Stage IV Disease
For patients with stage IV NSCLC, chemotherapy plus supportive care improves both survival and quality of life, compared with supportive care alone. Because 5-year survival in stage IV disease is 1% or less, discussions of outcomes in the literature often describe median survival, which can be measured in months. For a physician who is speaking with an individual patient, however, it is more meaningful to discuss the probability of living 1 or 2 years. A statement based on median survival in a large population of patients, such as "You have 6 months to live," does not help that patient understand the range of survival that occurs even in stage IV NSCLC. It gives the patient a better idea of the probabilities if the physician instead specifies the percentage of patients with advanced NSCLC who are alive at 1 year after diagnosis. According to a National Cancer Center database, untreated patients with stage IV disease had a 1-year survival of 9% to 11%, whereas patients receiving chemotherapy had a 1-year survival of 20% to 25%.49 These data are robust, because the population includes more than 700,000 patients with lung cancer, diagnosed between 1985 and 1995, and includes all stages and treatment categories for lung cancer. However, because this represents a database rather than a randomized comparison of groups, the survival data do not reflect the clinical factors that would guide the decision to forgo treatment in some patients. Such factors might include low performance status, comorbid disease, and advanced age, all of which may adversely affect survival.50
Which chemotherapeutic agents are best for stage IV disease? A meta-analysis of trials of supportive care alone versus supportive care with chemotherapy for advanced-stage disease demonstrated a 6% decrease in 1-year survival with the use of alkylating agents alone and a 4% improvement in 1-year survival with the use of vinca alkaloids or etoposide. Neither of these results reached statistical significance, however. By contrast, randomized trials of cisplatin-based combination chemotherapy versus supportive care showed an absolute increase of 10% in 1-year survival (P < 0.0001).28 These studies generally restricted eligibility to patients with higher performance status and enrolled a disproportionate number of younger patients with less comorbid disease than commonly seen in the community. Nevertheless, it is of interest that the magnitude of benefit in this trial is similar to that documented in the National Cancer Center database.
On the basis of this meta-analysis and additional data, an expert panel for the American Society of Clinical Oncology has concluded that in stage IV disease, platinum-based combination chemotherapy prolongs survival and is most appropriate for patients with good performance status, including an ECOG score of 0 or 1 or, possibly, 2.46 Although randomized trials of platinum-based chemotherapy have almost all involved cis-platin, carboplatin has a more favorable safety profile and lower toxicity. Furthermore, randomized trials comparing cisplatin and carboplatin with etoposide in NSCLC have shown comparable efficacy.51 Although a European study has suggested a small survival benefit for cisplatin therapy, compared with car-boplatin-based therapy,52 in the United States carboplatin continues to be the most widely used agent in the palliative management of patients with advanced lung cancer.
The 1990s brought the advent of newer agents in the treatment of NSCLC, including vinorelbine, paclitaxel, and gem-citabine, all three of which have received Food and Drug Administration approval for use (in combination with cisplatin) in the treatment of advanced disease. The approval of these newer agents was based on the results of randomized clinical trials that compared them, in combination with cisplatin, with either cisplatin alone or cisplatin in combination with older agents. In these trials, vinorelbine and cisplatin were associated with 1-year survival of 40%53 and 36%.54 These differences were statistically superior to vinorelbine alone, cisplatin and vindesine, or cisplatin alone; 1-year survival with cisplatin as a single agent was only 20%. Similarly, the combination of cisplatin and gem-citabine provided a 1-year survival of 39%, compared with 26% for cisplatin alone.55 In the study that led to the approval of pa-clitaxel for NSCLC, 1-year survival was 32% in the control groups that received cisplatin and etoposide, compared with 37% to 40% for patients receiving paclitaxel. Survival with pa-clitaxel depended on the dose used; patients receiving a higher dose required supportive therapy with granulocyte colony-stimulating factor.56 A randomized trial has demonstrated a better survival rate with cisplatin and docetaxel than with cis-platin and vinorelbine.57 In this study, survival for a carboplatin and docetaxel group was similar to that for a cisplatin and vi-norelbine group, but the former had a more favorable side-effect profile.
Overall, comparative trials have suggested that several different combination regimens may be equally effective in advanced disease, although toxicities vary.58-60 The optimal regimen should comprise two chemotherapy drugs, including a platinum agent and one of the newer chemotherapeutic agents. Of the newer agents, paclitaxel, vinorelbine, gemcitabine, or docetaxel would all be reasonable choices.
