prognostic value of n-myc immunoreactivity in medullary thyroid carcinoma

134

Prognostic Value of N-myc lmmunoreactivity in Medullary Thyroid Carcinoma Massirno Roncalli, M.D.,* Giuseppe Viafe , M.D.,* Lars Grimetius, M.D.,$ Henry Johansson, M.D.,§ Erik Wilander, M.D.,S Rosa Maria Alfano, D.Sc.,* David Springall, Ph.D., 1 1 Pier Maria Battezzati, M.D.,t Julia M . Polak, D.Sc.,II and Guido Coggi, M.D.*

Background. The analysis of N-myc expression in some neuroendocrine tumors has been reported to pro- vide prognostic information. To the authors' knowledge, no attempts have been made thus far to correlate N-myc expression with the clinical outcome of medullary thyroid carcinoma (MTC).

Methods. N-myc gene product immunoreactivity was evaluated in 34 patients with MTC with long term follow- up, using the OA-11-803 polyclonal antiserum. The results were related to patient age and sex, sporadic or familial disease, tumor size, stage, growth rate (as deter- mined by proliferating cell nuclear antigen [PCNA] immunostaining), and to clinical outcome.

Results. Patients harboring tumors with greater than 10% neoplastic cells immunoreactive to the N-myc antiserum (58% of the cases investigated) had significantly greater tumor size (P = 0,031) than patients with fewer or no N-myc immunoreactive cells. Deregulated expression of N-myc protein in tumor cells was not due to gene amplification, as demonstrated by multiplex polymerase chain reaction (PCR). In univariate analysis, patients with more than 10% immunoreactive neoplastic cells showed a significantly shorter disease free survival than did the remaining patients (P = 0.002). Among the other clinicopathologic parameters evaluated, male sex (P = 0.039) and sporadic disease (P = 0.035) also were associated with shorter disease free survival. In multivariate

From the I1 Department of *Pathology and the Department of tInternal Medicine, University of Milan School of Medicine, Milan, Italy; the Department of $Pathology and the Department of §Surgery, University Hospital of Uppsala, Uppsala, Sweden; the IlDepartment of Histochemistry, RPMS of the University of London, London, Great Britain.

The authors thank Ulla Bergholm, M.D., for collecting data on patient follow-up. The technical work of Mrs. Anna Comi is greatly acknowledged. Daniela Graziani, D.Sc., provided helpful assistance with the PCR experiments.

Address for reprints: Massimo Roncalli, M.D., I1 Department of Pathology, University of Milan School of Medicine, Via A. di Rudini 8, 20142 Milano, Italia.

Received November 18, 1993; revision received January 20, 1994; accepted March 8, 1994.

analysis, N-myc immunoreactivity (P = 0.039) and male sex (p = 0.050) retained a significant correlation with poor prognosis.

Conclusions. Our results suggest that immunoreactivity to the N-myc antiserum, but not tumor growth fraction as evaluated by PCNA immunostaining, is a novel and useful adjunct to predict clinical behavior of MTC. Cancer 1994; 74:134-41.

Key words: medullary thyroid carcinoma, N-myc, proliferating cell nuclear antigen, prognosis, immunocytochemistry, multiplex polymerase chain reaction.

The assessment of protooncogene (c-onc) and tumor suppressor gene product immunoreactivity in some common human malignancies has been shown to predict the clinical outcome of these malignancies. Indeed, c-erbB-2 gene product in breast and ovarian carcinomas, p21 ras product in lung adenocarcinomas, and p53 gene product in bladder and breast carcinomas are only a few examples of gene products abnormally expressed in more aggressive tumor The prognostic value of a deregulated c-onc expression in human neuroendocrine tumors has been less extensively investigated. A notable exception is represented by neuroblastomas, in which N-myc gene amplification has been correlated with a poor prognosis and requires a specific therapeutic a p p r ~ a c h . ~ - ~ Several studies have indicated that in some human tumors, and in embryonal development, N-myc gene overexpression possibly reflects the emergence of an immature cell population of puta- tive neurocrest

Medullary thyroid carcinoma (MTC) is a tumor of neurocrest derivatives, characterized by an unpredictable clinical course, irrespective of some conventional pathologic prognostic parameters such as tumor size, histology, and peptide immunocytochemical features."

