|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.03.074 on August 9 2004 © 2004 American Society of Clinical Oncology. Prognostic Factors of Thin Cutaneous Melanoma: An Analysis of the Central Malignant Melanoma Registry of the German Dermatological SocietyFrom the Department of Dermatology, Skin Cancer Program, Central Malignant Melanoma Registry of the German Dermatological Society, Eberhard-Karls-University, Tuebingen, Germany; Skin Cancer Research Group, School of Public Health and Tropical Medicine, James Cook University, Townsville, Australia Address reprint requests to Professor Claus Garbe, MD, Department of Dermatology, Eberhard-Karls-University, Liebermeisterstrasse 25, D-72076 Tuebingen, Germany; e-mail: claus.garbe{at}med.uni-tuebingen.de
PURPOSE: The increasing number of thin cutaneous melanomas (CM) with tumor thickness up to 1 mm demands a detailed analysis of prognostic factors for the classification and grading of these tumors. The aim of the present study was to identify prognostic factors in thin CM. PATIENTS AND METHODS: A series of 12,728 patients with thin incident primary invasive CM and follow-up data recorded between 1976 and 2000 by the German-based Central Malignant Melanoma Registry was analyzed using the multivariate Cox proportional hazard model to evaluate prognostic factors, and classification and regression trees analysis (CART) to define prognostic groups. RESULTS: Multivariate analysis found tumor thickness, sex, age, body site, and histopathologic subtype to be significant prognostic factors of thin CM. Ulceration and regression did not affect prognosis significantly. Prognostic classification based on the results of CART analysis resulted in three groups defined by tumor thickness, age, and sex. Ten-year survival rates of these groups varied between 91.8% and 98.1%, with improved classification as compared with subgroups by tumor thickness alone. CONCLUSION: Classification by tumor thickness identified prognostic subgroups with highest significance in thin CM, and the classification was improved by the introduction of age and sex. However, neither ulceration nor the level of invasion included in the new American Joint Committee on Cancer TNM system classification, revealed statistical significance as prognostic factors in thin CM.
Diagnosing cutaneous melanoma (CM) at an early stage is the key to improved prognosis. In most Western countries, this knowledge has encouraged prevention and early detection.1,2 Successfully implemented intervention programs led to an increased consciousness of risk factors in the general population and to respective modifications of behavior.3 As a likely consequence, a decrease of the median Breslow’s tumor thickness, and an increase in the diagnosis of thin CM, have been observed worldwide.4-11 In Central Europe, the median tumor thickness significantly decreased from 1.8 mm in 1976 to 0.5 mm in 2000, while the percentage of thin CM (tumor thickness  1 mm) increased from 39.0% in 1976 to 65.5% in 2000 (Buettner et al, submitted). There is therefore a growing need to focus on patients with thin primary CM, and to determine prognostic factors for these patients.9,12-15
The American Joint Committee on Cancer (AJCC) recently established a new staging system based on a large study that included 17,600 CM patients from 13 different cooperating centers.16,17 Within this new staging system, CM with a tumor thickness The present study is based on 12,728 patients with incident primary CM with thickness 1.00 mm or less. These patients were documented by the German-based Central Malignant Melanoma Registry (CMMR) of the German Dermatologic Society, to which more than 80 clinical centers throughout Germany, Austria, and Switzerland report, and which comprises more than 30% of all German incident invasive CM cases. The analysis of these data aimed to identify independent significant prognostic factors of thin CM and establish a more detailed prognostic classification for the growing population of patients with thin CM.