Palliative radiation therapy Radiation therapy is used for palliation of symptoms caused by metastatic NSCLC. These include obstructive symptoms, bone pain, and neurologic compromise from spinal cord compression or brain metastasis. In randomized trials, palliative radiation has been shown to produce some pain relief in 75% to 90% of patients and complete pain relief in at least 50%.61 Several fractionation schemes seem equally effective, but there is some evidence that prolonged treatment provides longer-lasting pain relief.
Superior vena cava syndrome responds to radiation treatment in approximately 50% of cases, but a substantial number of patients do not respond. Approximately 75% of patients have resolution of hemoptysis, and 50% will have cessation of cough after palliative radiation therapy. About 50% to 75% of patients with brain metastasis have a symptomatic response.
Spinal cord compression can also be treated with radiation therapy. Of patients who have only pain, 75% remain ambulatory, but only 30% to 35% of patients with muscle weakness improve.
Quality-of-life considerations Studies of chemotherapy for advanced NSCLC have largely focused on length of survival. In view of the modest benefits of treatment, however, the impact of chemotherapy on quality of life is a critical consideration. Although earlier trials often did not assess quality of life, virtually all of the current randomized clinical trials have qual-ity-of-life measures as a significant component. These studies suggest that combination chemotherapy often results in the improvement of symptoms such as cough, dyspnea, chest pain, and hemoptysis, often even when there is minimal evidence of tumor response.62 In general, these studies have shown that chemotherapy produces symptomatic improvement in more than 50% of patients. This figure is significantly higher than the objective response rate, which generally varies between 20% and 40% for combination regimens. The seeming discrepancy between these figures likely reflects the fact that even minor responses, or simply stabilizing the growth of the cancer, may bring at least short-term improvement in symptoms. With the improved toxicity profiles of the newer agents, improvements in quality-of-life differences may be easier to demonstrate.63
The elderly and other special populations Now that the benefits of chemotherapy have been established in younger patients with good performance status, researchers are evaluating its benefits in older patients and those with lower performance status. The first randomized trial of chemotherapy versus supportive care in the elderly (> 70 years), the Elderly Lung Cancer Vinorelbine Italian Study group (ELVIS) trial, assessed the effects of vinorelbine, as compared with supportive care alone, on both quality of life and survival of patients with advanced NSCLC.64 In this study, 1-year survival was 32% for the vinorel-bine-treated group, compared with 14% for best supportive care alone. On the basis of this trial, the FDA has approved vi-norelbine for single-agent use in NSCLC. Because of the superior outcome with the combination of cisplatin and vinorelbine, single-agent treatment has not been widely used in high-performance-status populations. However, the ELVIS trial suggests that single-agent chemotherapy is beneficial in elderly patients, many of whom may not be eligible for combination chemotherapy. However, for the fit elderly, a subset analysis of a randomized trial of carboplatin and paclitaxel compared with paclitaxel alone demonstrated a survival advantage for the combination regimen in patients older than 70 years; this result was comparable to that in younger patients.60
Relapse of disease For patients in whom initial chemotherapy has failed, two randomized trials have now demonstrated the benefit of salvage chemotherapy. A trial comparing docetaxel with either ifosfamide or vinorelbine showed a survival advantage for patients receiving docetaxel, particularly at the dose of 75 mg/m2.65 A trial comparing docetaxel with best supportive care showed that docetaxel produced improvement in both survival and quality of life.66 On the basis of these trials, the FDA has approved docetaxel for salvage chemotherapy in this setting.
Table 7 Therapy and Prognosis for Small Cell Lung Cancer
|
Stage |
Surgery |
Radiotherapy |
Chemotherapy |
Median Survival (months) |
Two-Year Disease-Free Survival (%) |
|
Limited I |
 |
 |
> 48 |
T1N0 > 60 |
|
 |
 |
> 14 |
T2N0 > 30 |
||
|
Limited II |
 |
Rarely used |
Several regimens available |
> 14 |
 |
|
Limited IIIA IIIB |
T3, N2, or both: research T4 or N3: not indicated |
Decreases local recurrences; used prophy-lactically for CNS |
Several regimens available; high-dose programs with autologous marrow rescue are being researched |
> 12 |
15-20 |
|
Complete response: 50%-80% |
|||||
|
Extensive IV |
Not indicated |
Useful for palliation of pain, atelectasis, SVC syndrome |
Several regimens available Complete response: 30%-50% |
> 8 |
s 5 |
SVC—superior vena cava
Gefinitib, a tyrosine kinase inhibitor, has received FDA approval as monotherapy for patients with locally advanced or metastatic NSCLC that has failed to respond to both platinum-based therapy and docetaxel. Approval was granted under accelerated-approval regulations; although no controlled trials have shown an improvement in symptoms or survival with the agent, gefinitib has produced objective responses and symptomatic improvement in this heavily pretreated population. Tox-icities include a skin rash, which is common, and interstitial lung disease, which is a rare but potentially fatal complication of gefinitib.67 There is reason to hope that this agent, which targets epidermal growth factor receptors, will usher in an era of more precise therapy for lung cancer.
