Clinically Meaningful Improvement in Symptoms and Quality of Life for Patients With Non-Small-Cell Lung Cancer Receiving Gefitinib in a Rand
http://www.100md.com
《临床肿瘤学》
the Evanston Northwestern Healthcare and Northwestern University Feinberg School of Medicine, Evanston, IL
University of Texas M.D. Anderson Cancer Center, Houston, TX
Massachusetts General Hospital Cancer Center, Boston, MA
University of California, Los Angeles Medical Center, Los Angeles
Cedars-Sinai Comprehensive Cancer Center, Beverly Hills, CA
University of Pittsburgh Cancer Institute, Pittsburgh, PA
University of Wisconsin Hospital, Madison, WI
AstraZeneca, Wilmington, DE
Memorial Sloan-Kettering Cancer Center, New York, NY
AstraZeneca, Macclesfield, United Kingdom
ABSTRACT
PURPOSE: Evaluation of disease-related symptom improvement rate by the Lung Cancer Subscale (LCS) of the Functional Assessment of Cancer Therapy-Lung (FACT-L) questionnaire was a coprimary end point of the pivotal phase II trial of gefitinib (Iressa; AstraZeneca, Wilmington, DE) conducted in the United States. This report includes the results of analyses exploring the relationship between weekly LCS scores and radiographic response and survival, as well as detailed protocol-specified analysis of symptom and quality-of-life data.
PATIENTS AND METHODS: In this trial, 216 symptomatic patients with advanced non-small-cell lung cancer (NSCLC) who had at least two prior chemotherapy regimens received gefitinib 250 or 500 mg/d. Disease-related symptoms were assessed weekly and quality of life was assessed monthly by LCS and FACT-L, respectively.
RESULTS: Symptom improvement was rapid and correlated with tumor response and survival. At the recommended gefitinib dose of 250 mg/d, median overall survival times were 13.6 and 4.6 months for patients with and without symptom improvement, respectively, and 9.7 months for patients with symptom improvement without tumor response. Among patients with stable disease or disease progression, those with symptom improvement had significantly better overall survival than those without improvement. At 250 mg/d, 30% of patients showed a quality-of-life improvement that was correlated with tumor response.
CONCLUSION: This triadic analysis of response, survival, and symptom data supports the hypothesis that tumor response and symptom response are related and that each predicts survival. Among these NSCLC patients treated with gefitinib, symptom improvement was complementary to and, for most patients, preceded evidence of radiographic regression.
INTRODUCTION
Improvements in quality of life and disease-related symptoms are key goals in the treatment of advanced non-small-cell lung cancer (NSCLC). With the introduction of agents, such as gefitinib (Iressa; AstraZeneca, Wilmington, DE), that have a more favorable tolerability profile than traditional chemotherapy agents, improvements in quality of life and symptoms are likely to be significant factors affecting treatment choice.
Gefitinib is an epidermal growth factor receptor tyrosine kinase inhibitor that blocks signal transduction pathways implicated in the proliferation and survival of cancer cells. Reported results of two pivotal phase II trials, one conducted internationally and the other in the United States (referred to in some publications as the Iressa Dose Evaluation in Advanced Lung Cancer 1 and 2 trials, respectively) showed that, in addition to radiographic regressions, gefitinib provides relief from the symptoms of advanced NSCLC.1,2 Both studies used the Functional Assessment of Cancer Therapy-Lung (FACT-L) questionnaire (version 4) and its Lung Cancer Subscale (LCS) to evaluate changes in quality of life and lung cancer-related symptoms, respectively.3,4 The FACT-L questionnaire was designed to be relevant to patient experience, to be sensitive to change, and to measure social and emotional well-being, as well as physical and functional well-being.4
The FACT-L has the following five components: physical well-being, functional well-being, social well-being, emotional well-being, and the LCS. The Trial Outcome Index of FACT-L, which is the sum of the physical well-being, functional well-being, and LCS scores, is a measure of physical aspects of quality of life. A 5- to 7-point change in Trial Outcome Index score has been suggested as the minimally important difference among patients with NSCLC.5 The LCS of FACT-L is an independently validated tool that measures the disease-related symptoms of lung cancer on a scale of 0 (most symptomatic) to 28 (asymptomatic). The Eastern Cooperative Oncology Group has demonstrated that a 2- to 3-point change in LCS score reflected a minimally important difference associated with performance status, weight loss, objective tumor response, and time to progression.5 On average, a 2- to 3-point difference in LCS at baseline was found to be comparable to the difference between a patient entering the study with performance status of 0 versus a performance status of 1 or with 5% weight loss versus less than 5% weight loss. The LCS was shown to have an equal or higher prognostic value than the established prognostic factors of weight loss and performance status.5
Symptom improvement data may be a valuable measure of clinical benefit in patients with advanced NSCLC treated with gefitinib. The results of phase I studies of gefitinib in patients with solid tumors showed that changes in LCS score seemed sensitive to clinical change in patients with NSCLC.6 Evaluation of the disease-related symptom improvement rate was a coprimary end point of the phase II study of gefitinib in the United States, in which patients had to have at least one severe or between two and four mild disease-related symptoms at entry (LCS score 24 at baseline). These results have been previously reported.2 In this article, we present the results of analyses to fully explore the relationship between symptom improvement, radiographic response, and survival. We also include detailed protocol-specified analysis of symptom and quality-of-life data from the trial. The previously reported results of both the international and US phase II trials1,2 established the recommended dose of gefitinib as 250 mg/d. In this article, we include comparative data for two dose groups (250 and 500 mg/d) but focus on data at the recommended dose.
PATIENTS AND METHODS
Trial Design
The US phase II trial of gefitinib was a randomized, double-blind, parallel-group, multicenter trial that evaluated two doses of gefitinib (250 and 500 mg/d) administered orally and once daily to patients with locally advanced or metastatic NSCLC. Eligible patients had received at least two prior chemotherapy regimens including platinum and docetaxel, which were administered either concurrently or as separate treatments, and had a baseline LCS score of 24. Patients were excluded if they had received chemotherapy or irradiation within 14 days. All patients gave informed consent, and approval was obtained from the investigational review boards at each trial center.
The primary end points were rates of LCS-assessed symptom improvement and objective tumor response. Secondary end points included the disease control rate (complete response plus partial response plus stable disease), overall survival, frequency and severity of adverse events, and overall quality of life using the FACT-L instrument. Full details of the trial design and statistical methods used in this study have been described by Kris et al.2 The intent-to-treat population who received any gefitinib is analyzed here.
