Synchronous lung cancers defined by deoxyribonucleic acid flow cytometry

J THORAC CARDIOVASC SURG 1991;102:418-24

Synchronous lung cancers defined by deoxyribonucleic acid flow cytometry When two pulmonary tumors are seen simultaneously in patients with lung cancer, it always raises a question as to whether the lesions are synchronous lung cancers or lung cancer with intrapulmonary metastasis. To settle this issue, we used deoxyribonucleic acid ftow cytometry. Deoxyribonucleic acid ploidy patterns of tumors were able to be analyzed in 14 patients with two simultaneous pulmonary tumors resected by operation. These two tumors, with completely different patterns of deoxyribonucleic acid ploidy, were defined as synchronous lung cancers. Tumors were defined as lung cancer with intrapulmonary metastasis when both tumors showed diploidy or when at least one deoxyribonucleic acid index of abnormal clones between two aneuploid tumors was the same or almost identical. Tumors of the five patients with stage I disease were classified as synchronous lung cancers according to the criteria involving deoxyribonucleic acid ftow cytometry. Both tumors were adenocarcinomas in four patients and Iarge-cell and squamous cell carcinomas in one. Both tumors in four patients were located in the same lobe but different segments. AU but one patient with different histologic types are alive without recurrence from 24 to 100 months after operation. Tumors of the nine patients with stage ill disease in whom intrapulmonary metastasis was clinically suspected were classified as lung cancer with intrapulmonary metastasis according to the criteria. These data suggest that deoxyribonucleic acid ftow cytometry of tumors may have diagnostic value in determining synchronous lung cancers.

Yukito Ichinose, MD, Nobuyuki Hara, MD, and Mitsuo Ohta, MD, Fukuoka, Japan

h e problem of evaluating two pulmonary tumors appearing simultaneously is complex. In patients without evidence of extrathoracic tumors the lesions are considered synchronous lung cancers or lung cancer with intrapulmonary metastasis. Martini and Melamed 1 reported pathologic criteria for distinguishing synchronous lung cancers from lung cancer with intrapulmonary metastasis. The 5-year survival rate of patients in whom synchronous stage I lung cancers were diagnosed according to the criteria, and who underwent operation, has been reported to range up to 28%.1,2 When compared with a 5-year survival rate of more than 59.0% in stage I disease with a single tumor," these results were poor. This indicates From the Department of Chest Surgery, National Kyushu Cancer Center, Fukuoka, Japan. Supported in part by a grant in aid of cancer research (62-S-I) from the Ministry of Health and Welfare, Japan. Received for publication March 29, 1990. Accepted for publication July 31,1990. Address for reprints: Yukito Ichinose, MD, Department of Chest Surgery, National Kyushu Cancer Center, I-I, 3-Chome, Notarne, Minami-ku, Fukuoka 815, Japan.

12/1/24588 418

that the patient group with the diagnosis of synchronous stage I lung cancers included patients with lung cancer accompanying intrapulmonary metastasis. In the present study we examined whether it is possible to differentiate between synchronous lung cancers and lung cancer with intrapulmonary metastasis by analyzing the deoxyribonucleic acid (DNA) ploidy pattern of tumors. Materials and methods Between 1974 and 1988, 19 patients with non-small-cell lung cancer simultaneously with other intrapulmonary tumors in ipsilateral lungs were operated on in the National Kyushu Cancer Center. The subjects in the present study were 14 patients whose tumors ena bled us to analyze DNA ploidy pattern by flow cytometry. There were 12 men and two women with a mean age of 62 years (range 40 to 82 years). Pathologic staging was done according to the new international staging system proposed by Mountain.' Five patients had stage I disease. Because our main purpose in the present study was to determine whether analysis of DNA ploidy pattern of tumors gave a clue to the differential diagnosis between synchronous lung cancers and lung cancer with intrapulmonary metastasis, nine patients with stage III disease in whom intrapulmonary metastasis was clinically suspected were examined as control subjects. Eight patients had N 2 disease and one had T3 disease. Preoperative examination included bronchoscopy in all

Volume 102

Number 3 September 1991

DNA flow cytometry

419

Table I. Synchronous lung cancers classified according to DNA ploidy pattern Histology Patient

2 3 4 5

First tumor

Second tumor

TN stage

Operation

Adenocarcinoma (P) (RUL:S2) Adenocarcinoma (P) (RUL:SI) Adenocarcinoma (M) (RUL:S2) Adenocarcinoma (P) (RUL:S2) Large-cell carcinoma (LUL:SI + 2)