N-myc in Medullary Thyroid Carcinoma/Roncalli e t al. 135

Though N-myc product immunoreactivity and mRNA expression have been recently and consistently demonstrated in different series of MTC,"-" no attempts have been made thus far to correlate N-myc expression with the clinical outcome of the disease.

In this study, a large series of MTC patients who had a long term follow-up was investigated with the aim of assessing the prognostic value of immunoreactivity to N-myc antiserum and its relationship with other clinicopathologic parameters, such as patient age and sex, sporadic or familial disease, tumor size and stage, and clinical outcome. The multiplex polymerase chain reaction (PCR) techniq~e '~ was used to assess whether N-myc gene was also amplified. Additionally, the immunocytochemical analysis of proliferating cell nuclear antigen (PCNA), a protein expressed through- out the cell cycle,I6 was performed to determine whether PCNA correlates with a deregulated N-myc immunoreactivity and whether PCNA, per se, may be of prognostic relevance.

Materials and Methods

The records of thirty-four patients with MTC who had long term follow-up (mean, 10.3 years; range, 1-26 years) were studied from the files of the University Hos- pital of Uppsala, Sweden. One paraffin block per patient was selected, based on good morphologic preser- vation. There were 16 men and 18 women (mean ages, 56 and 51 years, respectively; range, 18-82 years). Clin- icopathologic information included clinical type (sporadic or familial disease), tumor size, and stage (tumor appearance as confined to the thyroid gland or lymph node metastasis), and patient outcome.

Immunocytochemistry

Formalin fixed, paraffin embedded tissue sections were immunostained for N-myc gene product using the polyclonal antibody OA-11-803 (Cambridge Research Bio- chemicals, Northwich, UK) and the streptavidin-biotin system. The antibody specificity was characterized previously either in immunocytochemistry'4 or in Western blot analy~is, '~ where it showed binding to a 63-66 kD doublet, corresponding to the normal positions of the N-myc protein(s). According to N-myc immunoreactivity, tumors were subdivided in four groups: negative tumors; tumors with less than or equal to 10% immunoreactive neoplastic cells; tumors with 11-50% immunoreactive cells; and tumors with greater than 50% immunoreactive cells.

Immunostaining for PCNA was performed using the PClO monoclonal antibody (Dakopatts, Glostrup, Denmark) and the avidin-biotin system, as previously reported.'* Immunostaining results were scored by counting at least 2000 neoplastic cells in 10 randomly

selected, high-power fields (X400). l8 The percentage of PCNA-immunoreactive cells was recorded as PCNA labeling index (LI).

Negative control experiments were performed by substituting the primary antibody with normal sheep or mouse serum. For N-myc antibody, control experiments with the antiserum preadsorbed with an excess of the purified antigen were also run.14 The results of these reactions were always negative. Known N-myc immunoreactive and nonimmunoreactive neuroblastomas were also stained with the MTC samples.

The stained slides were independently evaluated by two of the authors (M.R. and G.V.); when different evaluations were obtained, a consensus interpretation was reached after reexamination of the slides with a double-headed microscope.

Multiplex Polymerase Chain Reaction

Three consecutive 10-p-thick paraffin sections (34 MTCs and the positive control patient with neuroblastoma) adjacent to those immunostained for N-myc were dewaxed and lysed overnight at 55OC in 50 mM Tris-HC1 (pH 8), 5 mM EDTA, 0.5% Tween 20 and 400 ng/ml Proteinase K (Sigma, St. Louis, MO).

