Prognostic Significance of Tumor Proliferative Fraction and DNA Content in Stage I Non-Small Cell Lung Cancer1 , 2

ANDREW E. FILDERMAN, GERARD A. SILVESTRI, CONSTANTINE GATSONIS, DANIEL J. LUTHRINGER, JULIE HONIG, and STUART D. FLYNN

Introduction

The prognosis for patients with localized (tumor, node, and metastasis: T 1 NoMo and T2NoMo) non-small cell carcinoma (NSCC) is generally favorable. However, despite having a localized, resectable tumor, up to 50070 of patients may relapse within 5 yr (1). It is apparent that factors in addition to the initial anatomic stage may affect tumor response to therapy and risk of relapse. It would be of great value to be able to identify patients with a favorable TNM stage who are at risk for tumor relapse. This would improve prognostic prediction, identify patients for close clinical follow-up, and allow for trials examining the efficacy of adjuvant therapies in a subset of patients with a less favorable prognosis. The technique of flow cytometry has enabled the study of tumor DNA content (ploidy) and proliferative fraction (percentage S phase) both in fresh tumor specimens as well as archival specimens (2). Evaluations of tumor DNA content and proliferative index have been found to be significant prognostic factors in a variety of tumors (3-5). In general, tumors with abnormal DNA content (aneuploidy) or a high proliferative fraction appear to havea more aggressivebiologic behavior and poorer outcome. The clinical utility of these measurements has proven to be useful in localized breast carcinoma in identifying patients at high risk for relapse and thus helping to stratify patients for adjuvant therapy (6, 7). A number of studies have examined the relationship between tumor DNA content and survival in all stages and cell types of NSCC (8-16). The majority of studies have shown a poorer prognosis for those patients with aneuploid tumors (9-12, 15, 16). However, there have been conflicting results in studies of localized, surgically resectable tumors (13, 14). Fewer studies have examined the value of the proliferative fraction as a prognostic variable in lung cancer (8, 9, 14).

SUMMARY Analyses of tumor DNA content and proliferative fraction by flow cytometry have been useful as prognostic determinants In a variety of solid tumors; The significance of this analysis in Stage I (T 1N oMo and T 2NoMo) non-small cell lung carcinoma (NSCC) Is unestabllshed. We determined DNA content (ploidy) and proliferative fraction (percentage S phase) on 44 surgically resected Stage I NSCC specimens obtained between 1m and 1982. All cases had a minimum follow-up of 5 yr. Of the 44 cases, 27 were adenocarcinomas, 15 squamous cell carcinomas, and 2 large cell carcinomas. Of these, 32 (73%) had T 1NoMo lesions and 12 (27%) had T 2NoMo lesions. Overall5-yr survival was 70%. All patients surviving 5 yr were free of detectable tumor. Patients with T1N oMo lesions had an 81% 5-yr survival, but those with T2NOMo lesions had a 42% 5-yr survival (p = 0.009). Analysis of tumor DNA content revealed 35 diploid tumors (79%) and 9 aneuploid tumors (21%). The 5-yr survival for diploid tumors was n% compared with a 44% 5-yr survival In aneuploid lesions (p = 0.048). The median proliferative fraction was 6%. All patients with a percentage S phase less than 6% survived 5 yr, and 41% (9 of 22) of those greater than 6% survived 5 yr (p < 0.001).When 8% S phase was used as a eutott, 93% (28 of 30) below the cutoff survived 5 yr but only 21% (3 of 14) above the cutoff survived 5 yr (p < 0.001). Within the T 1NoMo SUbgroup (n = 32), all patients with a percentage S phase less than 6% survived past 5 yr but only 54% (7 of 13) of those greater than 6% survived 5 yr (p = 0.002). Thus, measurement of tumor proliferative fraction and, to a lesser extent DNA content, In Stage I NSCC may provide significant prognostic Information apart from TNM stage and help identify a subset of patients at high risk for tumor relapse. AM REV RESPIR DIS 1992; 146:707-710