Adenocarcinoma (W) (RML:S5) Adenocarcinoma (P) (RUL:S2) Adenocarcinoma (M) (RUL:S3) Adenocarcinoma (P) (RUL:S3) Squamous cell carcinoma (M) (LUL:S3)

T2 NO

Bilobectomy

T2 NO

Lobectomy

Tl NO

Lobectomy

TI NO

Lobectomy

T2NO

Lobectomy

Outcome (mo) NED (100) NED (30) NED (30) NED (24) Died

(I8)

P, Poorly differentiated: W, well differentiated; NED, alive with no evidence of disease: RUL, right upper lobe; RML, right middle lobe; M, moderately differentiated; LUL. left upper lobe; SI, apical segment; Sl + 2, apical-posterior segment; S2, posterior segment; S3, anterior segment; S5, medial segment.

patients, computed axial tomography (CT) of the thorax in six, CT or scintigraphy of the brain in six, CT of the abdomen or liver scintigraphy in 12, bone scintigraphy in II, and gallium scintigraphy in two patients, Two separate lesions were identified by chest x-ray films or CT scans in eight patients. In the other six patients the second lesion was found during thoracotomy (n = 5) and in pathologic examination (n = I), Flow cytometry analysis of cellular DNA content was performedwith paraffin-embedded blocks of surgically resected specimens. Well-preserved areas ofthe tumor were chosen on slides, and the corresponding areas on paraffin blocks were detected, The single cell suspension was made and the cells stained with propidium iodide according to the method described by Tosi and colleagues" and Vindelev, Christensen, and Nissen.' The stained cells were analyzed on an FACS-CAN device (Becton Dickinson, Sunnyvale, Calif.), At least 20,000 cellswere examined in each specimen, and the DNA content per cell, measured by fluorescence intensity, was recorded and displayed in a histogram. The resulting DNA histograms were interpreted without knowledge of patient details. The DNA index (DI) was calculated as the ratio of the aneuploid GoGt peak to the diploid GoGt peak. When only one GoGt peak (DI = I) was observed, the tumor was considered diploid. Tumors that had evidence of at least one distinct GOG l population (DI > 1.0) in addition to a diploid stem line were classified as aneuploid. The lowest peak was assumed to represent diploid GoGt peak (DI = 1). Samples whose coefficient of variation of the diploid GoGt peak exceeded 10%were not used to determine DNA ploidy of the tumor. If the two tumors had completely different patterns of DNA ploidy, they were defined as synchronous lung cancers. Synchronous lung cancers fulfilled the following criteria: (1) One tumor showed diploidy and the other aneuploidy and (2) in case of both aneuploid tumors, each DI of abnormal clones between two tumors was different. When both tumors showed diploidy or at least one DI of abnormal clones between two aneuploid tumors was the same or almost identical, those tumors were defined as lung cancer with intrapulmonary metastasis. The criterion that defined an identical abnormal clone between two aneuploid tumors was the following: a mean DI of the subject clones X 0.975 < Dis of each subject clone < a mean DI of the subject clones X 1.025.

Table II. DIs other than diploid GOG 1 peak (DI = 1) in synchronous lung cancers Patient

First tumor

Second tumor

I 2 3 4 5

2.12 Diploid 1.85 1.27,2.46 1.40

Diploid 2.30 1.35,2.42 Diploid 1.60

Survival curves were computed by the Kaplan-Meier method.

Results Synchronous lung cancer defined by DNA ploidy pattern. Tumors of the five patients with stage I disease were diagnosed as synchronous lung cancers by the criteria involving DNA ploidy pattern. Table I shows patient details. The larger tumor was designated the first tumor and the smaller the second tumor. TN stage was determined according to the histologic characteristics of the first tumor. Histologic examination of both tumors showed adenocarcinomas in four patients and large-cell and squamous cell carcinomas in one. The tumors in four patients were located in the same lobe but different segments. Although one patient (patient 5) with different histologic patterns had recurrence and died 18 months after operation, the other four patients are alive without recurrence from 24 to 100 months after operation. The land 2-year disease-free postoperative survival rates of this group were 80%. Fig. 1shows DNA histograms oftumors in all five patients. DNA ploidy patterns of the first and second tumors in all these patients were completely different from each other. DIs of each tumor, other than diploid GoG, peak (01 = 1), are shown in Table II.

The Journal of Thoracic and Cardiovascular Surgery

420 Ichinose, Hara, Ohta

First tumor

Case 1

0

Second tumor

J-J~~-' 1013

a

200

I lJ

..,,-..1....

