|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 27 (September 20), 2006: pp. 4485-4491 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.07.1126 Longitudinal Study of the Relationship Between Chemoradiation Therapy for Non–Small-Cell Lung Cancer and Patient Symptoms
From the Departments of Symptom Research and Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX; and the Department of Preventive Medicine and Biometry, Colorado Health Outcomes Center, University of Colorado Health Science Center, Denver, CO Address reprint requests to Xin Shelley Wang, MD, MPH, The University of Texas M.D. Anderson Cancer Center, Department of Symptom Research, 1515 Holcombe Blvd, Unit 221, Houston, TX 77030; e-mail: swang{at}mdanderson.org
PURPOSE: Cancer patients undergoing aggressive therapy suffer from multiple nonspecific treatment-related symptoms. The goal of this prospective study was to establish a profile of the development of different symptoms over the time of therapy and to examine symptom-related functional interference in patients with non–small-cell lung cancer (NSCLC) undergoing concurrent chemoradiation therapy (CXRT). PATIENTS AND METHODS: Patients with locally advanced unresectable (stage II-IIIB) NSCLC were recruited for the study (N = 64). The M.D. Anderson Symptom Inventory (MDASI) was used to measure multiple symptoms before and weekly for 12 weeks after the start of CXRT. Mixed-effect growth curve models were used to estimate symptom development during CXRT.
RESULTS: Approximately 63% of patients suffered from moderate to severe levels of multiple symptoms by the end of CXRT. Symptom clusters with four development patterns appeared over the time of CXRT. With some variation between patients, all symptoms had a significant impact on the level of interference (all P < .001). Fatigue, distress, and sadness were the single strongest predictors of total symptom interference (each R2 CONCLUSION: Longitudinal analysis identified symptom clusters that have different development patterns in NSCLC patients receiving CXRT, providing a base for more accurate symptom management and suggesting the need for further study to identify potential mechanisms that might lead to better symptom control or prevention.
Patients experience multiple physical and psychological symptoms during and after a substantial insult, such as aggressive cancer treatment. Non–small-cell lung cancer (NSCLC) is the leading cause of cancer-related death in the United States,1 with a 5-year survival rate of only 15%. Symptoms, including those that are disease based or treatment related, are major aspects of morbidity in cancer care.2 Concurrent chemoradiation therapy (CXRT), a standard curative regimen for patients with stage III NSCLC for whom no surgery is planned,3-5 is frequently associated with acute radiation and chemotherapy effects, including hematologic suppression, pain, weight loss, esophagitis, and pneumonitis.6,7 Beyond these specific toxicities, many nonspecific symptoms (including fatigue, sadness, distress, sleep disturbance, drowsiness, and poor appetite) that are the result of treatment may also contribute to the patient's general distress during the course of treatment.8 Reducing the burden of treatment-related symptoms on a patient's functional status, especially during aggressive curative therapy, should be an important treatment goal. However, symptom control is dependent on understanding the trajectory of physical and psychological symptom development during treatment and on knowing which symptoms have the greatest impact on a patient's function. Although accurate symptom profiles are critical for establishing effective symptom management, the paucity of empirical research conducted on multiple-symptom development in patients undergoing aggressive cancer treatment9 gives rise to unanswered questions. How do clusters of symptoms vary during the course of treatment in a cohort of patients with the same diagnosis receiving the same treatment? What is the mechanism for such variation? Do physical and psychological symptoms interfere with the patient's daily life in similar ways? A study of the process of cancer-related symptom development should provide critical knowledge for developing an understanding of the mechanisms underlying these symptoms and lead to mechanism-based symptom interventions. In this study, we used a multiple-symptom assessment tool, the M.D. Anderson Symptom Inventory (MDASI), to longitudinally study cancer-related symptoms in patients with NSCLC undergoing CXRT. The MDASI is a patient-reported outcomes tool validated for use in the cancer population.10 Our primary goals were (a) to determine the levels of prevalence, severity, and longitudinal patterns of dynamic change in both physical and psychological symptoms that were present before, during, and after CXRT, and (b) to assess how each symptom affected the patient's daily activities as a function of the type and severity of the symptom.
