumors, PCI, and chemotherapy as a radiosensitizing agent and systemic treatment to eliminate micrometastases. Successful treatment of patients with lung cancer depends on recognition of pathologic findings, the local and distant extent of disease, the patient’s medical condition and medical history, and a multidisciplinary approach with experienced surgical, medical, and radiation oncologists.
|
Footnotes |
|---|
Abbreviations: CNS = central nervous system, NSCLC = non–small cell lung cancer, PCI = prophylactic cranial irradiation, RT = radiation therapy, RTOG = Radiation Therapy Oncology Group, SCLC = small cell lung cancer
|
References |
|---|
|
TopIntroductionCase 1Case 2Case 3ConclusionsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