In other tumors this variable has proven to be very accurate in predicting diseasefree and overall survival (7, 17). In this study weexamined the significance of tumor DNA content and proliferative fraction in patients with surgically resected Stage I NSCC. All specimens studied were archival, paraffin-embedded tissue, and all patients had a 5-yr clinical follow-up. Methods Patients Patients for this study were identified from the Thmor Registry at Yale-New Haven Hospital as having had surgically resected Stage I NSCC between the years 1977through 1982. During this 6-yr period 160 consecutive patients underwent surgical resection and had pathologically confirmed Stage I disease. Of these 160 patients, adequate surgical pathologic material for cytometric analysis was available in only 56 patients. The majority of patients were excluded because of the lack of tissue blocks for analysis, and 3 had less than 10,000cells on the histogram. Survivors in the study weredefined as patients who lived greater than 5 yr following surgery; nonsur-

vivors died of recurrent tumor within 5 yr of surgery. Of the 56 patients studied, 12 were excluded for the following reasons: five had carcinoid tumors, one died of a non-tumor-related cause within 5 yr, one developed lung metastases thought to be from a primary colon carcinoma, and fivehad uninterpretable DNA histograms. Thus, a total of 44 patients comprised the study populations.

Pathologic Examination Gross tumor characteristics (sizeand location) (Received in original form May 29, 1991 and in revised form April 13, 1992) 1 From the Pulmonary and Critical Care Section, Department of Internal Medicine, and the Department of Pathology, Yale University School of Medicine, the Department of Internal Medicine, Hospital of St. Raphael's, New Haven, Connecticut, and the Department of Health Care Policy, Harvard University Medical School, Boston, Massachusetts. 2 Correspondence and requests for reprints should be addressed to Dr. Andrew E. Filderman, Department of Pulmonary and Critical Care Medicine, St. Joseph Hospital, 2900 N. Lake Shore Drive, Chicago, IL 60657.

707

708

FILDERMAN, SILVESTRI, GATSONIS, WTHRINGER, HONIG, AND FLYNN

were determined by preoperative radiologic studies, surgical report, and surgical pathology description. All tumors examined were archival, paraffin-embedded specimens. Sections were cut from each tumor and reexamined to confirm the histopathologic diagnosis. Histologic type was assigned according to World Health Organization criteria (18). Postsurgical pathologic stage was determined according to the revised international TNM staging system (19).

fluorescent emission wasmeasured to 610nm. Single-parameter, 256-channel DNA histograms werederived from an analysisof a minimum of 10,000 cells (average 15,000 cells). Histogram analysis was performed using the software Verity Modfit'" (Verity Software House, Topsham, ME). Cell cycle distribution of the presynthetic growth phase (00 / 01 ) , synthetic phase (S), and postsynthetic and mitotic growth phases (02/M) was evaluated. A rectangle model, taking the best fit for the data, was used to determine S phase (20). The DNA content was measured and interpreted as diploid if only one 0 0 / 01 peak was present. An aneuploid peak was defined as a second peak separate from the diploid peak, containing at least 10% of the cells counted, and having a corresponding 02/M peak. The DNA index (DI) was calculated as the ratio of the peak channel number of the aneuploid population, if present, divided by the peak channel number of the diploid 0 0 / 01 peak. By definition, the DI of diploid tumors is 1. The coefficient of variation (CV) for the diploid 0 0 / 01 peak was also assessed.