The N-myc primers were: (1) TCACTGTGGA- GAAGCGGCGTT3'; and (2) 5'GTGCATCCTCACTCT- CCACGT3'. These primers amplify a 195-base pair sequence of the third exon corresponding to position 6846-7040 of the published sequence." The interferon gamma (IFN-7) gene was used as a single-copy reference gene. The IFN-./ primers were: (1) 5TCTTTTCTT- TCCCGATAGGT3'; and (2) 5'CTGGGATGCTCTTC- GACCTC3'. These primers amplify a 150-base pair sequence corresponding to position 4582-4731 of the published sequence.20 Oligonucleotide primers were synthesized on an Applied Biosystem 392A DNA/RNA synthesizer (Foster City, CA), using standard beta-cya- noethyl-phosphoramidite chemistry. PCR was performed in a final volume of 50 p1, containing 10 mM Tris HCl pH 8.3, 50 mM KC1, 4 mM MgC12, 200 mM each of deoxynucleotide triphosphate, 0.1 mM each of N-myc primers, and 0.5 mM each of IFN--, primers. Five microliters of the digestion supernatant, and 0.5 U of Taq polymerase (Perkin Elmer, ) were added to the reaction mixture, which was finally overlaid with 50 p1 of mineral oil. Amplification was performed in a Perkin Elmer Cetus Thermal Cycler (Branchburg, NJ) under the following conditions: denaturation at 94OC for 1 minute; annealing at 53OC for 1 minute; extension at 72OC for 1 minute for 39 cycles, followed by a final extension of 10 minutes at 72OC. Ten microliters of the amplified product were electrophoresed with 3% agarose-gel (Nusieve 3: 1 agarose; FMC Bioproducts, Rock- land, ME). Gels were stained with ethidium bromide at a final concentration of 0.5 mg/ml and were then pho-

136 CANCER July 1,1994, Volume 74, No. 1

Table 1. Patient Characteristics

Follow-up

Patient Age Diameter Clinical Duration no. (yr) Sex N-myc (I) PCNA (cm) Stage type (yr) Status

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

44 57 47 47 67 30 50 64 66 71 61 41 73 58 18 30 63 67 52 46 26 48 82 77 42 65 20 40 45 63 72 56 80

M M M F F F M F M M F F F M F F F M M M F M M F F F F M F M F M F

5 10 0

> 50 I 10

11-50 11-50 5 10 > 50 > 50 11-50 11-50 > 50

11-50 0

11-50 0

I 10 11-50 5 10

11-50 11-30 I 10 11-50 > 50 I 1 0

11-50 5 10 > 50 I 1 0

11-50 0

11-50 5 10

4 10 40

< I 2 5

40 25 15 50 15 50 3 0

25 15 40 40 35 30

< I 80

< l 35 50

< 1 15 65 70 10 1

10 <1

3 NA

> 3 > 3 > 3 > 3

1 2

> 3 2.5

> 3 > 3 > 3

1 > 3

2 1 3 > 3

2 3 1 1

> 3 > 3

3 > 3

3 NA

3 3

> 3 3 3

Met*

Met

Met*

Met*

Met

Sporadic Familial Sporadic Familial Sporadic Familial Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Familial Sporadic Sporadic Sporadic Sporadic Familial Familial Sporadc Familial Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic Sporadic

22 17 1

26 18 1 1

12 1

12 2

18 2 3 8

24 19 13

7 5

14 24 10 2

19 14 5 7 9 1

16 8 9

DNRD ANED DOD ANED ANED DOD AWD DNRD DOD DNRD DOD ANED DOD DOD DOD ANED DNRD DOD DNRD DOD AWD AWD AWD DOD AWD AWD DOD AWD ANED DOD ANED DOD ANED

1 DOD 34 58 M 11-50 50 > 3 Met* Sporadic - NA: not available; Met: presence of lymph node metastases on diagnosis; ANED: alive, no evidence of disease; AWD: alive with disease; DOD: dead of disease; DNRD: dead of nonrelated disease. * Only lymph node metastases available for study.

tographed. N-myc amplification was identified by visual comparison of the relative intensities of the N-myc- and IFN-y-amplified bands. Where the IFN-y band was of equal or greater intensity than the N-myc band, the sample was scored as single copy. Where the N-myc band was of unequivocally greater intensity, the result was scored as amplified. A DNA sample from a patient with a neuroblastoma with an amplified N-myc gene was included as a positive control, and it consistently showed a band of greater intensity than that of the reference gene. Normal human placental DNA and a tem- plate-free control were run in every amplification ex- periment to document the specificity of the amplification reaction and to exclude contamination between samples.