Small cell carcinoma of the lung
Whereas the management of NSCLC changed substantially during the past decade, the management of SCLC has evolved more slowly. Important refinements in therapy have occurred, however. Currently, new directions in therapy are being developed, on the basis of better understanding of the biology of this disease.
As in NSCLC, the outcome in patients with SCLC is determined largely by stage at presentation [see Table 7]. The difference between the two diseases lies in the stage distribution. In NSCLC, 25% to 30% of patients present with stage I or II disease; in SCLC, that figure is well below 5%. Several factors may contribute to this difference, but a principal one appears to be that small cell carcinoma arises from neuroendocrine cells, which normally reside below the bronchial epithelium in the mucosa and submucosa. This region is much more heavily supplied with lymphatics, which would facilitate the earlier spread of these cancer cells to regional and distant lymph nodes, as well as hematogenous spread. In addition, in the 1960s it was learned that SCLC was more sensitive to chemotherapy than NSCLC. Because of the advanced stage at presentation and this chemosensitivity, SCLC fell largely into the domain of medical oncologists and radiation therapists rather than that of thoracic surgeons.
Early-Stage Disease
Early-stage solitary pulmonary nodules have been referred to as peripheral small cell carcinoma. In patients without medi-astinal nodal involvement, this disease is associated with a 5-year survival in the 30% to 60% range, depending on the series. Long-term follow-up of these patients suggests a high tendency for distant relapse, both systemically and in the CNS. Surgery should be the therapy of choice for patients with no evidence of distant disease or of mediastinal nodal involvement. Because the number of patients presenting at this stage is small, no definitive randomized trials of adjuvant therapy have been performed and there is no consensus on adjuvant therapy. In the United States, it has been standard to recommend adjuvant chemotherapy, either alone or in combination with prophylactic cranial radiation or chest radiation.
Surgery with mediastinal lymph node dissection, followed by four cycles of adjuvant platinum and etoposide, was the subject of a phase II trial in patients with stage I to IIIA SCLC conducted by the Japanese Collaborative Oncology Group.68 In patients with stage I disease, who constituted the bulk of the study population, 5-year survival was 69%. In the limited number of patients with stage II or IIIA disease, survival was 38% to 40%. No prophylactic cranial irradiation was done, and the rate of CNS relapse was 17%, even in the stage I population, which suggests that prophylactic cranial radiation should have a role in this population.
Limited-Stage or Extensive-Stage Disease
At least 95% of patients with SCLC present with either limited-stage or extensive-stage disease. Limited-stage SCLC corresponds to stage IIIA or IIIB NSCLC without malignant pleural effusion but, generally, with supraclavicular lymph node involvement. Simplistically, limited-stage SCLC patients include those in whom the primary tumor and involved nodes, including supraclavicular nodes, can be encompassed in a standard radiation port. By contrast, patients with extensive disease have a malignant pleural effusion, contralateral lung involvement, or metastases at more distant sites. This staging separation has been important for selecting treatment and determining prognosis. Patients with limited-stage disease have a 5-year survival of 15% to 20% when treated with the combination of chemotherapy and radiation. In contrast, patients with extensive-stage disease generally have a 5-year survival of 1% to 2% and have been treated primarily with chemotherapy, plus ad-junctive radiation as needed. The distinction between limited-and extensive-stage disease has largely been based on clinical examination, as well as on the standard imaging studies of chest CT through the adrenal glands, bone scan, and brain CT or MRI. Of patients with no evidence of distant metastases on these studies, an additional 5% will be found to have distant disease by bone marrow evaluation. Currently, approximately 25% to 30% of patients with SCLC present with limited-stage disease, and the remainder present with extensive-stage disease; however, with the development of more sophisticated imaging techniques and the increased use of PET imaging, some degree of stage migration is likely to occur.