(27070). All patients had complete resection of the primary tumor, with 2 requiring pneumonectomy, 38 lobectomy, and 4 wedge resection. Overall survival at 5 yr for all patients was 70070. All patients surviving for 5 yr were free of detectable tumor. Patients with TtNoMo lesions had an 81070 5-yr survival, but those with T2NoMolesions had a 42070 5-yr survival (P = 0.009) (figure 1). Analysis of tumor DNA content reFlow Cytometry vealed 35 diploid tumors (79070) and 9 The nuclear DNA content was determined aneuploid tumors (21070). Distribution of from formalin-fixed, paraffin-embedded arploidy within the T tNoMolesions showed chival tissue by flow cytometry using a modifi(810/0) to be diploid and 6 (19070) to 26 cation of the method described by Hedley and be aneuploid. Among the T2NoMo lesions colleagues (2). Alternating 50 and 6 J.1m thick there were 9 (750/0) diploid tumors and sections werecut from the paraffin blocks that 3 (25070) aneuploid tumors. There was no were selected to best represent the lung carcinoma in each case. The 6 J.1m thick sections significant correlation between tumor were stained with hematoxylin and eosin for DNA content and age, sex, tumor histolhistologic evaluation. The 50 J.1m thick secogy, or tumor stage. tions weredeparaffinized in Americlears (SciThe 5-yr survival for the 35 diploid leentific Products, McOaw Park, IL) and resions was 770/0 (27 of 35) compared with hydrated through a series of 100, 95, 70, and a 44070 (4 of 9) 5-yr survival in patients 50070 ethanol solutions to double-distilled waData Analysis with aneuploid tumors (two-sided exact ter. Following two washes in double-distilled Survival curves were estimated using the log-rank p value = 0.048; asymptotic p water, the specimen was washed with 20 ml glycineand 0.9% NaCI solution (pH 1.5).The Kaplan-Meier estimator (21). The effect of value = 0.041) (figure 2). The mean tumor size (stage), ploidy, and percentage S coefficient of variation for the Gs/G, tissue was then minced into 1 mm cubes and incubated in a 37° C water bath for 30 min phase on survival were assessed in two ways: peak was 4.70/0 (range 1.8to 7.8070). With(with intermittent vortex mixing) with the gly- (1) using Cox regression models (21)in which in the aneuploid group, the median DI appropriately defined covariates were includcine solution, which also contained 1 ml of was 1.35 (range 1.2 to 2). 1% pepsin (Sigma Chemical Company, St. ed, and (2) using an exact logarithmic rank The median percentage S phase was test procedure (22). The latter is appropriate Louis, MO). Immediately following this in6070. There was no significant correlation for the comparison of survival curves when cubation, the specimen was centrifuged for between proliferative fraction and age, the sample sizes are small. 5 min (1,000 x g), the supernatant discardsex, or tumor histology. However, there ed, and the pellet washed with RPMI 1640 was an indication that higher percentage media (OIBCO Laboratories, Orand Island, Results S phase values were associated with a NY) and filtered through alSO J.1m metal mesh. The cells were then incubated at 4 C A total of 44 patients were evaluable for higher tumor stage. Among T tNoMo leovernight with 1 ml of a solution containing the study. The patient population con- sions, 19 had a percentage S phase less 0.05% propidium iodide (Sigma), 0.1% sodi- sisted of 19 females and 25 males with than 6070 and 13 greater than 6070; in the um citrate, 180 units/ml of RNase A (Wor- an age range from 39 to 82 yr (mean ± T2NoMo group, 3 had a percentage S thington Biochemical, Freehold, NJ), 0.2% standard deviation, 62 ± 11 yr). Histo- phase less than 6070 and 9 were greater 'Iritorr" X-100 (Sigma), and 0.5% Nonider" logic cell types included 27 adenocarcithan 6070. The p value of Fisher's exact P40 (Sigma) in 5 ml Tris buffer. nomas, 15 squamous cell carcinomas, test for these data is 0.044, and a logistic The nuclei were analyzed on a Coulter and 2 large cell carcinomas. Distribution regression analysis in which percentage EPICS Profiles flow cytometer (Coulter Electronics, Hialeah, FL). An argon laser was ad- of Stage I lesions included 32 with S phase was entered as a continuous exTtNoM o (73070) and 12 with T 2NoMo planatory variable and T2 stage was the justed to produce excitation at 488 nm, and response variable resulted in a coefficient of 0.09 for the percentage S phase (stanSurvival by Size group dard error of the mean, SEM = 0.06, C! corresponding to an odds ratio of 1.4 for L.~:._.--;, an increase of 4070 in the S phase). 1."1 ._.,._.-:.....All 22 patients with a percentage S ci .......... '-'-'-'-'L~ phase below 6070 survived past 5 yr, but Fig. 1. Overall survival and survival ac~ only 41070 (9 of 22) did so among patients :0 lO cording to TNM group in Stage I ci .0 '" with a percentage S phase above the me[1. _.. :..._...:.. . e non-small cell carcinoma of the lung. a. dian (two-sided exact p value < 0.001). Patients with 1;NoMo lesions had a sig~ "1" ci nificantly better 5-yr survival than those When 8070 S phase was used as a cutoff, en with T2N oMo disease (p = 0.009). 93070 (28 of 30) of patients below the _._.- overall (0=44) cutoff survived 5 yr, while only 21070 (3 ci ........... T2NOMO (0=12) of 14)above the cutoff survived 5 yr (two- - T1NOMO (0=32) sided exact p value < 0.001) (figure 3). ci Using 10070 S phase as a cutoff, 88070 (30 2 4 o 3 5 of 34) of patients below the cutoff surYears vived 5 yr, but only 10070 (1of 10)ofthose 0