Symptom Improvement
Disease-related symptoms were measured weekly using the LCS of the validated quality-of-life instrument, the FACT-L questionnaire4 (Table 1). Patients completed a diary card recording the severity of symptoms on days 8, 15, and 22 of each 28-day cycle, and the LCS was evaluated at baseline and every 28 days as part of the full FACT-L questionnaire. Missing data points were counted as no change in symptoms. The maximum attainable score on the LCS is 28, where the patient is considered to be asymptomatic. Patients were eligible for the trial if they had a baseline LCS score of 24. Patients were rated as improved, no change, or worsened according to their scores compared with baseline, and their best overall response was determined. A best overall LCS response of improved was defined as an increase in LCS score of 2 points sustained for 28 days with no worsening at any interim weekly assessment.5 A best overall response of other was allocated if the data did not meet the criteria for improved, no change, or worsened or if there were missing or incomplete data. Time to symptom improvement was measured from randomization to the first visit response of improved. Patients were withdrawn from the study because of disease progression, which was documented by radiographic studies, or toxicity. The first postbaseline radiographic assessment took place after 28 days; thus, up to four LCS evaluations were obtained before patients were aware of their postbaseline disease status. All patients were evaluated during the trial for the use of the following concomitant medications that might improve pulmonary symptoms: narcotics, bronchodilators, antidepressants or anxiolytics, oxygen, systemic corticosteroids, transfusions or erythropoietin, systemic antibiotics, and cough syrups.
Quality of Life
The FACT-L questionnaire was completed at baseline, on day 28 of each treatment period, and at withdrawal; at each of these time points, the questionnaires were completed before clinical assessment and before patients were informed of their disease status. Patients were given as much time as necessary to complete the forms and did so without help from relatives. Trial Outcome Index and overall FACT-L scores were used to assess quality of life. The highest scores attainable for Trial Outcome Index and FACT-L were 84 and 136, respectively. A best overall response of improved was defined as an increase in Trial Outcome Index or FACT-L score of 6 points sustained for 4 weeks. This degree of change has been shown to be clinically meaningful.5 A best overall response of other was allocated if the data did not meet the criteria for improved, no change, or worsened or if there were missing or incomplete data.
Radiographic Response and Survival
Radiographic assessments were performed 14 days before randomization, after approximately 28 and 56 days, and every 8 weeks thereafter, in accordance with the Southwest Oncology Group-modified Union Internationale Contre le Cancer/WHO criteria.7 Overall survival was measured from the randomization date to the date of death. Patients who were alive at data cutoff were censored at the last date the patient was known to be alive.
RESULTS
Patients
A total of 216 patients were randomly assigned to either gefitinib 250 mg/d (102 patients) or gefitinib 500 mg/d (114 patients). The patient characteristics were well balanced between the two dose groups (Table 2). 2
Disease-Related Symptoms and Quality of Life at Baseline
The median and range of baseline LCS, Trial Outcome Index, and FACT-L scores for this sample and for a normative reference sample are listed in Table 3. 8 Normative scores provide a context for how this sample was characterized relative to a normative reference population of patients with lung cancer. Patients from this sample had lower LCS, FACT-L, and Trial Outcome Index baseline scores compared with normative data. The eligibility criteria for participation in the trial (LCS score 24) are consistent with the observation that the study sample had lower quality-of-life scores on these three indices compared with a normative sample. Symptoms that were most frequently severe at baseline were cough and shortness of breath (Fig 1). 2
Compliance
The compliance rates (ratio of the number of assessable weekly forms to the number of forms expected) for the LCS were 86% and 82% in patients receiving 250 and 500 mg/d, respectively. The overall compliance rates for FACT-L were 88% and 84% in patients receiving 250 and 500 mg/d, respectively.
Symptom Improvement
The symptom improvement rates were 43.1% (95% CI, 33.4% to 53.3%) for patients receiving 250 mg/d and 35.1% (95% CI, 26.4% to 44.6%) for patients receiving 500 mg/d.2 Symptom improvement rates were similar when the criterion of a 2-point change in LCS score was increased to a 3-point change (39.2% at 250 mg/d and 31.6% at 500 mg/d; Fig 2). 8 Increasing the criterion to a 7-point change (more than double the established minimally important difference) resulted in symptom improvement rates of 16.7% and 11.4% for patients receiving 250 and 500 mg/d, respectively. Similar rates of symptom improvement were observed in patients receiving gefitinib as third-, fourth-, or fifth-line and above therapy; at 250 mg/d, symptom improvement rates were 41.5%, 54.8%, and 35.7%, respectively; at 500 mg/d, these rates were 37.5%, 36.6%, and 28.0%, respectively.
Those patients who had symptom improvement were off treatment for a median of 53.5 days between stopping their last chemotherapy until the first dose of gefitinib (250 mg/d: median, 56.0 days; range, 15 to 410 days; and 500 mg/d: median, 52.0 days; range, 15 to 338 days). Similarly, those patients without symptom improvement were also off treatment for a median of 53.0 days between stopping their last chemotherapy until the first dose of gefitinib (250 mg/d: median, 50.0 days; range, 13 to 374 days; and 500 mg/d: median, 53.0 days; range, 15 to 434 days).
Symptom improvement was rapid; the median time to onset was less than 2 weeks at both doses (250 mg/d: median, 10 days; 95% CI, 8 to 22 days; and 500 mg/d: median, 9 days; 95% CI, 9 to 16 days; the times of the first postbaseline assessments). The median duration of symptom improvement for the 250 mg/d group had not been reached at 7 months, but the median duration was 5.4 months for the 500 mg/d group.
Improvement from baseline in LCS score was observed for each individual LCS item (Fig 1). Similar results were obtained for the 250 and 500 mg/d groups. Generally, more patients showed an improvement in the pulmonary items of the LCS than the nonpulmonary items. Of the 44 patients who were assessable for improvement (defined as an increase in symptom score of 1 point sustained for 28 days) in each of the LCS items at 250 mg/d, 91% had improvement in shortness of breath, 94% had improved cough, 89% had improvement in tightness in chest, and 95% had improvement in ease of breathing. Appetite, weight loss, and clear thinking improved in 84%, 88%, and 86% of patients, respectively. The mean changes in the LCS items in assessable patients with LCS improvement are listed in Table 4. Pulmonary symptoms had the greatest magnitude of change, with shortness of breath and cough being the most improved. Although pulmonary symptoms tend to co-occur in patients, occurrence of any one of pulmonary symptoms did not predict the occurrence of any other with sufficient accuracy to consider any two pulmonary symptoms as providing redundant information.