Patients The present study was based on a total of 12,728 patients with incident invasive (Clark’s level of invasion  II) primary CM, with a tumor thickness of 1.00 mm or less. These patients were documented between January 1, 1976, and December 31, 2000, by the CMMR. The data of the CMMR are not population based. Nevertheless, the data can be considered representative because patients of a respective area that is covered by one center are recorded almost completely. In Germany, the majority of patients are referred to the hospitals, and cooperative surveillance of the patients between the hospital and the dermatologist in private practice is conducted. The Central Malignant Melanoma Registry records approximately 35% to 50% of all melanoma patients in Germany. Informed consent was obtained from all patients. Tumor characteristics and case history were recorded in a standardized manner, and patients were examined regularly every 3 to 6 months for a period of 10 years. In Germany, follow-up examinations are recommended to be performed every 3 months during the first 5 years after resection of the primary tumor, and every 6 months during the 6 to the 10 years after surgery; examinations comprise a physical examination and once-yearly lymph node ultrasound, abdomen ultrasound, chest x-ray, and blood examination.25 All patients included in the present analysis had a follow-up time of at least 3 months. Follow-up time was cut to a maximum of 10 years, as patients usually participate in the CM follow-up program for a maximum of 10 years,26 while afterwards, only patients with progressive disease are regularly documented, and little information can be retrieved for patients without recurrences. The body site of the primary CM was classified into five anatomic sites: head, scalp, and neck; anterior trunk; posterior trunk; upper extremities; and lower extremities. According to a classification described in a study published in 1995, body site was also subdivided according to the thorax, upper arm, neck, and scalp (TANS) concept.27 CM localized at the head or neck, back, breast, and upper arm were classified as TANS regions, while CM of the lower trunk, lower extremities, forearm, and hands were non-TANS regions. Tumor thickness was analyzed in four subgroups ( 0.25 mm, 0.26 to 0.5 mm, 0.51 to 0.75 mm, and 0.76 to 1.00 mm). Age was classified into five subgroups (< 40 years, 40 to 49 years, 50 to 59 years, 60 to 69 years, and 70 years). Ulceration was diagnosed by histopathology and was defined as the absence of intact epidermis overlying the major portion of primary melanoma. Histopathologic reports of the responsible dermatopathologists of the respective hospitals were documented within the present study. No independent review process of the histopathologic reports was performed; however, there is a continuous medical education of dermatopathologists as organized by the German Dermatologic Society.
Statistical Analysis Multivariate Cox proportional hazard models were used to analyze the prognostic impact of age, sex, Breslow’s tumor thickness, Clark’s level of invasion, ulceration, regression, histologic subtypes, and body sites (anatomic and TANS). Forward and backward stepwise procedures of the multivariate modeling process resulted in the same model. There were no confounding effects detected. The Cox model was described by means of relative risks together with 95% CIs and P values. Throughout the analysis, P values less than .05 were considered statistically significant. Explorative classification and regression tree (CART) analysis was used to define meaningful prognostic groups with respect to survival probability. During CART analysis, first, the entire sample, and thereafter, all newly defined subgroups, were investigated at every step of the analysis to determine which variable yielded the most significant division into prognostic groups with respect to estimates of survival probabilities according to Kaplan-Meier, judged by means of log-rank test statistics. The CART procedure stopped when either no further significant factors were detected or when the sample size was below 1% (total sample size). The resulting final groups were most homogenous with respect to prognosis; that is, some groups presented with low survival probabilities, and other groups, with high survival probabilities. Final groups defined by CART analysis were presented with 10-year survival probabilities (S10) and approximate 95% CIs.
Description of Sample The sample consisted of 41.4% male patients, and the mean age at diagnosis was 50 years (SD ± 15.7). The mean tumor thickness was 0.58 mm (SD ± 0.23), and the median tumor thickness was 0.57 mm (IQR = 0.40 to 0.77). The distribution of all prognostic factors of thin primary CM considered are presented in Table 1. Overall, patients with thin CM had a median follow-up time of 4.0 years (IQR = 2 to 77.1). The disease-related mortality was 1.4% (n = 165). The overall 5-year survival rate was 98.8% (95% CI, 98.5% to 99.0%), and the overall 10-year survival rate was 96.5% (95% CI, 95.9% to 97.2%). Patients who died of CM (n = 165) had a median follow-up of 4.4 years (IQR = 2.4 to 6.5), while patients who were considered censored cases had a median follow-up of 4.0 years (IQR = 2.0 to 7.1). Follow-up information was available for a period between 5 and 10 years for 31.8% of patients, and for 10 years for 9.2% of patients.
Results of Comparisons of Survival Probabilities According to Kaplan-Meier Cumulative survival rates were significantly different for age (P = .0175), sex (P < .0001), tumor thickness (P < .0001), level of invasion (P = .0025), histologic subtype (P = .0010), and body site (P = .0110 and P = .0026 for TANS; Table 2). The four categories of tumor thickness ( 0.25 mm, 0.26 to 0.50 mm, 0.51 to 0.75 mm, and 0.76 to 1.00 mm) resulted in well-separated survival rates (Fig 1). Ulceration (P = .2580) and regression (P = .9051) were nonsignificant factors.