Statistical Analysis

To estimate the agreement beyond chance among the two different observers that evaluated N-myc immunoreactivity, the kappa index was used.” The classifica- tion of N-myc immunoreactivity as either less than or equal to 10% immunoreactive cells or greater than 10% immunoreactive cells to be analyzed was agreed upon, because 10% had been chosen as the cutoff level for prognostic modeling.

Differences between groups of patients were analyzed using the chi-square test or the Fisher’s exact test for categoric variables, and the unpaired Student’s t-test was used for continuous variables.

For survival analysis, variables that were not natu-

N-myc in Medullary Thyroid Carcinoma/RoncaJli e t aJ. 137

rally dichotomous were dichotomized choosing the median value as cutoff level. For N-myc immunoreactivity, the cutoff level of 10% immunoreactive cells was chosen. Observed disease free survival time was the interval between diagnosis of MTC and disease relapse, death from any cause, or final follow-up and was calcu- lated using the Kaplan-Meier estimate.22 Comparison of univariate disease free survival curves between groups defined on the basis of morphologic or clinical variables was made using the log rank test.23 The variables that achieved a significance level of 0.10 in the univariate analysis were included in a multivariate analysis using the Weibull distribution to fit the model to disease free survival time.24

Data were expressed as percentages, means, and standard deviations. Two-tailed P values were used for significance testing. Statistical comparisons were obtained by using the SAS statistical package.25

Results

The individual results of N-myc and PCNA immunostaining and the main clinicopathologic features of the 34 patients are reported in Table 1. N-myc-immunoreactive neoplastic cells were identified in 30 of 34 patients with MTC (88%). Ten tumors showed 10% immunoreactive neoplastic cells or less, whereas the remaining 20 tumors (58% of the patients in the study) exhibited more than 10% immunostained cells (Figure 1, top). Known positive and negative neuroblastomas exhibited the expected immunoreactivity (Figures 1, center and bottom).

The concordance rate between the two observers who assessed N-myc immunoreactivity was 94%. The agreement beyond chance, as expressed by the kappa index, was 0.88, which can be interpreted as almost per- fect agreement.”

N-myc immunoreactivity was consistently localized to the cell cytoplasm and, in a few patients, to the nuclei also. Tumors always demonstrated a certain degree of heterogeneity for N-myc immunostaining (Figure 1, top). The staining intensity was increased commonly at the infiltrative tumor edges and in cells entrapped within the amyloid stroma.

Foci of C-cell hyperplasia were not identified in the tissue sections under study. Normal peritumoral thyroid was present in 10 patients; N-myc immunoreactive cells were not encountered, though normal C-cells, vis- ualized by Grimelius staining, were scattered within the normal glands in 3 of the 10 patients. Occasionally, normal follicular cells showed a dotlike golgian immunoreactivity for N-myc gene product.

Multiplex PCR provided DNA amplification in 28 Of the 34 patients with MTC; in none Of the patients was the intensity of the amplified N-myc band stronger than that of the amplified IFN-7 (Figure 2).

Figure 1. (Top) N-myc immunostaining in a patient showing more than 10% immunoreactive neoplastic cells. Notice the high degree of cell heterogeneity far N-myc staining (original magnification X400). (Center and bottom) N-myc immunoreactivity in a known-positive (center) and a known-negative (bottom) neuroblastoma (original magnification X400).