Only nine of the 44 patients with symptom improvement at 250 mg/d had received concomitant medications that might improve pulmonary symptoms after starting treatment with gefitinib but before symptom improvement occurred. An additional 32 of these 44 patients had medications added after the improvement of symptoms occurred. In comparison, 48 of the 58 patients who did not have self-reported symptom improvement had additional concomitant medications added while they were on study.
Symptom Improvement by Objective Tumor Response
The objective response rates were 11.8% and 8.8% at 250 and 500 mg/d, respectively, as previously reported.2 In addition, a further 30.4% and 27.2% of patients achieved stable disease at 250 and 500 mg/d, respectively. A correlation was observed between symptom improvement and objective tumor response. At 250 mg/d, all patients who had a partial response also had symptom improvement, as did 81% of patients who had stable disease; at 500 mg/d, symptom improvement was experienced by 90% of patients with a partial response and 61% of patients with stable disease. Of those patients with progressive disease, only 12% and 20% of patients at the 250 and 500 mg/d doses, respectively, had symptom improvement. Because symptom improvement occurred within 2 weeks in most cases (and, therefore, before first follow-up radiographic assessment), these results suggest that most of the observed symptom improvement preceded detection of objective tumor response and could not have been influenced by the patient's knowledge of clinical changes in his or her disease. The magnitude of improvement from baseline in mean on-study LCS score also correlated with the magnitude of tumor response (Table 5).
Survival by Radiographic Response and Symptom Improvement
One-year survival rates were 29% and 24% at 250 and 500 mg/d, respectively.9 Median progression-free survival time was 1.9 months (95% CI, 1.8 to 2.8 months) and 2.0 months (95% CI, 1.6 to 2.6 months) at 250 and 500 mg/d, respectively.
Longer survival was observed for patients with an objective response compared with patients without an objective response at both dose levels. At 250 mg/d, in patients who survived for 2 months, median survival time for patients with partial response was 17.9 months compared with 9.4 months for patients with stable disease and 5.2 months for patients with disease progression (Fig 3A); at 500 mg/d, the comparable median survival times were 13.6, 10.2, and 6.0 months, respectively. Patients with symptom improvement had a longer overall survival than patients without symptom improvement (13.6 v 4.6 months, respectively, at 250 mg/d, Fig 3B; and 10.5 v 5.2 months, respectively, at 500 mg/d). At 250 mg/d, those patients with symptom improvement but no partial response had a median overall survival time of 9.7 months, whereas patients with no partial response and no symptom improvement had a median overall survival time of only 4.6 months (Fig 3C). Comparable results were found at 500 mg/d. Similarly, patients with stable disease or progressive disease also had a longer overall survival if they had symptom improvement than if they did not (Fig 4A and 4B).
Physical Dimension of Quality of Life
Quality-of-life improvement rates by Trial Outcome Index (a summary score of physical and functional domains of FACT-L and LCS) were 33.3% (95% CI, 24.3% to 43.4%) for 250 mg/d patients and 20.2% (95% CI, 13.2% to 28.7%) for 500 mg/d patients. As measured by FACT-L total score (ie, overall multidimensional quality of life), improvement rates were 34.3% (95% CI, 25.2% to 44.4%) and 22.8% (95% CI, 15.5% to 31.6%) in patients treated at 250 and 500 mg/d, respectively. Of the patients who demonstrated improvement in Trial Outcome Index, 94.1% and 95.6% receiving 250 and 500 mg/d, respectively, had improved by the time of the first assessment (at 4 weeks). The FACT-L results were similar; 94.3% and 73.1% of patients on 250 and 500 mg/d, respectively, had improved by the first assessment. Improvements were durable; of those patients with Trial Outcome Index improvement, 47.1% of the 34 patients at 250 mg/d and 52.2% of the 23 patients at 500 mg/d remained improved at 3 months. This was similar for FACT-L; 45.7% of the 35 improved patients and 46.2% of the 26 improved patients at 250 and 500 mg/d, respectively, still had an improvement in quality of life at 3 months. In general, there was a relationship between the FACT-L and Trial Outcome Index best responses and the LCS best response; less than 5% of the patients with an LCS best response of improved had either a FACT-L or Trial Outcome Index best response of worsened.
As with symptom improvement, there was a correlation between radiographic regression and physical quality of life. At 250 mg/d, 92% of patients who had a partial response and 58% of patients with stable disease had an improvement in Trial Outcome Index score; at 500 mg/d, 70% of patients with a tumor response had improvements in quality of life, as did 45% of patients with stable disease (Fig 5). Similar results were seen for overall quality of life (FACT-L).
The FACT-L asked patients whether they had ever smoked, enabling a post hoc exploratory analysis of a possible relationship between smoking history and response to gefitinib. Indeed, those patients who reported they never smoked (n = 48) had a higher symptom improvement rate than former or current smokers (n = 168; 52.1% v 35.1%, respectively; P = .035).
DISCUSSION
This study demonstrates that symptomatic patients with lung cancer can reliably report changes in their symptoms and quality of life and that those changes correlate well with tumor response and survival. Symptom improvement measured using the FACT-L LCS is a robust assessment and may be a useful measure of efficacy in patients with symptomatic, advanced NSCLC. Compliance rates for LCS and FACT-L exceeded current standards in multicenter studies and allow reasonably robust conclusions to be drawn regarding LCS end points and the other end points derived from FACT-L.