ClO

.~

::l

C\l

0

709

PROLIFERATIVE FRACTION AND PLOIDY IN STAGE I WNG CANCER

Survival by Ploidy group

co ci

Fig. 2. Overall survival and survival according to DNA content (ploidy) in Stage I non-small cell carcinoma of the lung. Patients with diploid lesions had a significantly better 5-yr survival than those with aneuploid lesions (p = 0.048). N

_._ .• Overall (n-44)

ci

........... Aneuploid (n=9) - - Diploid (n=35) o ciL...-,-_ _----r

. . -_ _.....--_ _--.-_ _~

o

2

4

3

5

Years

above the cutoff did so (two-sided exact p value < 0.001). Finally, when percentage S-phase was entered as a continuous covariate in a Cox regression model, the estimated coefficient was 0.25 (SEM = 0.05, twosided p value < 0.001, corresponding to a hazard ratio of 2.7 for an increase of 4070 in the S phase). All aspects of this analysis point to a strong relationship between survival and percentage S phase. The limited sample size and the heavy amount of censoring preclude a full multivariate analysis of survival as a function of several predictors. However, we were able to assess the predictive value of the percentage S phase within the larger of the two stage groups (TtNoMo, n = 32). All patients in this group with a percentage S phase below 6070 survived past 5 yr, but only 54% (7 of 13)did so among patients with a percentage S phase greater than 6070 (two-sided exact p value = 0.002). Discussion

The anatomically based TNM staging system traditionally has been used to predict prognosis in patients with lung can-

cer (19). The advantages of this staging classification are well-defined categories and the relative ease of TNM classification. However, this system is limited by reliance on anatomic tumor site without taking into account tumor-specific biologic factors that may affect tumor growth and response to therapy (23). The influence of these additional biologic variables may explain the heterogeneous prognosis in patients with a seemingly favorable TNM classification (1). Because of this heterogeneity within the TNM classification, other factors have been sought that may have prognostic significance. In a retrospective study, Lipford and coworkers (24) analyzed a number of clinical, microscopic, and pathologic variables and found that large cell undifferentiated histology, thoracic lymph node involvement, larger primary tumor size, presence of giant cells in tumors, and lack of a tumor plasma cell infiltrate had a negative impact on patient survival. In a study of patients with small (~2 em) peripheral lung cancer lesions, negative prognostic variables included adenocarcinoma celltype, pleural involvement, nodal involvement, higher Survival by S-phase group F..., :·..1· _·-;,

Fig. 3. Overall survival and survival according to proliferative fraction (S phase) in Stage I non-small cell carcinoma of the lung. Patients with a percentage S phase less than 8% had a significantly better survival than those with a proliferative fraction greater than 8% (p < 0.001).