Subjects Sixty-four patients were recruited from clinic in the Department of Radiation Oncology at The University of Texas M.D. Anderson Cancer Center in Houston, TX, between January 2002 and May 2005. To be eligible for this study, patients had to be scheduled for curative CXRT, be at least 18 years old, and have a pathologic diagnosis of NSCLC with unresectable (stage III) disease. Patients who could not understand the intent of the study, who refused to participate, or who were currently diagnosed with a major psychiatric illness were excluded from the study. The study was approved by the institutional review board of M.D. Anderson Cancer Center, and all participants gave informed consent. Patients were newly diagnosed at the start of CXRT, and all 64 completed the planned therapy. The total radiation dose was 50 to 70 Gy at 1.8 to 2 Gy per fraction daily over 5 to 7 weeks (mean, 6.5 weeks). The chemotherapy regimen administered concurrently with radiation was carboplatin plus paclitaxel.
Patient-Reported Outcomes Tool: MDASI The MDASI also contains six items that describe how much symptoms have interfered with the patient's life during the last 24 hours: general activity, mood, walking ability, normal work, relations with other people, and enjoyment of life. The interference items also are assessed on a 0 to 10 scale, with 0 being "does not interfere" and 10 being "completely interferes." An interference composite score was computed from the mean of the six individual item scores.
Statistical Analysis To estimate the change in symptom severity during and after CXRT, we used mixed effect growth-curve models with a random subject effect and linear splines to approximate the change in each symptom item within each of four periods: early (first 2 weeks of CXRT), mid (the weeks between the early and late periods), late (last 2 weeks of CXRT), and after completion of CXRT. The interpretation focuses on the average change in the outcome along the 0-to-10 scale. Observations within 12 weeks of the commencement of CXRT were included in this analysis. The same mixed effect growth-curve models were used to estimate the association of specific symptoms with interference. For this study, we used the symptom interference composite score to present the functional burden of cancer-related symptoms. The proportion of variation in symptom interference between individuals and within each individual was estimated by contrasting a model with and without the time-varying measures of symptom severity. To define the impact of symptom severity level on interference, the prediction of total symptom interference score as a function of level of symptom severity (0 to 3, 3 to 5, 5 to 7, or 7 to 10 on the 0-to-10 scale) of each symptom was calculated.
Patient and Treatment Characteristics Patient demographics are shown in Table 1. All participants had stage III NSCLC at enrollment; 3% subsequently developed stage IV disease but remained on study. All patients had good Eastern Cooperative Oncology Group (ECOG) performance status (0-1) before the start of CXRT, although ECOG scores of 2 were seen in 18% of patients during therapy and in 26% of patients after therapy. Approximately 10% of eligible patients approached declined to participate.
The missing data rate ranged from 8.4% to 8.8% for the MDASI symptom items, stemming primarily from either patient fatigue or administrative error. All of the patients contributed symptom data at baseline and during CXRT, but 10 patients (15% of data) either withdrew from the study or were lost to follow-up after completing CXRT. Of these, two died as a result of pneumonia within 1 to 2 weeks of completing CXRT. Data on toxicities were not used in the analysis because records were not consistent.
Prevalence of Moderate to Severe Symptoms and Symptom Interference On average, the highest levels of symptom interference were seen in general activity and work (including work around the house) during and after therapy (Table 2). With the exception of enjoyment of life, all measures of symptom interference increased significantly during and after the course of treatment.
Longitudinal Patterns of Symptom Development Four clusters of symptoms demonstrated different development patterns during and after treatment when we estimated the average rate of symptom change across the four therapy periods (early, mid, late, and after therapy). Table 3 shows the rate of weekly change for each symptom over the four periods. Figure 1 presents longitudinal symptom patterns by cluster during and after CXRT.