The triadic analysis of response, survival, and symptom data clearly shows that those patients with symptom improvement or radiographic regression have longer survival than those patients without symptom improvement or radiographic response. At gefitinib 250 mg/d (the recommended dose), all patients with partial response also experienced symptom improvement. Of patients with stable or progressive disease, those who also experienced symptom improvement had a longer median survival time compared with those in the same tumor response category without symptom improvement. The early onset of symptom improvement seen with gefitinib, its association with partial response, progression-free survival, and overall survival, and its magnitude and duration suggest that this measured symptom improvement is an accurate rendering of the patient experience with disease symptoms in patients who are not yet influenced by any knowledge of clinical (radiographic) change. Similar results were also obtained in the international phase II trial of gefitinib.10
Many of the patients in this trial had improvements in disease-related symptoms, including the most heavily pretreated patients. Because symptom improvement generally occurred within the first 2 weeks of treatment and the first tumor assessment was at 4 weeks, patient-reported symptom improvement data were blind to tumor response results in this period. Up to four LCS evaluations were obtained before patients were aware of their postbaseline disease status. Increasing the criterion for symptom improvement from a 2-point change in LCS score to a 3-point change had little effect on the calculated symptom improvement rate. Increasing to a 7-point change (more than twice the minimally important difference for this instrument) resulted in more than 10% of patients with symptom improvement. Although only a 2-point improvement in LCS score sustained for 4 weeks signified symptom improvement, the mean LCS score improvement over the time on study (at the recommended 250 mg/d dose) was 6.1 for patients with a partial response and 3.0 for patients with stable disease.
All patients in this study had prior chemotherapy; therefore, it is possible that the symptom improvement that is noted is, in part, a result of recovery from chemotherapy side effects. However, several factors mitigate this conclusion. First, the median time since last chemotherapy dose was almost 2 months (53 days), with a minimum time since last dose of 2 weeks. Second, the median time since completing chemotherapy was no different between symptom improvers (53.5 days) and nonimprovers (53 days). Third, the majority of patients who responded symptomatically did so rapidly (within 1 week), making it unlikely that it was a reflection of gradual recovery from chemotherapy in the majority of patients. Fourth, the observed symptom improvement rate correlated with subsequent response to gefitinib treatment. Finally, there are no published data suggesting that patients with stage IIIb/IV NSCLC feel better during the weeks to months after chemotherapy discontinuation for progressive disease. Nevertheless, because this study was not controlled with a no-treatment (best supportive care) condition, we cannot fully exclude the possibility that stopping chemotherapy contributed to some degree to the symptom improvement observed in this trial.
Because of its potential for contributing important clinical information, we conducted an exploratory analysis of the impact of smoking history on symptom improvement rate and found evidence of a relationship suggesting that a higher proportion of patients who never smoked experienced symptom improvement. We emphasize that this was an exploratory, hypothesis-generating analysis that warrants further study as to possible mechanisms of action and clinical implications.
Most patients with stable disease had improvements in disease-related symptoms and quality of life. This suggests that the achievement of durable stable disease, although challenging to label a result of therapy, can be a valuable patient-centered goal in advanced NSCLC if it is tied to symptom relief. We note that most (n = 32) of the 44 symptom-responding patients had supportive medication added after symptom response. We are unable to determine why the medication was added or if the medication could have contributed to lasting improvement.
In conclusion, treatment with gefitinib offers clinically significant benefits in symptom improvement and quality of life for many heavily pretreated patients with advanced NSCLC. The results of this study indicate that symptom improvement alone is an appropriate end point for future randomized trials. Quality of life is one of the secondary end points in an ongoing international (not including the United States) phase III NSCLC study of gefitinib versus best supportive care after failure of platinum- and taxane-based therapy.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Anne Heyes, AstraZeneca; Judith S. Ochs, AstraZeneca; Michael K. Wolf, AstraZeneca; Andrea C. Kay, AstraZeneca, Novartis. Leadership Position: Roy S. Herbst, AstraZeneca. Consultant/Advisory Role: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Chandra P. Belani, AstraZeneca; Joan H. Schiller, AstraZeneca, Genentech; Mark G. Kris, AstraZeneca. Stock Ownership: Anne Heyes, AstraZeneca; Michael K. Wolf, AstraZeneca; Andrea C. Kay, AstraZeneca. Honoraria: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Diane Prager, AstraZeneca; Chandra P. Belani, AstraZeneca; Joan H. Schiller, AstraZeneca. Research Funding: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Joan H. Schiller, AstraZeneca. Expert Testimony: Thomas J. Lynch, AstraZeneca; Mark G. Kris, AstraZeneca. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
NOTES
Presented in part at the 152nd Annual Meeting of the American Medical Association, Chicago, IL, June 12-14, 2003; and at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Fukuoka M, Yano S, Giaccone G, et al: Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol 21:2237-2246, 2003
Kris MG, Natale RB, Herbst RS, et al: Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: A randomized trial. JAMA 290:2149-2158, 2003
Cella DF, Tulsky DS, Gray G, et al: The Functional Assessment of Cancer Therapy Scale: Development and validation of the general measure. J Clin Oncol 11:570-579, 1993
Cella DF, Bonomi AE, Lloyd SR, et al: Reliability and validity of the Functional Assessment of Cancer Therapy-Lung (FACT-L) quality of life instrument. Lung Cancer 12:199-220, 1995
Cella D, Eton DT, Fairclough DL, et al: What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire Results from Eastern Cooperative Oncology Group (ECOG) Study 5592. J Clin Epidemiol 55:285-295, 2002
LoRusso PM, Herbst RS, Rischin D, et al: Improvements in quality of life and disease-related symptoms in phase I trials of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 in non-small cell lung cancer and other solid tumors. Clin Cancer Res 9:2040-2048, 2003
Green S, Weiss GR: Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Invest New Drugs 10:239-253, 1992
Brucker PS, Yost K, Cashy J, Cella D, et al: Normative reference values for the FACT-G, in Cella D (ed): Functional Assessment of Chronic Illness Therapy (version 4). Evanston, IL, Evanston Northwestern Healthcare, 2004
Cella D, Natale RB, Lynch TJ, et al: Disease-related symptoms in advanced non-small-cell lung cancer as measured by the Lung Cancer Subscale of the FACT-L questionnaire: Clinically meaningful improvement with gefitinib (Iressa, ZD1839). 39th Annual Meeting of the Am Soc Clin Oncol, Chicago, IL, May 31-June 3, 2003 (poster 2531)
Gatzemeier U, Douillard JY, Kris M, et al: Rapid and durable objective responses in patients with advanced non-small-cell lung cancer in phase II trials (IDEAL 1 and IDEAL 2) treated with gefitinib. Eur J Cancer 1:S20, 2003 (suppl 1, abstr 52)
Douillard J-Y, Giaccone G, Horai T, et al: Improvement in disease-related symptoms and quality of life in patients with advanced non-small-cell lung cancer (NSCLC) treated with ZD1839 (Iressa) (IDEAL 1). Proc Am Soc Clin Oncol 21:299a, 2002 (abstr 1195)(David Cella, Roy S. Herbs)
University of Texas M.D. Anderson Cancer Center, Houston, TX
Massachusetts General Hospital Cancer Center, Boston, MA
University of California, Los Angeles Medical Center, Los Angeles
Cedars-Sinai Comprehensive Cancer Center, Beverly Hills, CA
University of Pittsburgh Cancer Institute, Pittsburgh, PA
University of Wisconsin Hospital, Madison, WI
AstraZeneca, Wilmington, DE
Memorial Sloan-Kettering Cancer Center, New York, NY
AstraZeneca, Macclesfield, United Kingdom
ABSTRACT
PURPOSE: Evaluation of disease-related symptom improvement rate by the Lung Cancer Subscale (LCS) of the Functional Assessment of Cancer Therapy-Lung (FACT-L) questionnaire was a coprimary end point of the pivotal phase II trial of gefitinib (Iressa; AstraZeneca, Wilmington, DE) conducted in the United States. This report includes the results of analyses exploring the relationship between weekly LCS scores and radiographic response and survival, as well as detailed protocol-specified analysis of symptom and quality-of-life data.