'00

2130

''''0

2813

Case 2

I

200

Case 3

Case 4

.-J 0

'00

2""

"

I

1~0

2""

.... .... Q) III

o

Case 5

"'0"

..

Q)

.0

e

"

Z

f·-'

a

.1

''''0

1~~" \

~.,j 2itlO

0

leo

2GO

Channel number (relative DNA content)

Fig. 1. DNA histograms of synchronous lung cancers classified according to DNA ploidy pattern.

Lung cancer with intrapulmonary metastasis defined by DNA ploidypattern. Tumorsof ninepatients with stage III diseasewerediagnosed as lungcancerwith intrapulmonary metastasis by the DNA ploidy criteria, unlike the stage I tumors. As shown in Table III, intrapulmonarymetastasiswas suspected on clinical grounds

intheseninepatientswithstageIII disease. TherewasN2 disease in eight patients,and the second tumor waslocatedin the samesegmentofthe maintumorin threeofthose eightpatients.Furthermore,the thirdtumor wasfound by pathologic examination in twopatients(patients4 and 7). Histologic examination of both tumors showed adeno-

Volume 102 Number 3

DNA flow cytometry

September 1991

421

II

Case 1

II

*

*

Case 4 III

*

* Case 5

e

l'l'l

2'l'l

........III ai

Case 8

I

II

o ..... 0

\.< ai

*

.Q

e

!"~'I /~\--'-"""....}.-.,

:l

z

\

'l

........ l'l'l

*

'.,

2'l'l

Channel number (relative DNA content)

Fig. 2. Representative examples of the DNA histograms of lung cancer with intrapulmonary metastasis classified according to DNA ploidy pattern. Asterisk represents an identical abnormal DNA clone between tumors; I, first tumor; II, second tumor; III, third tumor.

carcinomas in six patients and squamous cell carcinomas in three. Histologic characteristics of the third tumor in two patients (patients 4 and 7) was the same as in the original tumor. The 1- and 2-year disease-free survival rates of this group were 44% and 0%, respectively. Medi-

an survival was 13.5 months. All deaths were related to recurrent tumors. Representative examples of DNA histograms are shown in Fig. 2. Table IV shows all DIs of each tumor other than diploid GoGl peak (01 = 1) in all nine

422

The Journal of Thoracic and Cardiovascular Surgery

Ichinose, Hara, Ohta

Table III. Lung cancer with intrapulmonary metastasis classified according to DNA ploidy pattern Histology Patient

2 3

4 5 6 7 8 9

Second tumor

First tumor

TN stage

Adenocarcinoma (M) (RML:S5) Adenocarcinoma (M) (RUL:SI) Adenocarcinoma (M) (LLL:S6)

Adenocarcinoma (P) (RML:S5) Adenocarcinoma (M) (RUL:S2) Adenocarcinoma (M) (LUL:SI + 2)

Squamous cell carcinoma (P) (RML:S5) Squamous cell carcinoma (M) (RLL:S6) Adenocarcinoma (P) (RLL:S8) Adenocarcinoma (M) (RUL:S3) Adenocarcinoma (P) (RUL:S2) Squamous cell carcinoma (W) (RUL:S2)

Squamous cell carcinoma (P) (RML:S5) Squamous cell carcinoma (M) (RLL:S6) Adenocarcinoma (P) (RUL:S3) Adenocarcinoma (M) (RML,RLL) Adenocarcinoma (P) (RLL:SIO) Squamous cell carcinoma (W) (RUL:S3)

Operation

T3 N2

Pneumonectomy

T2N2

Lobectomy

T2 N2

Pneumonectomy

T2 N2

Pneumonectomy

T2N2

Pneumonectomy

T2N2

Lobectomy + wedge resection Pneumonectomy

T2N2 T2 N2 T3 NO

Bilobectomy + wedge resection Lobectomy

Outcome (mo) Died (35) Died (35) Alive with recurrence (35) Died (16) Died (14) Died (I I) Died (9) Died (6) Died (5)

S6, Superior segment; S8, anterior basal segment; SIO, posterior basal segment.See Table I for other abbreviations. Patients 4 and 7 had a third tumor. which was also analyzed regarding ON A ploidy pattern.