In pattern 1, we observed that the rates of change for both pain and sore throat steadily and significantly increased in severity over the course of therapy (early, mid, and late periods of CXRT, weeks 1 through 7), and then decreased in the post-CXRT period. Pain increased significantly in the early (0.32 point on a 0-to-10 scale; SE, 0.17), mid (0.49 point), and late (0.46 point) CXRT periods. The two symptoms were correlated, with the severity of sore throat explaining 38% of the variation in reported pain. Sore throat and pain most likely result from radiation-induced inflammation in local tissue. Pattern 2 showed a rapid increase in therapy-related GI toxicities, including nausea and vomiting, in the early and mid CXRT periods. Pattern 3, seen in nonspecific symptoms that included fatigue, lack of appetite, drowsiness, sleep disturbance, dry mouth, and distress, reflected an increase in severity in both the early and late therapy periods. Pattern 4 included affective symptoms (sadness), cognitive issues (difficulty remembering), and other disease- or treatment-related symptoms (shortness of breath, cough, and numbness) for which we observed no major change in severity over the course of therapy.
Impact of Each Symptom on Interference
In a cohort of patients with the same disease and treatment background, should we expect only a small variation in symptom interference among all patients, and will they share a similar pattern of symptom development over the time of CXRT? The proportion of variation in symptom interference between patients and within each patient from week to week was estimated by contrasting models with and without the time-varying measures of symptom severity. For most items, the between-subject variation in symptom interference explained more of the overall level of interference than the week-to-week within-subject variation (Table 4; eg, R2 = 0.62 v R2 = 0.40 for fatigue, R2 = 0.66 v R2 = 0.25 for sadness, R2 = 0.63 v R2 = 0.29 for distress, respectively). Fatigue, sadness, distress, and drowsiness explained the greatest proportion of the variation in overall levels of interference between subjects. The symptom that explained the most week-to-week variation in interference within subjects was fatigue (R2 = 0.40). Does every symptom create its greatest interference in a patient's daily life only when it is severe? When we categorized reported symptoms as mild, moderate, or severe, we found that these three categories had an unequal impact on interference by specific symptom (Table 5). Some symptoms, such as fatigue, drowsiness, nausea, and sore throat, had the greatest impact on interference scores when patients reported them at a severe level. For example, patient-reported ratings of interference increased almost 1 point (0.91) on the 0-to-10 scale for every point increase in fatigue in the severe range (7 to 10) but increased only a half point for every point increase in fatigue in the mild to moderate range (3 to 7). Other symptoms, including pain, lack of appetite, shortness of breath, cough, and drowsiness, had a relatively constant effect on interference over the mild to severe range (3 to 10). In contrast to these physical symptoms, the two affective symptoms (distress and sadness) had a stronger impact at lower severity levels. Sadness significantly affected interference only when it was mild (0 to 5 on the 0-to-10 scale). Finally, a few symptoms (difficulty remembering, dry mouth, vomiting, and numbness) had a constant significant effect on symptom interference over the entire 0-to-10 range of symptom severity (all P < .001).
This study presents a method of longitudinal symptom data acquisition and analysis that allows for the examination of the temporal pattern of emergence of several symptoms over the course of aggressive cancer therapy. Although most of the patients appeared to tolerate their therapy well, our study showed that by the end of CXRT, the peak of symptom severity, approximately 63% of patients experienced moderate to severe levels of multiple treatment-related symptoms. Further, four clusters of symptoms with unique temporal relationship to the course of CXRT therapy were identified in this homogenous NSCLC patient sample. We also found that specific symptoms and specific symptom clusters produced significantly more symptom interference in this patient sample. The evidence from this study suggests that accumulated effects over time from combined radiation and chemotherapy could severely affect a patient's daily functioning, evidencing the need for better and more efficient symptom management. One symptom cluster (pattern 3) included several nonspecific symptoms (fatigue, lack of appetite, drowsiness, sleep disturbance, dry mouth, and