Results of Multivariate Cox Proportional Hazard Modeling Tumor thickness was identified as the strongest significant prognostic factor (Table 3). Tumors with a thickness between 0.76 and 1.00 mm had a relative risk of 3.9 (95% CI, 2.6 to 5.8) compared with tumors 0.5-mm thick or thinner. Other independent prognostic factors were age, sex, histologic subtype, and body site (also when categorized as TANS: relative risk = 1.5; 95% CI, 1.1 to 2.2; model not shown). The results of the final Cox proportional hazard model were based on 11,927 patients. Initially, all potential prognostic factors were considered in the modeling process, reducing the sample size considerably. However, because neither ulceration nor regression showed any significant effects on survival, these characteristics were not included in the final model. In the multivariate analysis, ulceration, level of invasion, and regression failed to be independent significant prognostic factors.
Results of CART Analysis CART analysis based on 11,688 CM cases with complete data sets resulted in eight final prognostic groups (Fig 2). Separate missing variables were defined for ulceration and regression, allowing the CART analysis to deduce the patients with missing data for the two variables. Tumor thickness, with the cutoff of 0.75 mm, was the most significant factor within the entire collective. In CM 0.75-mm thick or thinner, patients younger than 60 years with level II CM were the group with the best prognoses (n = 2,957; S10 = 98.8%; 95% CI, 97.9 to 99.6), while patients 60 years or older with CM on the posterior trunk were the group with worst prognosis (n = 675; S10 = 93.3%; 95% CI, 89.0 to 97.6). In CM 0.76 mm to 1.00 mm thick, female patients with no nodular CM formed the group with best prognosis (n = 1,517; S10 = 96.8%; 95% CI, 95.2 to 98.4), while male patients with acral lentigenous melanoma were the group with worst prognosis (n = 25; S10 = 80.9%; 95% CI, 61.0 to 100). Table 4 gives the definition of three proposed prognostic subgroups for patients with primary CM with tumor thickness 1.00 mm or thinner based on the result of the CART analysis. Overall and pairwise, the three final groups were significantly different with respect to survival (overall: P < .0001; high survival probablility v medium: P = .0017; high survival probability v low: P < .0001; medium survival probability v low: P < .0001). Survival rates for each subgroup are shown in Figure 3.
In the present study, the multivariate evaluation revealed tumor thickness, histologic subtypes, body site, sex, and age as significant prognostic factors, whereas ulceration, regression, and Clark’s level of invasion did not seem to have an independent significant effect on the prognosis of thin CM. Ulceration and regression were not significant factors in the bivariate analysis using the Kaplan-Meier model. Concerning the level of invasion, the bivariate analysis revealed significantly lower 10-year survival probabilities for level III and IV CM compared with level II, but level of invasion did not prove to be an independent prognostic factor in the multivariate analysis.
