Evaluation and Definitive Management of Medically Inoperable Early-Stage Non-Small-Cell Lung Cancer
Evaluation and Definitive Management of Medically Inoperable Early-Stage Non-Small-Cell Lung Cancer
Lung cancer is estimated to be the second most commonly diagnosed cancer in both men and women in 2006, and the leading cause of cancer mortality. Non-small-cell histologies represent the majority of cases. Despite clinical investigation into screening high-risk populations, most patients have locally advanced disease at presentation and are not eligible for curative resection. For the fewer than 20% of patients with stage I or II disease, and some portion of those with stage III, surgery is the treatment of choice. The 5-year overall survival rates of patients managed with primary surgery in the modern era can be predicted as a function of clinical staging criteria: 61%, 38%, 37%, and 24% for stages IA, IB, IIA, and IIB, respectively.
While surgical resection with pathologic nodal staging remains the standard of care in patients with early-stage disease, the high rate of comorbid medical illness in this population often raises concern about perioperative morbidity, postoperative pulmonary function, and long-term quality of life. An evidence-based multidisciplinary evaluation of patient age, cardiovascular health, and baseline pulmonary function can accurately predict which patients may benefit from lobectomy.
In the absence of a curative surgical option, many patients and physicians appropriately opt for either a palliative or an observational approach, but there are a substantial number of patients for whom a definitive, nonsurgical approach is appropriate. To date, definitive radiotherapy has been the most commonly employed regimen, based on data suggesting a modest survival benefit.
McGarry et al analyzed the outcomes of 128 patients with stage I/II non-small-cell lung cancer (NSCLC), 47 of whom received no treatment. The median survival time with observation was 14.2 months, and the cause of death was cancer in 53% of the cases. Patients treated with radiotherapy in this series, with either palliative or curative intent, had significantly longer median survival, implying that a nonsurgical option conferred a survival benefit. Chadha et al similarly reported a relatively poor median survival of 11.9 months (13.7 months for stage I and 8.4 months for stage II) for untreated early-stage NSCLC. Again, the most common cause of death was progressive disease, either local or metastatic.
The largest evaluation of the utility of radiation was conducted from a population-based registry by Wisnivesky et al. The authors evaluated 4,357 patients diagnosed with stage I or II NSCLC who did not undergo surgical resection. Median survival was improved in patients treated with radiotherapy, although 5-year survival was not significantly different. Of note, the dataset did not distinguish between definitive and palliative radiation treatment courses.
For patients ineligible for curative resection, conventional single-modality radiotherapy has been the primary definitive option. Numerous retrospective reports demonstrate long-term disease-free and overall survival data that are modestly superior to that expected after observation, but both local and distant failure continue to be significant risks. Ongoing trials of dose escalation may improve local control, and the addition of systemic therapy may help to decrease metastatic failure. Additionally, emerging evidence suggests that new modalities, such as stereotactic radiosurgery and radiofrequency ablation (RFA), may offer curative treatment alternatives. These options will be discussed in the concluding part of this article, which will appear in the July issue of ONCOLOGY.
Defining "operability" in patients with lung cancer often presents a significant clinical challenge. Surgery is the treatment of choice for patients with stage I or II NSCLC. Recent studies support the use of postoperative adjuvant chemotherapy as well for treatment of early-stage NSCLC, but resection remains the primary therapeutic modality and is associated with the best long-term outcomes.[6-10] Many factors may contribute to the determination of whether an individual is suitable for lung resection. The factors that most commonly cause concern regarding morbidity or mortality following a surgical procedure are older age, the presence of significant cardiovascular risk, and the presence of underlying pulmonary disease.
Age is increasingly a consideration in lung cancer treatment. Lung cancer generally affects an older population. Data from the National Cancer Institute Surveillance, Epidemiology and End Results (SEER) program indicate that from 1998 to 2002, the median age at diagnosis of lung cancer was 70 years. The most recent SEER data show that at initial diagnosis of lung cancer, 33.1% of patients were between ages 65 and 74, 27.9% were between ages 75 and 84, and 6.9% were age 85 years and older. As the population ages, an even larger number of patients will predictably fall into older age groups.
That said, an expanding body of evidence shows that age per se should not be a contraindication to surgery. Two recent single-institution retrospective series of pulmonary resections in octogenarians reported by Brock and colleagues (N = 68) and Port and colleagues (N = 61) noted 5-year survival rates in patients with resected stage IA NSCLC of 61% and 82%, and 30-day mortality rates of 8.8% and 1.6%, respectively.[12,13] With proper preoperative evaluation of functional status (such as the ability to perform activities of daily living), comorbidities, and cognitive function, it is clear that appropriately selected elderly patients can safely be offered curative surgical resection.[12-14]
Patients with lung cancer are often at higher risk of cardiovascular disease because of shared risk factors, including cigarette smoking and older age. Since surgery for resection of lung cancer is rarely done emergently, preoperative cardiac evaluation should be possible in almost all patients. Perioperative cardiovascular risk assessment for noncardiac surgery, including thoracic surgery, has been studied extensively.