Table IV. DIs other than diploid GoG] peak (DI = 1) in lung cancer with intrapulmonary metastasis Patient

First tumor

I 2 3 4

Diploid 1.74* 1.70* 2.77*

5 6 7

1.14,1.97,* 2.51 1.13, 1.78* Diploid

8 9

1.72,* 2.88 1.49*

Second tumor Diploid 1.74* 1.66,* 2.96 1.62, 2.79* (2.82*) 1.98* 1.83* Diploid (Diploid) 1.69* 1.51*

Parentheses indicate OJ of the third tumor. '01 of identical abnormal ONA clonebetween twoaneuploid tumors.

patients, as well as an identical DI of abnormal clones between the first and second tumors. In the case of patients I and 7, all tumors showed diploidy and were diagnosed as lung cancer with intrapulmonary metastasis. The third tumor in patients 4 and 7 was also defined as being identical to the first tumor, according to DNA ploidy pattern. Discussion We used DNA flow cytometry to attempt to differentiate between synchronous lung cancers and lung cancer

with intrapulmonary metastasis. When both tumors had completely different patterns of DNA ploidy, they were defined as synchronous lung cancers. However, as shown in the present study and other previous reports."? all specimens had normal (diploid) GOGI peaks, which were made by tumor cells and normal epithelial and/or stromal cells within tumors, Even with cytofluorometry of lung cancer, which could theoretically select only tumor cells, most tumors contained normal (diploid) GOG I peak.!? Therefore we focused on the variation of DIs of aneuploid clones in tumors. The prevalence of aneuploid tumors in lung cancer varies for each report."? In the present study, 22 of 30 specimens (73%) showed aneuploidy. When both tumors showed diploidy, or at least one DI of abnormal clones between two aneuploid tumors was the same or almost identical, those tumors were defined as lung cancer with intrapulmonary metastasis. In the case of both being diploid tumors there are no abnormal clones in terms of DNA ploidy; thus their biologic characters are thought to be similar. When at least one DI of abnormal clones between aneuploid tumors is the same or almost identical, both tumors are thought to be related partly to each other. Volm and colleagues J I reported analysis of DNA flow cytometry of 18 primary lung cancers and their lymph node metastases. They found a large variation between primary tumors and lymph node metastases with regard to DNA stem lines. They hereby concluded that flow

Volume 102

DNA flow cytometry 423

Number 3 September 1991

cytometric analysis of lymphnodes provides onlylimited information about primary tumors. However, when our criteriawereapplied in analyzing the data, wefoundthat the biologic character of most metastatic lesions was partly related to that of primary tumors. Namely, one identical abnormalDI existed between primaryand metastatic tumors in 12 patients, 2 identical abnormal DIs between both tumors in one patient, and both tumors showed diploidy in one patient. Therefore in 14 of 18 patients (77.8%) the metastatic lesions werepartly related to primarylesions in terms of DNA ploidy pattern. A similar observation was alsofound in a report by Salvati andcolleagues," inwhichmultiple-site sampling wasdone in lung cancer specimens. In their representative figure showing DNA histograms in core and peripheralsitesof primary tumors and four different sites of lymph node metastases in the same patient, all specimens had an identical abnormalDI. Thosefindings lendsupportto our criteriatodifferentiate between synchronous lungcancers and lung cancer with intrapulmonary metastasis. According to criteria proposed by Martini and Melamed,' ifthe histologic characteristics of bothtumors are the same,cases withtumor permeation into lymphaticvessels sharedby twotumorshaveto be excluded in the analysis of synchronous lung cancers. In cases of unilateraltumors, patientswithhilaror mediastinal lymphnode metastases haveto be excluded. However, this strict rule does notseemto be necessarily usedin all reports.i In the present study lung cancer with intrapulmonary metastasiswas diagnosed by analysis of DNA ploidy pattern in all patients with N2 disease. Therefore our data may support the criteria proposed by Martini and Melamed.' In previous reportssurvival of patientshaving a diagnosis of synchronous stage I lung cancers with no regional lymph nodemetastasis was poor. In the reportof Martini and Melamed,'the 5-yearsurvival rate of 15patientswho underwent operation (13 patientswithstageI disease and two patients with regional lymph nodes metastasis) was only28%. Ferguson and colleagues/ reported that the 2yearsurvival rate of 10patientswith stage I synchronous lung cancers was 60% and median survival time was 27 months. According to their publication there were no 5yearsurvivors, and one and two patientswerealive 3 and 24 months after operation, respectively. There was no information as to whetherthosethree patientswerealive with or without recurrence. However, even if those patients survived 5 years, the 5-yearsurvival rate would beonly30%. Whencomparedwith a 5-yearsurvival rate of morethan 59%in stage I disease with a singletumor;' those results were poor, indicating that a patient group diagnosed as having synchronous stage I lung cancers