Overall, tumor thickness dominated as the strongest prognostic factor. The subgroup of 0.76- to 1.00-mm thickness was significantly worse than the group with 0.51- to 0.75-mm thickness, and both groups showed an increased risk (3.9 and 1.9, respectively) compared with those CM with a Similar to the results for incident primary CM of all sizes, the classifications of body site according to anatomic regions, as well as in TANS/non-TANS regions, proved to be significant independent prognostic factors for thin tumors.27 Concerning the histologic subtypes, acral lentigineous melanoma was associated with an increased relative risk of death in thin CM, though the number of acral lentigineous melanoma was relatively low, with only 1.9%. Concordant with previously published results, the effect of sex seems to be an important independent prognostic factor in thin CM, with a better outcome for female patients.20 The prognostic factors of thin CM as established by the present analysis were similar to the prognostic factors for primary CM of all sizes.27
In this analysis of 12,728 patients, data on ulceration and regression were documented in the great majority of the patients analyzed, and approximately 9,000 patients could be included in the multivariate analysis with complete information on all potential prognostic factors. Of these, only 229 patients (2.5% of all cases) showed histologic ulceration. In the present study, the percentage of thin CM with ulceration was lower as compared with two studies with rates of ulcerated tumor between 4.2% and 6.0%.28 Possibly, the percentage of primary tumors that were very thin tumors (44.7% McKinnon et al published a study including 2,746 patients from the Sydney Melanoma Unit.28 In this study, ulceration proved to be an independent significant prognostic factor in the multivariate Cox regression analysis for thin CM (P = .0045). Patients with ulcerated CM had a 10-year survival rate of 84%, compared with 92.3% for tumors without ulceration (P = .0002). Again, the outcome of these patients seems to be less favorable as compared with the present study. The databases reported by Balch and by McKinnon seem not to be independent, as the AJCC Melanoma Database for thin CM consisted in great part of data from the Sydney Melanoma Unit. The CART analysis of the present data resulted in a pruned regression tree with eight final subgroups. Tumor thickness was found to be the strongest factor on the first level of the regression tree, which was concordant with the results of the multivariate analysis. Consideration of additional prognostic factors contributed to a better estimation of the patient’s outcome. In tumors with a thickness of more than 0.75 mm, sex played an important role for the prognosis. Men with CM with a greater than 0.75-mm thickness had a worse prognosis than did female patients with the same thickness. This subgroup with the most unfavorable prognosis (males with tumor thickness > 0.75 mm) may be considered for sentinel node biopsy. In patients with a tumor thickness of 0.75 mm or less, age was an important factor in the second level of the regression tree. This second level of the CART analysis provided the basis for the suggested clinical classification of patients with thin CM. Ten-year survival rates among the defined prognostic groups ranged from 91.8% to 98.1%. For a further characterization of thin CM, this suggested that classification allows the definition of more homogenous subgroups for comparisons and could easily be considered in the follow-up of thin CM. Further studies are required to address whether the difference in survival between the prognostic groups may be clinically significant. Published studies focusing on prognostic factors of thin CM showed variant results and proved to be difficult to compare. In all studies considered, there was general agreement that tumor thickness is an independent prognostic factor for thin CM,6,16,17,20,24,27,28,30-34whereas body site, level of invasion, histologic subtype, sex, and age showed contradicting results.
The level of invasion was found to be a significant independent prognostic factor in 585 patients with CM In contrast to our analysis, sex was not a significant prognostic factor in the studies of Balch and McKinnon.28,29 In contrast, the analysis from the Scottish Melanoma Registry showed sex to be independently influencing survival in the entire collective; however, the subgroup of patients with CM having a thickness of 1.5 mm or less was not further analyzed.19 In concordance with our results, age proved to be an independent prognostic factor in the multivariate Cox regression analyses performed by McKinnon et al on 2,746 patients with thin CM, and also in the Scottish study of 6,288 patients with CM of all sizes.19,28 Body site also proved to be an independent prognostic factor in the multivariate analyses of the Swedish study 29 the Scottish study,19 and the study by Balch et al.24 McKinnon did not further investigate body site or histologic subtype.28 Histologic subtype proved to be a significant prognostic factor in the study conducted by Owen et al, but was nonsignificant or not investigated in the other studies cited.19 Overall, patients with thin CM have an extremely high survival probability. Thus, for a factor to show significance in a multivariate model, either the effect of the factor or the sample size needs to be extremely large. Thus, variations in the composition of the samples of patients might result in one factor being significant over other factors in one study, but not in other studies. Also, variations in the composition of the samples of patients might in itself lead to contradicting results. For example, in the present sample of patients, ulceration was less prevalent (probably because patients presented with very thin tumors), and thus, less likely to be significant.
In conclusion, in thin CM (
The authors indicated no potential conflicts of interest.