138 CANCER Jury 2,2994, Volume 74, No. 1

N - MYC 195 bp INT-8 15obp

7 10 11 14 24 28 30 Figure 2. Multiplex polymerase chain reaction in seven patients with of medullary thyroid carcinoma. Two patients (7 and 10) were immunocytochemically negative for N-myc, and five patients (1 1, 14, 24,28, and 30) exhibited more than 10% N-myc immunoreactive cells. The band of the amplified reference gene was always of the same or of greater intensity than that of the amplified N-myc gene sequence.

PCNA immunoreactive cells were detected in all but one patient (Table 1, Figure 3, top and bottom), and 43% of the patients showed a higher than 15% PCNA LI .

The main clinicopathologic features of patients grouped according to N-myc immunoreactivity (i.e., patients with tumors showing more or less than 10% N- myc immunoreactive neoplastic cells) or PCNA LI (i.e., patients with PCNA LI of greater or less than 15%) are reported in Table 2. Patients with tumors showing more than 10% N-myc immunoreactive cells had a significantly higher tumor size (P = 0.031), whereas the PCNA LI did not correlated with any of the above parameters. In particular, no relationship was found between N-myc immunoreactivity and PCNA LI.

Patients had been followed for a mean period of 10.3 years (range, 1-26 years). The cumulative proba- bility of disease recurrence or death at the end of the 5th and 10th years were 35 and 50%, respectively. At univariate analysis, the 21 patients having tumors with more than 10% N-myc immunoreactive cells showed a statistically significant shorter disease free survival than did the remaining patients (Figure 4). Among the other clinicopathologic variables studied including PCNA LI, age, sex, tumor stage and size, and type of disease, male sex and sporadic disease were significantly associated with a shorter disease free survival (Figure 4).

Multivariate analysis indicated that immunoreactivity to N-myc antiserum in more than 10% neoplastic cells (P = 0.039) and male sex (P = 0.050) were inde- pendent predictors of poor prognosis.

Discussion

This study documented a high prevalence of immunoreactivity to N-myc antiserum in MTC, with 58% of the patients with tumors showing more than 10% immunoreactive cells. This figure was in agreement with previous immunocytochemical finding^.'^ Though N-myc proteins are expected to show a nuclear localization, in these patients, we observed mainly a cytoplasmic im-

munoreactivity. This finding, in agreement with previous investigations,26 most likely was due to the cytoplasmic diffusion of the nuclear antigen after formalin fixation. Unfortunately, we could not compare the intensity and localization of the staining in frozen versus formalin fixed, paraffin embedded samples of the patients in this study, most of which have been stored for several years, without any frozen tissue available for in- vestigation. N-myc immunoreactivity was not due to gene amplification, as documented by multiplex PCR analysis performed on DNA extracted from consecutive serial sections adjacent to those immunostained for N- myc protein. These data confirm the results of a previous study, failing to detect N-myc gene amplification in seven patients with MTC.27 Therefore, N-myc immunoreactivity may mirror an upregulated gene transcrip-

Figure 3. (Top and bottom) proliferating cell nuclear antigen immunoreactivity in a case of medullary thyroid carcinoma showing more than 15% immunoreactive neoplastic cells (top, original magnification X400) and in a case with less than 10% immunoreactive cells (bottom, original magnification X400).

N-myc in Medullary Thyroid Carcinoma/Roncalli et al. 139

Table 2. Clinical Features of Diagnosis of MTC in Patients Grouped According to N-myc and PCNA Immunoreactive (IR) Cells

N-myc IR cells PCNA IR ceIls All patients 5 10% > 10% 5 15% > 15% (n = 36) (n = 14) (n = 20) (n = 19) (n = 15)

Mean age (yr) 53.7? 16.7 50.6 & 15.5 55.9 k 17.6 55.5 * 18.2 51.4k 14.9 Males 16 (47) 6 (43) 10 (50) 8 (42) 8 (53) With lymph node metastases 6 (18) 2 (14) 4 (20) 5 (26) 1(7)