PATIENTS AND METHODS: In this trial, 216 symptomatic patients with advanced non-small-cell lung cancer (NSCLC) who had at least two prior chemotherapy regimens received gefitinib 250 or 500 mg/d. Disease-related symptoms were assessed weekly and quality of life was assessed monthly by LCS and FACT-L, respectively.
RESULTS: Symptom improvement was rapid and correlated with tumor response and survival. At the recommended gefitinib dose of 250 mg/d, median overall survival times were 13.6 and 4.6 months for patients with and without symptom improvement, respectively, and 9.7 months for patients with symptom improvement without tumor response. Among patients with stable disease or disease progression, those with symptom improvement had significantly better overall survival than those without improvement. At 250 mg/d, 30% of patients showed a quality-of-life improvement that was correlated with tumor response.
CONCLUSION: This triadic analysis of response, survival, and symptom data supports the hypothesis that tumor response and symptom response are related and that each predicts survival. Among these NSCLC patients treated with gefitinib, symptom improvement was complementary to and, for most patients, preceded evidence of radiographic regression.
INTRODUCTION
Improvements in quality of life and disease-related symptoms are key goals in the treatment of advanced non-small-cell lung cancer (NSCLC). With the introduction of agents, such as gefitinib (Iressa; AstraZeneca, Wilmington, DE), that have a more favorable tolerability profile than traditional chemotherapy agents, improvements in quality of life and symptoms are likely to be significant factors affecting treatment choice.
Gefitinib is an epidermal growth factor receptor tyrosine kinase inhibitor that blocks signal transduction pathways implicated in the proliferation and survival of cancer cells. Reported results of two pivotal phase II trials, one conducted internationally and the other in the United States (referred to in some publications as the Iressa Dose Evaluation in Advanced Lung Cancer 1 and 2 trials, respectively) showed that, in addition to radiographic regressions, gefitinib provides relief from the symptoms of advanced NSCLC.1,2 Both studies used the Functional Assessment of Cancer Therapy-Lung (FACT-L) questionnaire (version 4) and its Lung Cancer Subscale (LCS) to evaluate changes in quality of life and lung cancer-related symptoms, respectively.3,4 The FACT-L questionnaire was designed to be relevant to patient experience, to be sensitive to change, and to measure social and emotional well-being, as well as physical and functional well-being.4
The FACT-L has the following five components: physical well-being, functional well-being, social well-being, emotional well-being, and the LCS. The Trial Outcome Index of FACT-L, which is the sum of the physical well-being, functional well-being, and LCS scores, is a measure of physical aspects of quality of life. A 5- to 7-point change in Trial Outcome Index score has been suggested as the minimally important difference among patients with NSCLC.5 The LCS of FACT-L is an independently validated tool that measures the disease-related symptoms of lung cancer on a scale of 0 (most symptomatic) to 28 (asymptomatic). The Eastern Cooperative Oncology Group has demonstrated that a 2- to 3-point change in LCS score reflected a minimally important difference associated with performance status, weight loss, objective tumor response, and time to progression.5 On average, a 2- to 3-point difference in LCS at baseline was found to be comparable to the difference between a patient entering the study with performance status of 0 versus a performance status of 1 or with 5% weight loss versus less than 5% weight loss. The LCS was shown to have an equal or higher prognostic value than the established prognostic factors of weight loss and performance status.5
Symptom improvement data may be a valuable measure of clinical benefit in patients with advanced NSCLC treated with gefitinib. The results of phase I studies of gefitinib in patients with solid tumors showed that changes in LCS score seemed sensitive to clinical change in patients with NSCLC.6 Evaluation of the disease-related symptom improvement rate was a coprimary end point of the phase II study of gefitinib in the United States, in which patients had to have at least one severe or between two and four mild disease-related symptoms at entry (LCS score 24 at baseline). These results have been previously reported.2 In this article, we present the results of analyses to fully explore the relationship between symptom improvement, radiographic response, and survival. We also include detailed protocol-specified analysis of symptom and quality-of-life data from the trial. The previously reported results of both the international and US phase II trials1,2 established the recommended dose of gefitinib as 250 mg/d. In this article, we include comparative data for two dose groups (250 and 500 mg/d) but focus on data at the recommended dose.
PATIENTS AND METHODS
Trial Design
The US phase II trial of gefitinib was a randomized, double-blind, parallel-group, multicenter trial that evaluated two doses of gefitinib (250 and 500 mg/d) administered orally and once daily to patients with locally advanced or metastatic NSCLC. Eligible patients had received at least two prior chemotherapy regimens including platinum and docetaxel, which were administered either concurrently or as separate treatments, and had a baseline LCS score of 24. Patients were excluded if they had received chemotherapy or irradiation within 14 days. All patients gave informed consent, and approval was obtained from the investigational review boards at each trial center.
The primary end points were rates of LCS-assessed symptom improvement and objective tumor response. Secondary end points included the disease control rate (complete response plus partial response plus stable disease), overall survival, frequency and severity of adverse events, and overall quality of life using the FACT-L instrument. Full details of the trial design and statistical methods used in this study have been described by Kris et al.2 The intent-to-treat population who received any gefitinib is analyzed here.