Joint evidence-based practice guidelines from the American College of Cardiology and American Heart Association have been available since 1980, with the most recent update published in 2002. These guidelines make the important point that the purpose of the preoperative evaluation is not to merely grant medical clearance for surgery, but to assess the need for further preoperative testing, to plan for management of the patient's cardiac needs during and after surgery, and to guide treatment decisions.
Intrathoracic surgery falls into the category of intermediate cardiac risk, with the reported overall risk of cardiac complications usually less than 5%. Clinical predictors of increased perioperative risk of myocardial in-farction, heart failure, or death related to cardiac causes have been well described. The presence of unstable coronary syndromes (acute or recent myocardial infarction or unstable angina), decompensated congestive heart failure, high-grade arrhythmias, or severe valvular disease may delay lung cancer surgery until appropriate evaluation and planning for cardiac management can be determined. Patients whose evaluations raise issues about limited life expectancy related to underlying cardiovascular disease or in whom intrathoracic surgery is deemed of unacceptable cardiovascular risk should be evaluated by appropriate specialists before a decision to deny surgery is made.
As with cardiovascular disease, older age and a high prevalence of cigarette smoking increase the risk for concomitant pulmonary disease. Chronic obstructive pulmonary disease (COPD) related to cigarettes is most commonly associated with lung cancer, but other lung diseases such as pulmonary fibrosis and asbestosis appear to contribute to increased risk of lung cancer as well.
Many patients with lung cancer have abnormal pulmonary function. The challenge of determining operability for these patients dates to the early days of thoracic surgery. In 1955, Gaensler and colleagues addressed the risk of respiratory failure and death related to chest surgeries in patients with severe pulmonary tuberculosis. Their landmark study was the first to report surgical outcomes in relation to preoperative pulmonary physiologic measurements (vital capacity and maximal breathing capacity) and to define thresholds for operability based on such measurements.
In the 50 years that have passed since their observations, physiologic measurements have remained the cornerstone of operative risk assessment in patients undergoing lung resection. Those measurements typically used to determine resectability include absolute and percent predicted values of forced expiratory volume in 1 second (FEV1) and diffusing capacity of the lung for carbon monoxide (DLCO), as well as exercise capacity. The latter is usually expressed as maximal oxygen consumption (VO2max) measured during formal cardiopulmonary exercise testing.
A number of algorithms using these measurements to identify patients who can safely undergo lung resection have been proposed.[17-21] A summary of recommendations relating to assessment of resectability from the evidence-based guidelines for lung cancer proposed by the American College of Chest Physicians is outlined in Figure 1.
• Assessment ConsiderationsIt is widely accepted that patients with absolute FEV1 > 2 L are candidates for pneumonectomy, and those with FEV1 > 1.5 L, candidates for lobectomy without further physiologic testing. These recommendations are based on older series done largely in men, with reported operative mortality < 5%. These studies typically did not report FEV1 as a percentage of predicted normal value, but a normal FEV1 (> 80% predicted) is also accepted as a criterion for resectability.[18,22] Since cigarette smoking is so common a factor in lung cancer, obstructive airway disease is typically the accompanying pulmonary disease, with measurement of FEV1 providing a reasonable means of assessing severity.
In patients with evidence of interstitial lung disease on radiographic evaluation or in whom dyspnea is a prominent symptom, DLCO is also a useful measurement. Impairments in DLCO correlate with increased surgical morbidity as well as with worse quality of life after resection.[23,24] Patients who have normal FEV1 and DLCO (ie, both > 80% predicted) should be considered suitable candidates for resection, including pneumonectomy, without further pulmonary evaluation.
The demographics of lung cancer have changed considerably over the past several decades, with women and older persons now comprising a larger percentage of patients. Persons who are female, older, of certain ethnicities (including African or Asian descent), and who are of smaller stature will have smaller lung volumes at their normal baseline. With this consideration, using percent predicted values of FEV1 and other physiologic measurements rather than absolute values are in general more reliable in assessing lung function.