~ 0.. ~

..,. ci

(/) N

ci

~.-

'\LL_:::::::_:~::

~

~ ~

~

'-"._.-.

: : :

co ci

_._ .• overall (n-44)

........... S-phase >- 8% (n-14) o ci

- - S-phase < 8% (n-30)

o

3

2 Years

4

5

pathologic stage, male gender, and incomplete resections (25). Recently, studies in a variety of tumors have suggested that analysis of cellular factors, including tumor cell DNA content and proliferative growth fraction, may correlate with cell growth and risk of tumor relapse (3-5). This has been well studied in breast carcinoma in which DNA content and proliferative fraction are significant independent prognostic variables (6, 7, 17) and may soon be included in the initial staging process of this tumor (26). The value of tumor ploidy and proliferativefraction analysis in bronchogenic carcinoma is less well established. Several studies have reported DNA content to be a significant prognostic variable in a variety of NSCC cell types and stages. Zimmerman's (10) and Asamura's (12) groups have reported that tumor DNA content accurately predicts prognosis in Stage I NSCC. In contrast, other studies either did not find tumor DNA content to be a significant independent prognostic variable (13, 14)or found tumor ploidy to be valuable only for the squamous cell carcinoma subtype (15). In our study we found that aneuploid tumors occurred in 21% of cases. There was no correlation between tumor size and DNA content. The frequency of aneuploid tumors was less than seen in other studies, which have reported aneuploidy in 45 to 96070 of cases, and may reflect that localized tumors are less aggressive biologically than higher stage tumors. Alternatively, the low rate of aneuploid cases may be a particularity of our study sample, which was a subset from a total of 160patients. It is also possible that the frequency of aneuploidy may be underestimated because of possible heterogeneity in tumor DNA content (27). Sections for ploidy analysis were taken from representative areas of the tumor, however, and distinction between diploid and aneuploid histograms was clear. Overall, we found that tumor ploidy correlated significantly with survival. Analysis of this relationship was difficult, though, because of the small number of aneuploid cases examined. The importance of the apparent correlation between ploidy and survival should also be tempered by the fact that cases with uninterpretable ploidy results were not included in the analysis. In our final sample, this group accounted for 9% (5 of 56) of all cases. Only a limited amount of information exists regarding the significance of the proliferative fraction in lung cancer (8,

710

9, 14). In some other tumors the percentage S phase has been found to be a significant prognostic variable apart from tumor ploidy (5, 7, 17). The value of this variable may stem from the direct correlation of the percentage S phase with tumor cell growth characteristics (28, 29). The few studies of lung cancer in which proliferative fraction was assessed have found conflicting results (8, 9, 14). In our study the percentage S phase correlated strongly (p < 0.(01) with risk of tumor relapse. The median percentage S phase was 6070; all patients with a proliferative fraction less than this value survived greater than 5 yr, but 41% (9 of 22) of patients with a proliferative fraction greater than 6% and only 21% (3 of 14) of patients with a percentage S phase greater than 8070 survived 5 yr (figure 3). When a 10070 cutoff is used for the proliferative fraction, only 10070 (l of 10)of patients with a percentage S phase greater than 10% survived longer than 5 yr. Because of the small number of cases, we were unable to perform a full multivariate analysis to determine whether these factors may be regarded as independent variables when determining prognosis. However, in the one instance when such an analysis was possible, that is, stratifying by stage, we found a strong suggestion that percentage S phase is an independent prognostic factor distinct from tumor stage. In the TtNoMo subgroup, all patients with a low proliferative fraction (less than 6070) survived past 5 yr, but only 54070 of those with a proliferative fraction greater than 6070 were long-term survivors. We fully recognize, however, that this is a small study requiring replication with a larger amount of data to allow a full multivariate analysis of the model. In addition, the data supporting DNA content as a significant prognostic variable are less clear-cut; a greater number of cases must be analyzed to establish this factor as an independent prognostic variable. A critical cutoff value for the proliferative fraction in NSCC is unknown. Using a median value of 6070 in this study allowed identification of patients who had an excellent prognosis (percentage S phase less than 6070) but also included patients with either poor or good prognoses in the group over 6070. In contrast, a cutoff of 8070 included some patients with a poor prognosis less than this value but identified a group (percentage S . phase greater than 8070) who had a particularly poor prognosis. Because of the small number of patients in this study,