Limited-stage disease Chemotherapy has been the mainstay of treatment for limited-stage disease for the past 30 years. For at least the past 10 years, the standard treatment has been based on etoposide, either as a single agent or combined with either cisplatin or carboplatin, generally for four to six cycles. Most of the advances in the understanding of treatment of limited-stage SCLC have come from trials evaluating the role of radiation therapy.69 These trials have shown that cure in limited-stage SCLC is associated with the use of full-dose combination chemotherapy, along with thoracic and cranial radiation therapy. Thoracic radiation appears to be more beneficial if it is done concurrently with chemotherapy, rather than sequentially. Secondly, starting radiation therapy during the first few cycles of chemotherapy appears to provide a survival advantage over beginning radiation therapy at the end of chemotherapy.70 The explanation for this is that drug resistance may evolve rapidly in small cell carcinoma, so in some patients, delaying thoracic radiation even for a few months may result in escape of drug-resistant cells from the lung and mediastinum to distant sites. Early intervention with radiation may result in the destruction of these cells and an increase in the rate of cure. However, as with NSCLC, the benefit of concurrent chemotherapy and radiation must be balanced against its higher tox-icity (e.g., esophagitis).
Although clinical practice varies, most oncologists initiate radiation therapy with chemotherapy cycle one or with cycles two or three. The choice of timing depends on the clinical situation. The advantage of delaying radiation for one or two cycles is that during this time the tumor mass may be significantly reduced, therefore easing the patient’s symptoms and often improving performance status, so that the patient is better able to cope with combination treatment. Most initial trials of concurrent therapy have used pretreatment tumor volumes to determine radiation fields. However, newer strategies of radiation therapy will take advantage of chemotherapy-induced tumor reduction and use higher doses of radiation delivered to a smaller tumor volume. In one major trial, twice-daily radiation, compared with once-daily treatment, has been associated with improved surgical outcome,71 but twice-daily treatment is also associated with significantly higher toxicity rates. The use of twice-daily radiation can be considered in some patients with high performance status. The standard dose of daily radiation therapy that is given concurrently with chemotherapy has increased from 45 Gy to 60 Gy.
Another significant advance in the understanding of SCLC involves the role of prophylactic cranial irradiation. Early trials that used prophylactic cranial irradiation along with chemotherapy reported a high incidence of both immediate and delayed neurologic toxicity. Studies have used prophylactic cranial irradiation at lower dose fractions and have delayed its initiation until chemotherapy has been completed. These adjustments seem to be associated with substantially diminished immediate and delayed neurotoxicity. Moreover, a meta-analysis of the use of prophylactic cranial irradiation from randomized clinical trials (largely composed of patients with limited-stage disease) has demonstrated not only a reduction in CNS relapse but also an increase in long-term survival by approximately 5%.72
Several issues involving prophylactic cranial irradiation remain to be settled. One is its role in elderly patients, who appear to be at higher risk for neurotoxicity. Also, although trials of SCLC have often included patients in complete remission from either limited or extensive disease, most of the patients have had limited disease. Therefore, the strength of the recommendation for prophylactic cranial irradiation in patients in complete remission from extensive-stage SCLC is less certain.
A number of trials have evaluated maintenance chemotherapy or the addition of other treatments after chemoradiothera-py for SCLC. Phase II studies of high-dose chemotherapy for patients in remission have suggested an encouraging long-term survival rate,73 and a potential long-term benefit has also been reported in a small number of patients who received a vaccine against a tumor ganglioside (BEC-2)74; the latter approach is currently the subject of a randomized phase III trial. No benefit for these approaches has yet been proved, however, so maintenance therapy after completion of standard therapy cannot be recommended for SCLC patients, except in the context of a clinical trial.
Extensive-stage disease Progress in the treatment of extensive-stage SCLC over the past 10 years has been slower than that of limited-stage SCLC.69 As with limited-stage disease, treatment centers on chemotherapy. Although several different chemotherapy combinations have been demonstrated to have equivalent survival outcomes, etoposide-based regimens have been shown to be superior to older regimens such as cy-clophosphamide, doxorubicin, and vincristine. Secondly, most clinical trials in SCLC have used cisplatin, but as noted above, carboplatin has a better therapeutic ratio, is equally effective, and causes significantly less nausea, vomiting, and neurotoxic-ity. 75,76 Currently, however, cisplatin and carboplatin are used with roughly equal frequency for SCLC in community practice in the United States. Both agents are generally administered every 3 to 4 weeks, depending on hematologic recovery. Although the combination of platinum and etoposide remains the most commonly used regimen in the United States, cis-platin and irinotecan has been approved for treatment of SCLC in Japan, and randomized trials of this combination are ongoing in the United States.