With diameter > 3 cm* 16 (50) 3 (23) With PCNA LI > 15% 15 (44) 6 (43) 9 (45)

With sporadic disease 29 (79) 10 (71) 17 (85) 14 (74) 13 (87)

13 (68)t 10 (56) 6 (43) - -

PCNA: proliferating cell nuclear antigen; LI: labeling index. Other than mean age, values are no. (YO). * Not available in two patients. t P = 0.031 versus patients with I 10% IR cells.

tion, as previously suggested for MTC and demonstrated in medulloblastoma.12~17 The appearance of N- myc immunoreactivity in MTC likely is related to the neoplastic transformation and/or to tumor progression, because N-myc protein and N-myc messenger RNA have not been detected in normal C cells." Unfortu- nately, data on N-myc expression in C-cell hyperplasia are not available thus far.

Immunoreactivity to the N-myc antiserum identi- fies patients with MTC who may have a shorter disease free interval. This is of particular relevance, because MTC usually is characterized by a variable, and often, unpredictable clinical behavior. A more favorable clinical course has been suggested for patients with multiple endocrine neoplasm 2a,28-30 though the apparently poorer prognosis of patients with sporadic disease may be related partly to their older age.29 Some authors also reported that in MTC, tumor size and decreased levels of tissue calcitonin indicated a less favorable prognosis,30,31 whereas other authors did not show any correlation between prognosis and histopathologic or immunocytochemical data." Rather, patient's age, tumor stage, and DNA content were shown to correlate with clinical outcome." However, Tenvall et al. did not confirm the prognostic role of age and sex in MTC.32 Col- lectively, these findings reemphasize the need for more reliable prognostic indicators.

In this study, immunostaining of more than 10% neoplastic cells by the N-myc antiserum was the parameter most strongly associated, both in univariate and multivariate analysis, with an unfavorable MTC outcome. N-myc immunoreactivity was also associated with tumors of larger size but, interestingly, the tumor size, per se, did not predict patient survival.

Among the other variables investigated, only male sex and sporadic disease were related, though less strongly than N-myc immunoreactivity, to poor prognosis in univariate analysis. The lack of a significant relationship between tumor onset as localized or meta-

static disease and survival in this series is particularly puzzling. However, only six patients had tumor onset as metastatic disease; therefore, a significant effect of stage might have been missed because of the small number of cases.

Immunoreactivity to N-myc antiserum was not related to PCNA LI, which is a specific parameter of tumor growth fraction and, therefore, of aggressiveness. Furthermore, this study documents that the clinical outcome of MTC cannot be predicted by the evaluation of the neoplastic growth fraction as indicated by PCNA immunostaining. The lack of association between N- myc and PCNA immunoreactivity in MTC is intriguing, because myc-related proteins are currently thought to be involved in inducing and maintaining the cell cy- c1e.33,34 However, additional functions of myc proteins include transactivation of genes critically involved in cell growth and differentiati~n~~ and the induction of apoptosis. Specifically, in neuroblastoma cells, Schweigerer et al.36 demonstrated that the increased activity of the N-myc gene may confer a more aggressive tumor phenotype through the induction of autocrine growth factor(s).

In conclusion, we have documented that immunoreactivity to the OA-11-803 polyclonal antiserum to N- myc provides a prognostic marker for MTC, and that it may, in itself, add useful prognostic information. Not- withstanding the lack of a definite evidence that this immunoreactivity actually reflects genuine N-myc overexpression (due to the unavailability of data from in situ hybridization experiments for N-myc mRNA or from other investigations on mRNA or protein characteriza- tion), we believe that immunostaining with the OA-11- 803 polyclonal antiserum to N-myc (which is to our experience the most suitable reagent for N-myc localization on routinely processed material) is a useful prognostic tool in MTC. Sex and type of disease are less strong predictors, whereas other clinicopathologic parameters, including the growth fraction as evaluated by

140 CANCER July I , 1994, Volume 74, No. 1

N-myc Irnmunonacllvlty 1M) 9

P ;P.M19 80-

60-

%

ra - 20-

T

0 5 10 15 m

I aa

80

20

yeam C I

5 10 15 .20

Sex 1w +

P 4.0395 m.