The following Departments of Dermatology were major contributors (> 500 cases) to the Central Malignant Melanoma Registry of the German Dermatological Society:University of Berlin (Charité) (Chair: W. Sterry), Free University of Berlin (Chair: C.E. Orfanos), Berlin-Neukölln (Chair: P. Kohl), Chemnitz (Chair: J. Koch), Dortmund (Chair: P. Frosch), Dresden-Friedrichstadt (Chair: U. Wollina), University of Dresden (Chair: M. Meurer), Erfurt (Chair: R. Linse), University of Erlangen (Chair: G. Schuler), University of Essen (Chair: H. Goos), University of Freiburg (Chair: L. Bruckner-Tudermann), Gera (Chair: J. Meyer), University of Göttingen (Chair: Ch. Neumann), University of Graz (Chair: H. Kerl), University of Greifswald (Chair: M. Jünger), University of Hamburg (Chair: I. Moll), Hamburg-St. Georg (Chair: C. Sander), Hildesheim (Chair: H. Vakilzadeh) University of Heidelberg (Chair: D. Petzold), University of Heidelberg/Mannheim (Chair: S. Goerdt), University of Homburg (Chair: W. Tilgen), University of Jena (Chair: P. Elsner), Kassel (Chair: R. Rompel), University of Kiel (Chair: E. Christophers), Krefeld (Chair: S. Wassilew), University of Leipzig (Chair: J. Simon), University of Lübeck (Chair: H.H. Wolf), University of Magdeburg (Chair: H. Gollnick), Minden (Chair: R. Stadler), University of Munich (Chair: G. Plewig), University of Münster (Chair: Th. Luger), Nürnberg (Chair: E. Paul), University of Tuebingen (Chair: M. Röcken), University of Ulm (Chair: K. Scharfetter-Kochanek), Wiesbaden (Chair: J. Metz), University of Würzburg (Chair: E.B. Bröcker), University of Halle (Chair: W. Marsch), University of Zürich (Chair: G. Burg).
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
1. Koh HK, Norton LA, Geller AC, et al: Evaluation of the American Academy of Dermatology’s National Skin Cancer Early Detection and Screening Program. J Am Acad Dermatol 34:971-978, 1996[CrossRef][Medline]
2. Koh HK, Geller AC, Miller DR, et al: Prevention and early detection strategies for melanoma and skin cancer: Current status. Arch Dermatol 132:436-443, 1996 3. Melia J, Cooper EJ, Frost T, et al: Cancer Research Campaign health education programme to promote the early detection of cutaneous malignant melanoma, II: Characteristics and incidence of melanoma. Br J Dermatol 132:414-421, 1995[CrossRef][Medline] 4. Balch CM, Soong SJ, Milton GW, et al: Changing trends in cutaneous melanoma over a quarter century in Alabama, USA, and New South Wales, Australia. Cancer 52:1748-1753, 1983[CrossRef][Medline]
5. Dennis LK: Analysis of the melanoma epidemic, both apparent and real: Data from the 1973 through 1994 surveillance, epidemiology, and end results program registry. Arch Dermatol 135:275-280, 1999 6. Garbe C, McLeod GR, Buettner PG: Time trends of cutaneous melanoma in Queensland, Australia and Central Europe. Cancer 89:1269-1278, 2000[CrossRef][Medline] 7. Garbe C, Blum A: Epidemiology of cutaneous melanoma in Germany and worldwide. Skin Pharmacol Appl Skin Physiol 14:280-290, 2001[CrossRef][Medline] 8. Herd RM, Cooper EJ, Hunter JA, et al: Cutaneous malignant melanoma: Publicity, screening clinics and survival—The Edinburgh experience 1982-90. Br J Dermatol 132:563-570, 1995[Medline]
9. Jemal A, Devesa SS, Hartge P, et al: Recent trends in cutaneous melanoma incidence among whites in the United States. J Natl Cancer Inst 93:678-683, 2001 10. Kittler H, Binder M, Wolff K, et al: A ten-year analysis of demographic trends for cutaneous melanoma: Analysis of 2501 cases treated at the University Department of Dermatology in Vienna (1990-1999). Wien Klin Wochenschr 113:321-327, 2001[Medline]