Special populations Adaptation of treatment to special populations is another very important principle in the management of SCLC. Standard combination chemotherapy regimens, which were developed for younger, healthier patients, pose a greater risk of severe myelosuppression and life-threatening complications in elderly patients with significant comorbid disease and in patients with poor performance status. Modified regimens should be considered in such patients. If combination therapy is employed, the use of carboplatin (dosed according to renal function) rather than cisplatin may produce less toxici-ty in this population. Although monotherapy with low-dose etoposide is certainly a consideration, most elderly patients should be considered for combination carboplatin and etopo-side therapy if permitted by their level of functional illness and degree of comorbid illness. Whether these patients should receive significantly reduced doses or near-standard doses plus hematopoietic growth factor has not been well studied.
Relapse of disease Although SCLC typically responds to therapy, the majority of patients with limited-stage disease and nearly all patients with extensive-stage disease will experience relapses. For limited-stage patients, the duration of remission may range from months to years. For most patients with extensive-stage disease, remission generally lasts only weeks to months after completion of chemotherapy. In both groups of patients, a major issue is the emergence of drug resistance, which has severely hampered the benefit of second-line or so-called salvage chemotherapy.
In this setting, a critical factor in choosing therapy is the time to relapse. For patients who have had a prolonged remission (longer than 1year), retreatment with the same regimen will often produce a response. On the other hand, in patients who fail to respond to initial therapy (primary drug resistance), no therapy has been convincingly shown to be beneficial, so these patients are best enrolled in experimental trials.
For patients who have had a remission of at least 2 months after initial therapy, topotecan has been shown to produce a significant improvement in disease-related symptoms—partic-ularly dyspnea, but also anorexia, hoarseness, and fatigue— and has been approved for second-line therapy for patients with SCLC.77 The other commonly used approach in relapses of SCLC has been prolonged courses of oral etoposide for 10 to 21 days per cycle. Other agents that have shown promise for salvage therapy include paclitaxel, irinotecan, vinorelbine, and gemcitabine.69,78 These agents, as well as more novel approaches, warrant further investigation.79
Supportive care
Supportive care is important across all stages of both NSCLC and SCLC. In patients with advanced or refractory disease, it often becomes the principal form of therapy. As the ability of oncologists to deliver both chemotherapy and radiation safely has improved, it is clear that one of the best supportive care approaches is to palliate disease-related symptoms with chemotherapy, radiation, or both.
Over the past 10 years, management of pain, depression, nausea, weight loss, and constipation in cancer patients has been improved by the development of pharmacologic agents with better efficacy and less toxicity [see CE:X Symptom Management in Palliative Medicine]. In patients who are receiving myelosuppressive chemotherapy and so are at high risk for neutropenic complications, the use of colony-stimulating factors has been demonstrated to substantially reduce the risk of fever, neutropenia, and infection.80 How to identify the populations who are at highest risk and are thus candidates for primary prophylaxis with colony-stimulating factors is an area that needs further investigation.
Anemia and fatigue are now appreciated as especially important concerns in patients with lung cancer.81 Analysis of the 1,748 lung cancer patients enrolled in the prospective trials of recombinant human erythropoietin for cancer patients with anemia has shown that amelioration of anemia led to significant improvements in quality of life; these improvements applied to patients with SCLC or NSCLC receiving a variety of chemotherapy regimens.82 Although all anemic cancer patients may benefit from higher hemoglobin levels, those with lung cancer may reap particular benefits because of dyspnea and fatigue related to the cancer itself, as well as comorbid diseases and the treatments used. In addition to substantially enhancing quality of life, normalization of hemoglobin levels may also improve therapeutic outcome, perhaps by improving tolerance to treatment or by reducing tumor hypoxia. Studies evaluating the role of hemoglobin maintenance in reducing cognitive dysfunction during treatment are ongoing.
Treatment updates
The National Cancer Center Network (NCCN) provides evidence-based, expert-panel guidelines for the management of cancer, as well as supportive care guidelines. Clinicians are encouraged to review the NCCN Web site (http://www.nccn.org) for annual updates of lung cancer management guidelines.
Prognosis
At present, the 5-year survival rate for lung cancer is approximately 14%. The chance of survival is heavily dependent on stage at presentation, however. For example, 5-year survival in patients with stage I NSCLC is approximately 63%. Unfortunately, more than 40% of all patients with NSCLC have stage IV disease at diagnosis, and in this population the survival at 5 years is 1% or less. In patients with SCLC, 5-year survival ranges from 30% to 60% for peripheral disease (stage I/II) to 1% to 3% for patients with extensive disease (stage IIIB/IV). Unfortunately, approximately 65% to 70% of patients with small cell lung cancer present with extensive disease.