Symptom Improvement
Disease-related symptoms were measured weekly using the LCS of the validated quality-of-life instrument, the FACT-L questionnaire4 (Table 1). Patients completed a diary card recording the severity of symptoms on days 8, 15, and 22 of each 28-day cycle, and the LCS was evaluated at baseline and every 28 days as part of the full FACT-L questionnaire. Missing data points were counted as no change in symptoms. The maximum attainable score on the LCS is 28, where the patient is considered to be asymptomatic. Patients were eligible for the trial if they had a baseline LCS score of 24. Patients were rated as improved, no change, or worsened according to their scores compared with baseline, and their best overall response was determined. A best overall LCS response of improved was defined as an increase in LCS score of 2 points sustained for 28 days with no worsening at any interim weekly assessment.5 A best overall response of other was allocated if the data did not meet the criteria for improved, no change, or worsened or if there were missing or incomplete data. Time to symptom improvement was measured from randomization to the first visit response of improved. Patients were withdrawn from the study because of disease progression, which was documented by radiographic studies, or toxicity. The first postbaseline radiographic assessment took place after 28 days; thus, up to four LCS evaluations were obtained before patients were aware of their postbaseline disease status. All patients were evaluated during the trial for the use of the following concomitant medications that might improve pulmonary symptoms: narcotics, bronchodilators, antidepressants or anxiolytics, oxygen, systemic corticosteroids, transfusions or erythropoietin, systemic antibiotics, and cough syrups.
Quality of Life
The FACT-L questionnaire was completed at baseline, on day 28 of each treatment period, and at withdrawal; at each of these time points, the questionnaires were completed before clinical assessment and before patients were informed of their disease status. Patients were given as much time as necessary to complete the forms and did so without help from relatives. Trial Outcome Index and overall FACT-L scores were used to assess quality of life. The highest scores attainable for Trial Outcome Index and FACT-L were 84 and 136, respectively. A best overall response of improved was defined as an increase in Trial Outcome Index or FACT-L score of 6 points sustained for 4 weeks. This degree of change has been shown to be clinically meaningful.5 A best overall response of other was allocated if the data did not meet the criteria for improved, no change, or worsened or if there were missing or incomplete data.
Radiographic Response and Survival
Radiographic assessments were performed 14 days before randomization, after approximately 28 and 56 days, and every 8 weeks thereafter, in accordance with the Southwest Oncology Group-modified Union Internationale Contre le Cancer/WHO criteria.7 Overall survival was measured from the randomization date to the date of death. Patients who were alive at data cutoff were censored at the last date the patient was known to be alive.
RESULTS
Patients
A total of 216 patients were randomly assigned to either gefitinib 250 mg/d (102 patients) or gefitinib 500 mg/d (114 patients). The patient characteristics were well balanced between the two dose groups (Table 2). 2
Disease-Related Symptoms and Quality of Life at Baseline
The median and range of baseline LCS, Trial Outcome Index, and FACT-L scores for this sample and for a normative reference sample are listed in Table 3. 8 Normative scores provide a context for how this sample was characterized relative to a normative reference population of patients with lung cancer. Patients from this sample had lower LCS, FACT-L, and Trial Outcome Index baseline scores compared with normative data. The eligibility criteria for participation in the trial (LCS score 24) are consistent with the observation that the study sample had lower quality-of-life scores on these three indices compared with a normative sample. Symptoms that were most frequently severe at baseline were cough and shortness of breath (Fig 1). 2
Compliance
The compliance rates (ratio of the number of assessable weekly forms to the number of forms expected) for the LCS were 86% and 82% in patients receiving 250 and 500 mg/d, respectively. The overall compliance rates for FACT-L were 88% and 84% in patients receiving 250 and 500 mg/d, respectively.
Symptom Improvement
The symptom improvement rates were 43.1% (95% CI, 33.4% to 53.3%) for patients receiving 250 mg/d and 35.1% (95% CI, 26.4% to 44.6%) for patients receiving 500 mg/d.2 Symptom improvement rates were similar when the criterion of a 2-point change in LCS score was increased to a 3-point change (39.2% at 250 mg/d and 31.6% at 500 mg/d; Fig 2). 8 Increasing the criterion to a 7-point change (more than double the established minimally important difference) resulted in symptom improvement rates of 16.7% and 11.4% for patients receiving 250 and 500 mg/d, respectively. Similar rates of symptom improvement were observed in patients receiving gefitinib as third-, fourth-, or fifth-line and above therapy; at 250 mg/d, symptom improvement rates were 41.5%, 54.8%, and 35.7%, respectively; at 500 mg/d, these rates were 37.5%, 36.6%, and 28.0%, respectively.
Those patients who had symptom improvement were off treatment for a median of 53.5 days between stopping their last chemotherapy until the first dose of gefitinib (250 mg/d: median, 56.0 days; range, 15 to 410 days; and 500 mg/d: median, 52.0 days; range, 15 to 338 days). Similarly, those patients without symptom improvement were also off treatment for a median of 53.0 days between stopping their last chemotherapy until the first dose of gefitinib (250 mg/d: median, 50.0 days; range, 13 to 374 days; and 500 mg/d: median, 53.0 days; range, 15 to 434 days).
Symptom improvement was rapid; the median time to onset was less than 2 weeks at both doses (250 mg/d: median, 10 days; 95% CI, 8 to 22 days; and 500 mg/d: median, 9 days; 95% CI, 9 to 16 days; the times of the first postbaseline assessments). The median duration of symptom improvement for the 250 mg/d group had not been reached at 7 months, but the median duration was 5.4 months for the 500 mg/d group.
Improvement from baseline in LCS score was observed for each individual LCS item (Fig 1). Similar results were obtained for the 250 and 500 mg/d groups. Generally, more patients showed an improvement in the pulmonary items of the LCS than the nonpulmonary items. Of the 44 patients who were assessable for improvement (defined as an increase in symptom score of 1 point sustained for 28 days) in each of the LCS items at 250 mg/d, 91% had improvement in shortness of breath, 94% had improved cough, 89% had improvement in tightness in chest, and 95% had improvement in ease of breathing. Appetite, weight loss, and clear thinking improved in 84%, 88%, and 86% of patients, respectively. The mean changes in the LCS items in assessable patients with LCS improvement are listed in Table 4. Pulmonary symptoms had the greatest magnitude of change, with shortness of breath and cough being the most improved. Although pulmonary symptoms tend to co-occur in patients, occurrence of any one of pulmonary symptoms did not predict the occurrence of any other with sufficient accuracy to consider any two pulmonary symptoms as providing redundant information.