In patients who have either abnormal FEV1 or DLCO, further evaluation is necessary to determine whether resection can be performed with acceptable operative mortality and postoperative morbidity and quality of life. Predictions of postoperative values of FEV1, DLCO, and VO2max require an estimation of how much lung function will be lost with resection. The usual means of measuring "split lung" function is by radionuclide quantitative lung scanning to assess perfusion to different sides and areas of the lung in combination with measurements of FEV1, DLCO, and VO2max. Predicted postoperative FEV1 (FEV1ppo) is calculated as follows:
FEV1ppo = preoperative FEV1 X (1 - fractional contribution of lung to be resected as estimated by lung perfusion scanning)
Several studies have demonstrated reliable correlation of predicted and measured postoperative lung function using this method.[25,26] Notably, there is evidence that patients undergoing lobectomy typically have more recovery of pulmonary function (measured as both FEV1 and VO2max) within 3 to 6 months after surgery than would be predicted by split lung function prediction, so that using this method, if anything, errs on the side of patient safety.[22,27,28]
The lower limit of acceptable FEV1ppo remains controversial. Typically, FEV1ppo of 40% of the predicted normal is felt to be the minimum threshold, although some groups have suggested that 30% may also be acceptable.[29,30] Similarly, percent predicted postoperative DLCO (DLCOppo) of < 40% appears to correlate with increased perioperative complications and worsened postoperative pulmonary quality of life.[23,27,31] Patients with both FEV1ppo and DLCOppo < 40% would generally be felt to have a high risk of surgical morbidity and postoperative severe pulmonary impairment.[18,22] For some patients in this group, further physiologic assessment with cardiopulmonary exercise testing may be warranted to assess whether resection can still be an option. Alternatively, nonsurgical therapeutic modalities should be considered.
• Cardiopulmonary Exercise TestingExercise capacity can be estimated by simple stair climbing, or may be measured with formal cardio-pulmonary exercise testing (CPET). Stair climbing has been used for years as a measure of operability, although the number of stairs required to predict successful surgical outcome has been variably reported. Olsen and colleagues reported that to achieve successful surgical outcome a patient needed to climb at least 76 stairs for lobectomy and at least 100 steps for pneumonectomy. Pollock and colleagues reported that 83 stairs would be acceptable for pneumonectomy, and correlated this amount of effort with a VO2max of 20 mL/kg/min.
Formal cardiopulmonary exercise testing may be less readily available than stair climbing, but offers the ability to measure VO2max in a more standardized fashion. Both the British Thoracic Society and the American College of Chest Physician guidelines for lung cancer evaluation and treatment recommend that CPET be performed to measure VO2max if FEV1ppo and DLCOppo are < 40%.[17,22] Based on a number of studies correlating VO2max with operative risk, it is generally accepted that patients with preoperative VO2max > 20 mL/kg/min can undergo pneumonectomy without increased risk of perioperative complications, while VO2max < 15 mL/kg/min is associated with an increased risk of perioperative complications.[31,34-42] Further, patients with preoperative VO2max < 10 mL/kg/min are at very high risk of perioperative complications even with lobectomy only.[18,35,40,43] Patients with VO2max in this range will need careful evaluation before a decision regarding resectability can be made.
As with measurements of FEV1, concern has been raised about predictions based on absolute values of VO2max, as women and persons of older age and shorter stature might have normal predicted values of VO2max < 15-20 mL/kg/min. Several recent studies suggest that percent predicted VO2max is a better predictor of surgical outcome than absolute preoperative VO2max.[36,37,41,44] These studies suggest that perioperative complications are substantially increased in patients with percent predicted VO2max < 50%-60%, but that patients who have exercise capacity above this threshold can undergo surgery with reasonable safety.
The issue of resectability in patients whose physiologic evaluation raises concerns about perioperative risk and postoperative pulmonary compromise clearly can be very challenging. Such patients should undergo evaluation by a multidisciplinary team at a center with the necessary expertise in pulmonary medicine and thoracic surgery before a final decision regarding the feasibility of resection is made.
It should be noted that alternatives to classical lobectomy and pneumonectomy may be a consideration for individual patients. Lung-sparing surgeries such as segmentectomy and wedge resection may be reasonable in patients with severely diminished pulmonary function, and may offer such patients good long-term outcomes.[45-47] Experience in patients with severe emphysema undergoing lung volume reduction surgery has shown that carefully selected patients with very poor lung function can safely undergo thoracotomy and may have functional benefit from removal of severely emphysematous regions of lung.[48,49] In some cases, lung cancers contained within such areas can also be resected, even though the physiologic parameters outlined in Figure 1 are not met.[50,51] However, these patients should be very carefully selected, with evaluation performed at centers with specific expertise in lung cancer and lung volume reduction surgery.
The question of what constitutes acceptable risk for potentially curative surgery remains extremely difficult to answer. Treatment evaluation in patients who have substantial risk for perioperative complications and postoperative compromise of quality of life related to limited pulmonary reserve should include careful consideration of alternatives to surgery. As with other aspects of lung cancer management, decisions relating to treatment for patients with impaired lung function should involve multidisciplinary input from an experienced team of specialists in relevant disciplines, including pulmonary medicine, thoracic surgery, medical oncology, and therapeutic radiology.