distress) that showed similar development patterns not characterized by a steady linear increase with accumulated dose during CXRT. This symptom cluster had the strongest impact on patients' daily functioning compared with other symptom clusters, yet the effect of these symptoms is often ignored in patient care. The variation in symptom interference between patients was larger than the week-to-week within-subject variation, indicating that certain patients were at greater risk for symptom development than others throughout their course of treatment. A symptom mechanism study might provide the explanation for such differences in patient response to CXRT. At a lower severity level, affective symptoms (distress and sadness) contributed more significantly to symptom interference than did physical symptoms. The severity distributions of these symptoms were not skewed, so the reason for this phenomenon is unknown. Nevertheless, this result indicates that symptom intervention for affective symptoms may need to be initiated at a lower symptom-severity threshold. Given that fatigue is a common disease-induced symptom in lung cancer11,13 and is a result of CXRT regardless of cancer type,14-16 it is an especially important symptom to examine when both factors are combined, as in the setting of NSCLC treated by CXRT. This study found that fatigue was the most severe symptom at baseline and over the course of CXRT and did not return to baseline levels even 5 to 6 weeks after the completion of CXRT. Fatigue also had the highest predictive value for interference in the patient's daily life. Accordingly, fatigue is a reasonable target for symptom management. Worsening pain and sore throat (pattern 1) during CXRT evidences the need for better management for acute esophagitis, a common complication of radiation for NSCLC.17,18 Because the volume effect from the radiation is needed to control local recurrence of the tumor and its spread to the local lymph nodes, an active plan for routinely using therapeutic agents against radiation-induced esophagitis should be integrated into the CXRT treatment course for NSCLC. For example, topical anesthetics as well as liquid hydrocodone and acetaminophen could be used beginning the second week of therapy. Once symptom reports become severe, a fentanyl transdermal patch could provide better pain control. Hydration and consultation with a dietician are also key in the management of esophagitis. There are methodologic barriers in symptom research, including inappropriate assessment tools, inappropriate schedules over the course of therapy, or lack of a suitable statistical modeling method. The MDASI, a brief, multiple-symptom assessment tool, can be completed with minimal effort by symptomatic cancer patients, a factor that is especially important for a longitudinal study in which repeated measures are required. The mixed-effects statistical model was used to handle time-varying covariates and other unbalanced designs in this longitudinal study, in preference to examining simple cross-sectional correlations where the various components are mathematically indistinguishable. Our study had certain limitations. First, inconsistent documentation of treatment-related toxicities did not allow us to compare patient-reported symptoms with items more typically evaluated by clinicians. Second, we did not use additional measures for cognitive testing of anxiety and depression. In future studies, such measures might be helpful for further understanding differences in physical, cognitive, and psychological symptom development over the course of therapy. The longitudinal symptom profile findings of this study indicate avenues for more accurate methods for patient monitoring that might lead to development of better clinical guidelines, and better information for patients as to when to expect the greatest impact from their symptoms. Although there is increasing literature on symptom clusters in cancer patients,19,20 many of the studies are cross-sectional. Identification of the developmental time course of symptom clusters and its impact on cancer patients undergoing aggressive therapies such as CXRT will provide greater understanding of the patterns of association and interactions of symptoms. The temporal patterning of these symptom clusters raises the possibility that the clusters might have somewhat different biologic mechanisms. This is worthy of further study, which could lead to mechanism-driven symptom management strategies that provide less distressing and more tolerable treatment.
The authors indicated no potential conflicts of interest.
We acknowledge the assistance of Maria Sanchez, LVN, and Yanira Lara, LVN, Marilyn Morrissey, MPH, Jeanie Woodruff, and Lorraine M. Cherry, PhD.