Only nine of the 44 patients with symptom improvement at 250 mg/d had received concomitant medications that might improve pulmonary symptoms after starting treatment with gefitinib but before symptom improvement occurred. An additional 32 of these 44 patients had medications added after the improvement of symptoms occurred. In comparison, 48 of the 58 patients who did not have self-reported symptom improvement had additional concomitant medications added while they were on study.
Symptom Improvement by Objective Tumor Response
The objective response rates were 11.8% and 8.8% at 250 and 500 mg/d, respectively, as previously reported.2 In addition, a further 30.4% and 27.2% of patients achieved stable disease at 250 and 500 mg/d, respectively. A correlation was observed between symptom improvement and objective tumor response. At 250 mg/d, all patients who had a partial response also had symptom improvement, as did 81% of patients who had stable disease; at 500 mg/d, symptom improvement was experienced by 90% of patients with a partial response and 61% of patients with stable disease. Of those patients with progressive disease, only 12% and 20% of patients at the 250 and 500 mg/d doses, respectively, had symptom improvement. Because symptom improvement occurred within 2 weeks in most cases (and, therefore, before first follow-up radiographic assessment), these results suggest that most of the observed symptom improvement preceded detection of objective tumor response and could not have been influenced by the patient's knowledge of clinical changes in his or her disease. The magnitude of improvement from baseline in mean on-study LCS score also correlated with the magnitude of tumor response (Table 5).
Survival by Radiographic Response and Symptom Improvement
One-year survival rates were 29% and 24% at 250 and 500 mg/d, respectively.9 Median progression-free survival time was 1.9 months (95% CI, 1.8 to 2.8 months) and 2.0 months (95% CI, 1.6 to 2.6 months) at 250 and 500 mg/d, respectively.
Longer survival was observed for patients with an objective response compared with patients without an objective response at both dose levels. At 250 mg/d, in patients who survived for 2 months, median survival time for patients with partial response was 17.9 months compared with 9.4 months for patients with stable disease and 5.2 months for patients with disease progression (Fig 3A); at 500 mg/d, the comparable median survival times were 13.6, 10.2, and 6.0 months, respectively. Patients with symptom improvement had a longer overall survival than patients without symptom improvement (13.6 v 4.6 months, respectively, at 250 mg/d, Fig 3B; and 10.5 v 5.2 months, respectively, at 500 mg/d). At 250 mg/d, those patients with symptom improvement but no partial response had a median overall survival time of 9.7 months, whereas patients with no partial response and no symptom improvement had a median overall survival time of only 4.6 months (Fig 3C). Comparable results were found at 500 mg/d. Similarly, patients with stable disease or progressive disease also had a longer overall survival if they had symptom improvement than if they did not (Fig 4A and 4B).
Physical Dimension of Quality of Life
Quality-of-life improvement rates by Trial Outcome Index (a summary score of physical and functional domains of FACT-L and LCS) were 33.3% (95% CI, 24.3% to 43.4%) for 250 mg/d patients and 20.2% (95% CI, 13.2% to 28.7%) for 500 mg/d patients. As measured by FACT-L total score (ie, overall multidimensional quality of life), improvement rates were 34.3% (95% CI, 25.2% to 44.4%) and 22.8% (95% CI, 15.5% to 31.6%) in patients treated at 250 and 500 mg/d, respectively. Of the patients who demonstrated improvement in Trial Outcome Index, 94.1% and 95.6% receiving 250 and 500 mg/d, respectively, had improved by the time of the first assessment (at 4 weeks). The FACT-L results were similar; 94.3% and 73.1% of patients on 250 and 500 mg/d, respectively, had improved by the first assessment. Improvements were durable; of those patients with Trial Outcome Index improvement, 47.1% of the 34 patients at 250 mg/d and 52.2% of the 23 patients at 500 mg/d remained improved at 3 months. This was similar for FACT-L; 45.7% of the 35 improved patients and 46.2% of the 26 improved patients at 250 and 500 mg/d, respectively, still had an improvement in quality of life at 3 months. In general, there was a relationship between the FACT-L and Trial Outcome Index best responses and the LCS best response; less than 5% of the patients with an LCS best response of improved had either a FACT-L or Trial Outcome Index best response of worsened.
As with symptom improvement, there was a correlation between radiographic regression and physical quality of life. At 250 mg/d, 92% of patients who had a partial response and 58% of patients with stable disease had an improvement in Trial Outcome Index score; at 500 mg/d, 70% of patients with a tumor response had improvements in quality of life, as did 45% of patients with stable disease (Fig 5). Similar results were seen for overall quality of life (FACT-L).
The FACT-L asked patients whether they had ever smoked, enabling a post hoc exploratory analysis of a possible relationship between smoking history and response to gefitinib. Indeed, those patients who reported they never smoked (n = 48) had a higher symptom improvement rate than former or current smokers (n = 168; 52.1% v 35.1%, respectively; P = .035).
DISCUSSION
This study demonstrates that symptomatic patients with lung cancer can reliably report changes in their symptoms and quality of life and that those changes correlate well with tumor response and survival. Symptom improvement measured using the FACT-L LCS is a robust assessment and may be a useful measure of efficacy in patients with symptomatic, advanced NSCLC. Compliance rates for LCS and FACT-L exceeded current standards in multicenter studies and allow reasonably robust conclusions to be drawn regarding LCS end points and the other end points derived from FACT-L.
The triadic analysis of response, survival, and symptom data clearly shows that those patients with symptom improvement or radiographic regression have longer survival than those patients without symptom improvement or radiographic response. At gefitinib 250 mg/d (the recommended dose), all patients with partial response also experienced symptom improvement. Of patients with stable or progressive disease, those who also experienced symptom improvement had a longer median survival time compared with those in the same tumor response category without symptom improvement. The early onset of symptom improvement seen with gefitinib, its association with partial response, progression-free survival, and overall survival, and its magnitude and duration suggest that this measured symptom improvement is an accurate rendering of the patient experience with disease symptoms in patients who are not yet influenced by any knowledge of clinical (radiographic) change. Similar results were also obtained in the international phase II trial of gefitinib.10
Many of the patients in this trial had improvements in disease-related symptoms, including the most heavily pretreated patients. Because symptom improvement generally occurred within the first 2 weeks of treatment and the first tumor assessment was at 4 weeks, patient-reported symptom improvement data were blind to tumor response results in this period. Up to four LCS evaluations were obtained before patients were aware of their postbaseline disease status. Increasing the criterion for symptom improvement from a 2-point change in LCS score to a 3-point change had little effect on the calculated symptom improvement rate. Increasing to a 7-point change (more than twice the minimally important difference for this instrument) resulted in more than 10% of patients with symptom improvement. Although only a 2-point improvement in LCS score sustained for 4 weeks signified symptom improvement, the mean LCS score improvement over the time on study (at the recommended 250 mg/d dose) was 6.1 for patients with a partial response and 3.0 for patients with stable disease.