FILDERMAN, SILVESTRI, GATSONIS, WTHRINGER, HONIG, AND FLYNN

we did not attempt to determine an optimal cutoff value for the proliferative fraction. Greater numbers of patients must be analyzed to establish the desired cutoff for the relationship between proliferative fraction and survival, and individual institutions may need to establish their own values for the optimal proliferative fraction (7). The strong relationship between proliferative fraction and survival suggests that this index of tumor growth may be a valuable prognostic variable within Stage I NSCC. Tumor DNA content also appears to be important, albeit less significant. There appear to be several reasons for including a "biologic" staging of the patient at the time of diagnosis. First, a better understanding of prognosis would be obtained. Second, tumors with a high proliferative fraction and/or aneuploidy may benefit from more intensive clinical monitoring. Finally, an improved stratification of patients with Stage I NSCC would identify a subgroup of patients with higher rates of tumor relapse who might benefit from additional therapy. To date, trials of adjuvant therapy in lung cancer have met with limited success (30, 31). References 1. Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988; 96:440-7. 2. Hedley DW, Friedlander ML, Taylor IW, Rugg CA, Musgrove EA. Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytometry. J Histochem Cytochem 1983; 31:1333-5. 3. Barlogie B, Raber MN, Schumann J, et al. Flow cytometry in clinical cancer research. Cancer Res 1983; 43:3982-97. 4. Friedlander ML, Hedley DW, Taylor IW. Clinical and biological significance of aneuploidy in human tumours. J Clin Pathol 1984; 37:961-74. 5. Merkel DE, Dressler LO, McGuire WL. Flow cytometry, cellular DNA content, and prognosis in human malignancy. J Clin Pathol 1987; 5: 1690-703. 6. Fallenius AG, Franzen SA, Auer ou Predictive value of nuclear DNA content in breast cancer in relation to clinical and morphologic factors: a retrospective study of 227 consecutive cases. Cancer 1988; 62:521-30. 7. Sigurdsson H, Baldetorp B, Borg A, et al. Indicators of prognosis in node-negative breast cancer. N Engl J Med 1990; 322:1045-53. 8. Bunn PA, Carney DN, Gazdar AF, Whang-Peng J, Matthews MJ. Diagnostic and biologic implications of flow cytometric DNA content analysis in lung cancer. Cancer Res 1983; 43:5026-32. 9. Volm M, Drings P, Mattern J, Sonka J, VogtMoykopf I, Wayss K. Prognostic significance of DNA patterns and resistance-predictive tests in non-small cell lung carcinoma. Cancer 1985; 56: 1396-403. 10. Zimmerman PV, Bint MH, Hawson GAT, Parsons PG. Ploidy as a prognostic determinant in SUf-