Supported in part by National Institutes of Health Grants No. R01 CA026582 and R21 CA109286. Presented in part at the 12th Annual Conference of the International Society of Quality of Life, San Francisco, CA, October 10-12, 2005. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55:10-30, 2005 2. Cleeland CS: Cancer-related symptoms. Semin Radiat Oncol 10:175-190, 2000[CrossRef][Medline] 3. Ohno T, Kato S, Wakatsuki M, et al: Incidence and temporal pattern of anorexia, diarrhea, weight loss, and leukopenia in patients with cervical cancer treated with concurrent radiation therapy and weekly cisplatin: Comparison with radiation therapy alone. Gynecol Oncol [Epub ahead of print March 6, 2006] 4. Sunpaweravong P, Magree L, Rabinovitch R, et al: A phase I/II study of docetaxel, etoposide, and carboplatin before concurrent chemoradiotherapy with cisplatin and etoposide in limited-stage small cell lung cancer. Invest New Drugs 24:213-221, 2006[CrossRef][Medline] 5. Yamamoto N, Nishimura Y, Nakagawa K, et al: Phase I/II study of weekly docetaxel dose escalation in combination with fixed weekly cisplatin and concurrent thoracic radiotherapy in locally advanced non-small cell lung cancer. Cancer Chemother Pharmacol 58:285-291, 2006[CrossRef][Medline] 6. Krupnick AS, Kreisel D, Hope A, et al: Recent advances and future perspectives in the management of lung cancer. Curr Probl Surg 42:540-610, 2005[CrossRef][Medline] 7. Choy H, Kim JS, Pyo H, et al: Topoisomerase I inhibitors in the combined modality therapy of lung cancer. Clin Lung Cancer 2:S34-S40, 2001 (suppl)[Medline] 8. Cox JD, Stetz J, Pajak TF: Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 31:1341-1346, 1995[CrossRef][Medline] 9. Patrick DL, Ferketich SL, Frame PS, et al: National Institutes of Health State-of-the-Science Conference Statement: Symptom Management in Cancer: Pain, Depression, and Fatigue, July 15-17, 2002. J Natl Cancer Inst 95:1110-1117, 2003 10. Cleeland CS, Mendoza TR, Wang XS, et al: Assessing symptom distress in cancer patients: The MD Anderson Symptom Inventory. Cancer 89:1634-1646, 2000[CrossRef][Medline] 11. Mendoza TR, Wang XS, Cleeland CS, et al: The rapid assessment of fatigue severity in cancer patients: Use of the Brief Fatigue Inventory. Cancer 85:1186-1196, 1999[CrossRef][Medline] 12. Serlin RC, Mendoza TR, Nakamura Y, et al: When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain 61:277-284, 1995[CrossRef][Medline] 13. Pater JL, Zee B, Palmer M, et al: Fatigue in patients with cancer: Results with National Cancer Institute of Canada Clinical Trials Group studies employing the EORTC QLQ-C30. Support Care Cancer 5:410-413, 1997[CrossRef][Medline] 14. Cleeland CS, Wang XS: Measuring and understanding fatigue. Oncology 13:91-97, 1999[Medline] 15. Hickok JT, Morrow GR, Roscoe JA, et al: Occurrence, severity, and longitudinal course of twelve common symptoms in 1129 consecutive patients during radiotherapy for cancer. J Pain Symptom Manage 30:433-442, 2005[CrossRef][Medline] 16. Wang XS, Janjan NA, Guo H, et al: Fatigue during preoperative chemoradiation for resectable rectal cancer. Cancer 92:1725-1732, 2001[CrossRef][Medline] 17. Bradley J, Deasy JO, Bentzen S, et al: Dosimetric correlates for acute esophagitis in patients treated with radiotherapy for lung carcinoma. Int J Radiat Oncol Biol Phys 58:1106-1113, 2004[CrossRef][Medline] 18. Fournel P, Robinet G, Thomas P, et al: Randomized phase III trial of sequential chemoradiotherapy compared with concurrent chemoradiotherapy in locally advanced non-small-cell lung cancer: Groupe Lyon-Saint-Etienne d'Oncologie Thoracique-Groupe Français de Pneumo-Cancerologie NPC 95-01 Study. J Clin Oncol 23:5910-5917, 2005 19. Barsevick AM, Whitmer K, Nail LM, et al: Symptom cluster research: Conceptual, design, measurement, and analysis issues. J Pain Symptom Manage 31:85-95, 2006[CrossRef][Medline] 20. Miaskowski C, Dodd M, Lee K: Symptom clusters: The new frontier in symptom management research. J Natl Cancer Inst Monogr 32:17-21, 2004[Medline] Submitted April 21, 2006; accepted July 19, 2006.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
0.49). Physical symptoms had greater impact on interference with function when they were moderate to severe, whereas affective symptoms had the largest effect on interference when they were mild to moderate. 