All patients in this study had prior chemotherapy; therefore, it is possible that the symptom improvement that is noted is, in part, a result of recovery from chemotherapy side effects. However, several factors mitigate this conclusion. First, the median time since last chemotherapy dose was almost 2 months (53 days), with a minimum time since last dose of 2 weeks. Second, the median time since completing chemotherapy was no different between symptom improvers (53.5 days) and nonimprovers (53 days). Third, the majority of patients who responded symptomatically did so rapidly (within 1 week), making it unlikely that it was a reflection of gradual recovery from chemotherapy in the majority of patients. Fourth, the observed symptom improvement rate correlated with subsequent response to gefitinib treatment. Finally, there are no published data suggesting that patients with stage IIIb/IV NSCLC feel better during the weeks to months after chemotherapy discontinuation for progressive disease. Nevertheless, because this study was not controlled with a no-treatment (best supportive care) condition, we cannot fully exclude the possibility that stopping chemotherapy contributed to some degree to the symptom improvement observed in this trial.
Because of its potential for contributing important clinical information, we conducted an exploratory analysis of the impact of smoking history on symptom improvement rate and found evidence of a relationship suggesting that a higher proportion of patients who never smoked experienced symptom improvement. We emphasize that this was an exploratory, hypothesis-generating analysis that warrants further study as to possible mechanisms of action and clinical implications.
Most patients with stable disease had improvements in disease-related symptoms and quality of life. This suggests that the achievement of durable stable disease, although challenging to label a result of therapy, can be a valuable patient-centered goal in advanced NSCLC if it is tied to symptom relief. We note that most (n = 32) of the 44 symptom-responding patients had supportive medication added after symptom response. We are unable to determine why the medication was added or if the medication could have contributed to lasting improvement.
In conclusion, treatment with gefitinib offers clinically significant benefits in symptom improvement and quality of life for many heavily pretreated patients with advanced NSCLC. The results of this study indicate that symptom improvement alone is an appropriate end point for future randomized trials. Quality of life is one of the secondary end points in an ongoing international (not including the United States) phase III NSCLC study of gefitinib versus best supportive care after failure of platinum- and taxane-based therapy.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Anne Heyes, AstraZeneca; Judith S. Ochs, AstraZeneca; Michael K. Wolf, AstraZeneca; Andrea C. Kay, AstraZeneca, Novartis. Leadership Position: Roy S. Herbst, AstraZeneca. Consultant/Advisory Role: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Chandra P. Belani, AstraZeneca; Joan H. Schiller, AstraZeneca, Genentech; Mark G. Kris, AstraZeneca. Stock Ownership: Anne Heyes, AstraZeneca; Michael K. Wolf, AstraZeneca; Andrea C. Kay, AstraZeneca. Honoraria: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Diane Prager, AstraZeneca; Chandra P. Belani, AstraZeneca; Joan H. Schiller, AstraZeneca. Research Funding: David Cella, AstraZeneca; Roy S. Herbst, AstraZeneca; Thomas J. Lynch, AstraZeneca; Joan H. Schiller, AstraZeneca. Expert Testimony: Thomas J. Lynch, AstraZeneca; Mark G. Kris, AstraZeneca. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
NOTES
Presented in part at the 152nd Annual Meeting of the American Medical Association, Chicago, IL, June 12-14, 2003; and at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Fukuoka M, Yano S, Giaccone G, et al: Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol 21:2237-2246, 2003
Kris MG, Natale RB, Herbst RS, et al: Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: A randomized trial. JAMA 290:2149-2158, 2003
Cella DF, Tulsky DS, Gray G, et al: The Functional Assessment of Cancer Therapy Scale: Development and validation of the general measure. J Clin Oncol 11:570-579, 1993
Cella DF, Bonomi AE, Lloyd SR, et al: Reliability and validity of the Functional Assessment of Cancer Therapy-Lung (FACT-L) quality of life instrument. Lung Cancer 12:199-220, 1995
Cella D, Eton DT, Fairclough DL, et al: What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire Results from Eastern Cooperative Oncology Group (ECOG) Study 5592. J Clin Epidemiol 55:285-295, 2002
LoRusso PM, Herbst RS, Rischin D, et al: Improvements in quality of life and disease-related symptoms in phase I trials of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 in non-small cell lung cancer and other solid tumors. Clin Cancer Res 9:2040-2048, 2003
Green S, Weiss GR: Southwest Oncology Group standard response criteria, endpoint definitions and toxicity criteria. Invest New Drugs 10:239-253, 1992
Brucker PS, Yost K, Cashy J, Cella D, et al: Normative reference values for the FACT-G, in Cella D (ed): Functional Assessment of Chronic Illness Therapy (version 4). Evanston, IL, Evanston Northwestern Healthcare, 2004
Cella D, Natale RB, Lynch TJ, et al: Disease-related symptoms in advanced non-small-cell lung cancer as measured by the Lung Cancer Subscale of the FACT-L questionnaire: Clinically meaningful improvement with gefitinib (Iressa, ZD1839). 39th Annual Meeting of the Am Soc Clin Oncol, Chicago, IL, May 31-June 3, 2003 (poster 2531)
Gatzemeier U, Douillard JY, Kris M, et al: Rapid and durable objective responses in patients with advanced non-small-cell lung cancer in phase II trials (IDEAL 1 and IDEAL 2) treated with gefitinib. Eur J Cancer 1:S20, 2003 (suppl 1, abstr 52)
Douillard J-Y, Giaccone G, Horai T, et al: Improvement in disease-related symptoms and quality of life in patients with advanced non-small-cell lung cancer (NSCLC) treated with ZD1839 (Iressa) (IDEAL 1). Proc Am Soc Clin Oncol 21:299a, 2002 (abstr 1195)(David Cella, Roy S. Herbs)