gically treated lung cancer. Lancet 1987; 2:530-3. 11. Tirindelli-Danesi D, Teodori L, Mauro F, et al. Prognostic significance of flow cytometry in lung cancer. Cancer 1987; 60:844-51. 12. Asamura H, Nakajima T, Mukai K, Shimosato Y. Nuclear DNA content by cytofluorometry of stage I adenocarcinoma of the lung in relation to postoperative recurrence. Chest 1989; 96:312-8. 13. Van Bodegom PC, Baak JP, Stroet-Van Galen C, et al. The percentage of aneuploid cells is significantly correlated with survival in accurately staged patients with stage I resected squamous cell lung cancer and long-term follow up. Cancer 1989; 63:143-7. 14. Cibas ES, Melamed MR, Zaman MB, Kimmel M. The effect of tumor size and tumor cell DNA content on the survival of patients with stage I adenocarcinoma of the lung. Cancer 1989; 63:1552-6. 15. SahinAA, Ro JY, El-Naggar AK, etal. Flow cytometric analysis of the DNA content of nonsmall cell lung cancer. Cancer 1990; 65:530-7. 16. Isobe H, Miyamoto H, Shimizu T, et al: Prognostic and therapeutic significance of the flow cytometric nuclear DNA content in non-small cell lung cancer. Cancer 1990; 65:1391-5. 17. Kallioniemi O-P, Blanco G, Alavaikko M, et al. Improving the prognostic value of DNA flow cytometry in breast cancer by combining DNA index and S-phase fraction. Cancer 1988;62:2183-90. 18. World Health Organization. The World Health Organization histologic typing of lung tumors, ed. 2. Am J Clin Pathol 1982; 77:123-36.' 19. Mountain CF. A new international staging system for lung cancer. Chest(Suppl) 1986; 89:225-33. 20. Lacombe F, Belloc F, Bernard P, Boisseau MR. Evaluation of four methods of DNA distribution data analysis based on bromodeoxyuridine/DNA bivariate data. Cytometry 1988; 9:245-53. 21. Cox DR, Oakes D. Analysis of survival data. New York: Chapman and Hall, 1984. 22. Statexact: Statistical software for exact nonparametric inference. Cambridge, MA: CYTEL Software Corporation, 1989. 23. Hajj C, Akoum R, Bradley E, Paquin F, Ayoub J. DNA alterations at proto-oncogene loci and their clinical significance in operable non-small cell lung cancer. Cancer 1990; 66:733-9. 24. Lipford EH, Eggleston JC, Lillemoe KD, Sears DL, Moore GW, Baker RR. Prognostic factors in surgically resected limited-stage, nonsmall cell carcinoma of the lung. Am J Surg Pathol 1984; 8:357-65. 25. Sagawa M, Saito Y, Takahashi S, et al. Clinical and prognostic assessment of patients with resected small peripheral lung cancer lesions. Cancer 1990; 66:2653-7. 26. Adjuvant chemotherapy of early breast cancer. Med Lett 1990; 32:49-50. 27. Carey FA, Lamb D, Bird CC. Intratumoral heterogeneity of DNA content in lung cancer. Cancer 1990; 65:2266-9. 28. Braylan RC, Diamond LW,Powell ML, HartyGolder B. Percentage of cells in the S phase of the cell cycle in human lymphoma determined by flow cytometry. Cytometry 1980; 1:171-4. 29. McDivitt RW, Stone KR, Craig RB, Meyer JS. A comparison of human breast cancer cell kinetics measured by flow cytometry and thymidine labeling. Lab Invest 1985; 52:287-91. 30. Legha SS, Muggia FM, Carter SK. Adjuvant therapy in lung cancer. Cancer 1977; 39:1415-24. 31. Holmes EC, Hill LD, Gail M. A randomized comparison of the effects of adjuvant therapy on resected stages II and III non-small cell carcinoma of the lung. Ann Surg 1985; 202:335-41.

Prognostic significance of tumor proliferative fraction and DNA content in stage I non-small cell lung cancer.

Analyses of tumor DNA content and proliferative fraction by flow cytometry have been useful as prognostic determinants in a variety of solid tumors. T...
541KB Sizes 0 Downloads 0 Views