included patients with lung cancer accompanying intrapulmonarymetastasis. In general,whenhistologic featuresof the two tumors are different, it may be accepted that they are synchronous lung cancers, regardless of lymph node involvement.l-? However, occasionally a histologic difference exists in the same tumor from one portion to another. Recently, the University of Toronto Lung Oncology Group'? reportedthat of 38 patients whohad beengiven a diagnosis of small-cell lung cancer and who underwent resection after chemotherapy, four and two patients had non-small-cell lung cancers and mixed small and nonsmall-cell lung cancers in residual tumors, respectively. Therefore, eveninthe caseofsmall-cell lungcancerinone tumor and non-smaIl-cell lung cancer in the other tumor, the possibility of lung cancer with intrapulmonary metastasis cannot be excluded. In the presentstudy wedemonstratedthat lungcancer with intrapulmonary metastasis was diagnosed by DNA flow cytometrycriteria in all patients in whomit wassuspectedon clinical grounds. Four of five patientswith synchronous stage I lungcancersdiagnosed by thesecriteria havesurvived withoutrecurrencefrom 24 to 1()() months after operation. Although a definitive conclusion cannot be drawn because of the small sample size, analysis of DNA flow cytometry of tumors, in addition to conventionalpathologic examination, may be a usefulmethodto differentiate between synchronous lung cancersand lung cancer with intrapulmonary metastasis. We thank Lesley Koustaff for critical review and Yumiko Ohshima and Yuko Ishibashi for help in the preparation of the manuscript.

REFERENCES I. Martini N, Melamed MR. Multiple primary lung cancers. J THORAC CARDIOVASC SURG 1975;70:606-12. 2. Ferguson MK, DeMeester TR, DesLauriers J, Little AG, Piraux M, Golomb H. Diagnosis and management of synchronous lung cancers. J THORAC CARDIOVASC SURG 1985;89:378-85. 3. Mountain CF. A new international staging system for lung cancer. Chest 1986;89(suppl):225s-33s. 4. Tosi P, Leoncini L, Cintorino M, et al. Flow cytometric analysis of DNA ploidy pattern from deparaffinized formalin-fixed gastric cancer tissue. Int J Cancer 1988;42:86871. 5. Vindelev LL, Christensen IL, Nissen NI. A detergenttrypsin method for the preparation of nuclei for flow cytometric DNA analysis. Cytometry 1983;3:323-7. 6. Zimmerman PY, Hawson GAT, Bint MH, Parsons PG.

424 Ichinose, Hara, Ohta

Ploidy as a prognostic determinant in surgically treated lung cancer. Lancet 1987;2:530-3. 7. Tirindelli-Danesi D, Teodori L, Mauro F, et al. Prognostic significance of flow cytometry in lung. cancer: a 5-year study. Cancer 1987;60:844-51. 8. Cibas ES, Melamed MR, Zaman MB, Kimmel M. The effect of tumor size and tumor cell DNA content on the survivalof patients with stage I adenocarcinoma of the lung. Cancer 1989;63:1552-6. 9. Salvati F, Teodori L, Gagliardi L, Signora M, Aquilini M, Storniello G. DNA flow cytometric studies of 66 human lung tumors analyzed before treatment. Chest 1989;

The Journal of Thoracic and Cardiovascular Surgery

10. 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. II. Volm M, Mattern J, Vogt-Schaden M, Wayss K. Flow cytometric analysis of primary lung carcinomas and their lymph node metastases. Anticancer Res 1987;7:71-6. 12. Shepherd FA, Ginsberg RJ, Patterson AG, Evans WK, Feld R, the University of Toronto Lung Oncology Group. A prospective study of adjuvant surgical resection after chemotherapy for limited small cell lung cancer. J THORAC CARDIOVASC SURG 1989;97:177-86.

96:1092-8.

Bound volumes available to subscribers Boundvolumesof THE JOURNAL OF THoRAcIC AND CARDIOVASCULAR SURGERY are availableto subscribers(only) for the 1991 issues from the Publisher,at a cost of $64.00for domestic,$86.48 for Canadian, and $82.00 for internationalsubscribersfor Vol. 101 (JanuaryJune) and Vol. 102 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertisingis removed. Copiesare shippedwithin60 days after publicationof the last issueof the volume. The bindingis durable buckram with the JOURNAL name, volumenumber, and year stamped in goldon the spine. Payment must accompany all orders. Contact MosbyYear Book,Inc., SubscriptionServices, 11830 Westline Industrial Drive,St. Louis, Missouri63146-3318, USA; phone (800) 325-4177, ext. 4351. Subscriptionsmust be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.