11. MacLennan R, Green AC, McLeod GR, et al: Increasing incidence of cutaneous melanoma in Queensland, Australia. J Natl Cancer Inst 84:1427-1432, 1992 12. Baccard M, Havard S, Souques M: Prospective study of the incidence of melanoma in the Paris region in 1994: The PETRI Melanoma Group. Melanoma Res 7:335-338, 1997[CrossRef][Medline] 13. Marks R: The public health approach to the burden of common skin diseases in the community. J Dermatol 28:602-605, 2001[Medline] 14. Marrett LD, Nguyen HL, Armstrong BK: Trends in the incidence of cutaneous malignant melanoma in New South Wales, 1983-1996. Int J Cancer 92:457-462, 2001[CrossRef][Medline] 15. van der Rhee HJ, van der Spek-Keijser LM, van Westering R, et al: Increase in and stabilization of incidence and mortality of primary cutaneous malignant melanoma in Western Netherlands, 1980-95. Br J Dermatol 140:463-467, 1999[CrossRef][Medline] 16. Balch CM, Soong S, Ross MI, et al: Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm): Intergroup Melanoma Surgical Trial. Ann Surg Oncol 7:87-97, 2000[CrossRef][Medline]
17. Balch CM, Buzaid AC, Soong SJ, et al: Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 19:3635-3648, 2001
18. Buzaid AC, Ross MI, Balch CM, et al: Critical analysis of the current American Joint Committee on Cancer staging system for cutaneous melanoma and proposal of a new staging system. J Clin Oncol 15:1039-1051, 1997 19. Owen SA, Sanders LL, Edwards LJ, et al: Identification of higher risk thin melanomas should be based on Breslow depth not Clark level IV. Cancer 91:983-991, 2001[CrossRef][Medline] 20. Buttner P, Garbe C, Bertz J, et al: Primary cutaneous melanoma: Optimized cutoff points of tumor thickness and importance of Clark’s level for prognostic classification. Cancer 75:2499-2506, 1995[CrossRef][Medline] 21. Heenan PJ, Yu L, English DR: Melanoma in the elderly: A neglected public health challenge. Med J Aust 170:394-395, 1999 22. Karakousis CP, Balch CM, Urist MM, et al: Local recurrence in malignant melanoma: Long-term results of the multiinstitutional randomized surgical trial. Ann Surg Oncol 3:446-452, 1996[CrossRef][Medline] 23. Keefe M, Mackie RM: The relationship between risk of death from clinical stage 1 cutaneous melanoma and thickness of primary tumour: No evidence for steps in risk—Scottish Melanoma Group. Br J Cancer 64:598-602, 1991[Medline] 24. Mansson-Brahme E, Carstensen J, Erhardt K, et al: Prognostic factors in thin cutaneous malignant melanoma. Cancer 73:2324-2332, 1994[CrossRef][Medline]
25. Garbe C, Paul A, Kohler-Spath H, et al: Prospective evaluation of a follow-up schedule in cutaneous melanoma patients: Recommendations for an effective follow-up strategy. J Clin Oncol 21:520-529, 2003 26. Garbe C, Buttner P, Bertz J, et al: Primary cutaneous melanoma: Prognostic classification of anatomic location. Cancer 75:2492-2498, 1995[CrossRef][Medline] 27. McKinnon JG, Yu XQ, McCarthy WH, et al: Prognosis for patients with thin cutaneous melanoma: Long-term survival data from New South Wales Central Cancer Registry and the Sydney Melanoma Unit. Cancer 98:1223-1231, 2003[CrossRef][Medline]
28. Balch CM, Soong SJ, Gershenwald JE, et al: Prognostic factors analysis of 17,600 melanoma patients: Validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 19:3622-3634, 2001 29. Cascinelli N, Marubini E, Morabito A, et al: Prognostic factors for stage I melanoma of the skin: A review. Stat Med 4:265-278, 1985[Medline] 30. Gamel JW, George SL, Edwards MJ, et al: The long-term clinical course of patients with cutaneous melanoma. Cancer 95:1286-1293, 2002[CrossRef][Medline]
31. Lipsker DM, Hedelin G, Heid E, et al: Striking increase of thin melanomas contrasts with stable incidence of thick melanomas. Arch Dermatol 135:1451-1456, 1999
32. Mackie RM, Hole D, Hunter JA, et al: Cutaneous malignant melanoma in Scotland: Incidence, survival, and mortality, 1979-94—The Scottish Melanoma Group. BMJ 315:1117-1121, 1997 33. Massi D, Franchi A, Borgognoni L, et al: Thin cutaneous malignant melanomas (< or =1.5 mm): Identification of risk factors indicative of progression. Cancer 85:1067-1076, 1999[CrossRef][Medline] Submitted March 9, 2004; accepted June 7, 2004.
Related Article
Related Editorial
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