60.

%

10.

20.

Women

Men -

Clinical type

I

80

w %

40

P IO.035.2

L

a 5 10 15 20 25

PCNA immunostaining, fail to add prognostic information.

References

1. Slamon DJ, Codolphin W, Jones LA, Halt JA, Wong SG, Keith DE, et al. Studies of the HER-Zneu proto-oncogene in human breast and ovarian cancer. Science 1987; 244:707-12.

2. Slebos RJC, Kibbelaar RE, Dalesio 0, Koiistra A, Stam J, Meijer C, et al. K-Ras oncogene activation as a prognostic marker in adenocarcinoma of the lung. N Engl ]Med 1990; 323:561-5.

Figure 4. Univariate survival estimates of patients according to N- niyc and proliferating cell nuclear antigen immunoreactivity, sex, and clinical type and size of medullary thyroid carcinoma.

3. Fujimoto K, Yamada Y, Okajima E, Kakizoe T, Sasaki H, Sugi- mura T, et al. Frequent association of p53 gene mutation in inva- sive bladder cancer. Cancer Res 1992; 52: 1393-8. Sawan A, Randall B, Angus B, Wright C, Henry ]A, Ostrowski J, et al. Retinoblastoma and p53 gene expression related to relapse and survival in human breast cancer: an immunohistochemical

Schwab M, Amlcr LC. Amplification of cellular oncogenes: a predictor of clinical outcome in human cancer. Gene Chrom Can- cer 1990; 1:181-93.

6 . Tsuda H, Shimosato Y, Upton ME, Yokota J, Terada M, Ohira M, et al. Retrospective study on amplification of N-myc and c-myc

4.

study.] Patho! 1992; 168~23-8. 5.

N-myc in Medullary Thyroid Carcinorna/Roncdi et al. 141

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

genes in pediatric solid tumors and its association with prognosis and tumor differentiation. Lab Invest 1988; 59:321-7. Hashimoto H, Daimaru Y, Enjoji M, Nakagawara A. N-myc gene product expression in neuroblastoma. J Clin Pathol 1989; 42:52- 5. Yokoyama T, Tsukahara T, Nakagawa C, Kikuchi T, Minoda K, Shimatake H. The N-myc gene product in primary retinoblasto- mas. Cancer 1989; 63:2134-8. Zimmerman KA, Yancopoulos GD, Collum RG, Smith RK, Kohl NE, Denis KA, et al. Differential expression of myc-family genes during murine development. Nature 1986; 319:780-3. Nisen PD, Zimmerman KA, Cotter SV, Gilbert F, Alt FW. En- hanced expression of the N-myc gene in Wilms tumors. Cancer Res 1986; 46:6217-22. Schroder S, Bocker W, Baisch H, Burk CG, Arps H, Meiners I, et al. Prognostic factors in medullary thyroid carcinomas: survival in relation to age, sex, stage, histology, immunocytochemistry and DNA content. Cancer 1988; 61:806-16. Boultwood J, Wyllie FS, Williams ED, Wynford-Thomas D. N- myc expression in neoplasia of human thyroid C-cells. Cancer Res 1988; 48:4073-7. Klinpfinger M, Ruhri C, Putz B, Pfragner R, Wirnsberger G, Hofler H. Oncogene expression in a medullary thyroid carcinoma. Virch Arch B Cell Pathol 1988; 54:256-9. Roncalli M, Springall DR, Varndell IM, Gaitonde VV, Hamid Q, Ibrahim NBN, et al. Oncoprotein immunoreactivity in human endocrine tumors. JPathol 1991; 163:117-27. Crabbe DCG, Peters J, Seeger RC. Rapid detection of MYCN gene amplification in neuroblastomas using the polymerase chain reaction. Diagn Mol Pathol 1992; 14:229-34. Hall PA, Levison DA, Woods AL, Yu CC, Kellock DB, Watkins JA, et al. Proliferating cell nuclear antigen (PCNA) immunolo- calization in paraffin sections: an index of cell proliferation with evidence of deregulated expression in some neoplasms. J Pathol

Garson JA, Pemberton LF, Sheppard PW, Varndell IM, Coak- ham HB, Kemshead JT. N-myc gene expression and oncoprotein characterisation in medulloblastoma. Br J Cancer 1989; 59:889- 94. Roncalli M, Doglioni C, Springall DR, Papotti M, Pagani A, Po- lak JM, et al. Abnormal p53 expression in lung neuroendocrine tumors: diagnostic and prognostic implications. Diagn Mol Pa-

Ibson JM, Rabbits PH. Sequence of a germ-line N-myc gene and amplification as mechanism of activation. Oncogene 1988; 2:

1990; 162:285-94.

thol 1992; 1:129-35.

399-402.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Grey PW, Goeddel DV. Structure of the human immune interferon gene. Nature 1982; 298:859-63. Landis RJ, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-74. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations.]Am Stat Assoc 1958; 53:457-81. Mantel N. Evaluation of survival data and new rank order sta- tistic arising in its consideration. Cancer Chemofher Rep 1966; 50:

Cox RJ, Oakes D. Analysis of survival data. London: Chapmann and Hall, 1984. SAS Institute. SAS/STAT. Cary, North Carolina: SAS Institute Inc., 1988. Loke S-L, Neckers LM, Schwab G, Jaffe ES. c-myc protein in normal tissue: effects of fixation on its apparent subcellular distribution. Am JPathol 1988; 131:29-37. Yang KP, Nguyen CV, Hickey RC, Samaan NA. C-myc, N-myc, N-ras and c-erB: lack of amplification or rearrangement in human medullary thyroid carcinoma and a derivative cell line. An- ticancer Res 1990; 10:189-92. Sizemore GW. Medullary carcinoma of the thyroid gland. Sem

Samaan NA, Schultz I", Hickey RC. Medullary thyroid carcinoma: prognosis of familial versus sporadic disease and role of radiotherapy. J Clin Endocrinol Metab 1988; 67:801-5. Bigner SH, Mendelsohn G, Wells SA, Cox EB, Baylin SB, Eggle- ston JC. Medullary carcinoma of the thyroid in the multiple endocrine neoplasia IIA syndrome. Am ] Surg Pathol 1981; 5:459- 72. Trump DL, Mendelsohn G, Baylin SB. Discordance between plasma calcitonin and tumor cell mass in medullary thyroid carcinoma. N EnglJMed 1979; 301:253-5. Tennvall J, Bjorklund A, Moller T, Ranstam J, Akerman M. Prog- nostic factors of papillary, follicular and medullary carcinomas of the thyroid gland: retrospective multivariate analysis of 216 patients with a median follow-up of 11 years. Acta Radiol Oncol

Kaczmarek L. Protooncogene expression during the cell-cycle. Lab Invest 1986; 54:365-76. Rosolen A, Whiteseell L, Ikegaki N, Kennett RH, Neckers LM. Antisense inhibition of single copy N-myc expression results in decreased cell growth without reduction of c-myc protein in a neuroepithelioma cell line. Cancer Res 1990; 50:6316-22. Collum RG, Alt FW. Are myc protein transcription factors? Can- cer Cells 1990; 2:69-75. Schweigerer L, Breit S, Wenzel A, Tsunamoto K, Ludwig R, Schwab M. Augmented MYCN expression advances the malig- nant phenotype of human neuroblastoma cells: evidence for induction of autocrine growth factor. Cancer Res 1990; 50:4411-6.

163-70.

Oncol 1987; 14~306-14.

1985; 24:17-24.

prognostic value of n-myc immunoreactivity in medullary thyroid carcinoma

Documents