1
2
3
Significant Correlation between Urinary N1, N12-diacetylspermine and
Tumor Invasiveness in Patients with Clinical Stage IA Non-small Cell
Lung Cancer
4
5
Yusuke Takahashi, MD, PhD 1, 2, Hirotoshi Horio, MD, PhD 1, Koji Sakaguchi, MD, PhD 1 , 4,
6
Kyoko Hiramatsu, PhD 3,
Masao Kawakita, PhD 3
7
8
1
9
Center Komagome Hospital, 3-18-22 Hon-komagome, Bunkyo-ku, Tokyo, Japan
Department of Thoracic Surgery, Tokyo Metropolitan Cancer and Infectious Diseases
10
2
11
2-11-1 Kaga, Itabashi-ku, Tokyo, Japan
12
3
13
Science, 2-1-6 Kami-kitazawa, Setagaya-ku, Tokyo, Japan
14
4
15
Oaza-suzaka, Suzaka, Nagano, Japan
16
Email
17
Yusuke Takahashi: yusuketakahashigts@gmail.com
18
Hirotoshi Horio: hirohori@cick.jp
19
Koji Sakaguchi: sakaguchi-kouji@pref-nagano-hosp.jp
20
Kyoko Hiramatsu: hiramatsu-kk@igakuken.or.jp
Department of General Thoracic Surgery, Teikyo University School of Medicine,
Center for Medical Research Cooperation, Tokyo Metropolitan Institute of Medical
Department of Thoracic Surgery, Nagano Prefectural Suzaka Hospital, 1332
1
1
Masao Kawakita: kawakita-ms@igakuken.or.jp
2
Short title: Urinary diacetylspermine predicts invasiveness in NSCLC
3
Category: Original article
4
Key words: Tumor invasiveness, Urine diacetylspermine, clinical stage IA, non-small
5
cell lung cancer
6
Abbreviations: NSCLC: Non-small cell lung cancer; GGO: Ground-glass opacity;
7
DiAcSpm: N1, N12-diacetylspermine; BSA: Bovine serum albumin; CT: Computed
8
tomography; PET: Positron Emission Tomography scan; HE: hematoxylin and eosin ;
9
VvG: Victoria blue van Gieson; CEA: Carcinoembryonic antigen; TDR: Tumor
10
disappearance rate; CI: Confidence Intervals; HR: Hazard Ratio; IASLC/ATS/ERS: the
11
International Association for the Study of Lung Cancer, American Thoracic Society, and
12
European Respiratory Society ; AIS: adenocarcinoma in situ; MIA: minimally invasive
13
adenocarcinoma; I-ADC: invasive adenocarcinoma;
14
15
CORRESPONDENCE TO:
16
Yusuke Takahashi, Department of General Thoracic Surgery, Tokyo Metropolitan
17
Cancer and Infectious Disease Center Komagome Hospital, 3-18-22 Honkomagome,
18
Bunkyo-ku,
Tokyo,
113-8677,
Japan.
2
Phone:
+81-03-3823-2101,
Fax:
1
2
+81-03-3823-5433, yusuketakahashigts@gmail.com
Competing interests: The authors have no potential conflicts of interest to disclose.
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
34
35
3
1
Abstract
2
3
Background: In order to select better candidates for limited resection, it is necessary to
4
accurately differentiate histological non-invasive tumors from small-sized lung cancer.
5
Urinary N1, N12-diacetylspermine (DiAcSpm) has been reported to be a useful tumor
6
marker in various cancers. The aim of our study was to examine the correlation
7
between preoperative urinary DiAcSpm level and clinicopathological characteristics,
8
especially histological invasiveness, of tumors in patients with clinical stage IA
9
NSCLC.
10
Methods: Urine samples were obtained before treatment from five hundred sixteen
11
consecutive patients who were diagnosed as operable NSCLC at our institution during
12
the period from April, 2008 to January, 2013. Urinary DiAcSpm values were measured
13
by colloid gold aggregation procedure. Among these patients, 171 patients with clinical
14
stage IA were consistent with our study cohort. We defined non-invasive tumors as
15
NSCLC showing no vascular invasion, lymphatic permeation, pleural invasion, or
16
lymph node metastasis. We investigated the correlation between non-invasive tumor and
17
clinicopathological factors including urinary DiAcSpm.
18
Results: The median urine DiAcSpm was 147.2 nmol/g creatinine in males and 161.8
19
nmol/g creatinine in females, which we set as the cut-off value separately by gender.
4
1
The high urinary DiAcSpm group had significantly more frequent elevated serum CEA
2
(p=0.023), lymph node metastasis (p=0.048), lymphatic permeation (p=0.046), and
3
vascular invasion (p=0.010). And tumor size >2.0 cm (p=0.001), serum CEA >5.0mg/dL
4
(p<0.001), high urinary DiAcSpm (p=0.002), and tumor disappearance rate (TDR)
5
<0.75 (p<0.001) were more frequent in patients with invasive tumors than in those with
6
non-invasive tumors. Multivariate analysis revealed that tumor 2.0 cm or smaller
7
(RR=2.901, 95%CI; 1.372-6.136, p=0.005), high urinary DiAcSpm (RR=3.374, 95%CI;
8
1.547-7.361, p=0.002), TDR<0.75 (RR=4.673, 95% CI; 2.178-10.027, p<0.001) were
9
independent predictive factors for invasive tumors.
10
Conclusions: We have successfully shown the significant correlation between urinary
11
DiAcSpm and pathological tumor invasiveness in patients with clinical stage IA
12
NSCLC. Further investigation should elucidate the clinical usefulness of DiAcSpm.
13
14
5
1
Background
2
Surgery is one of the major therapeutic choices for primary lung cancer if that is
3
possibly indicated. Especially in clinical stage IA non-small cell lung cancer (NSCLC),
4
the standard treatment remains lobectomy and systematic hilar and mediastinal lymph
5
node dissection [1]. Advancement in diagnostic techniques has detected small-sized
6
lung tumors more frequently in recent years [2]. Thus, a number of researchers tried to
7
prove the effectiveness of limited resection and a higher local recurrence rate after
8
limited resection was observed even though a negative surgical margin had been
9
confirmed pathologically [3-5]. Possible explanations for local recurrence may be
10
insufficient surgical margin, misdiagnosis of nodal involvement, or intrapulmonary
11
lymphatic spread [6]. Although two randomized control trials comparing limited
12
resection with standard lobectomy in patients with clinical T1aN0M0 NSCLC are
13
currently taking place in Japan, limited resection has often been performed in patients
14
with peripheral small-sized lung cancer. Therefore, it is necessary to differentiate
15
histological non-invasive tumors accurately from these small-sized lung cancers in order
16
to select better candidates. Several researchers have reported that a greater proportion of
17
ground-glass opacity (GGO) was a significant predictor of non-invasive lung cancer [7,
18
8]. Moreover, a recent report described that the presence of a micropapillary component
6
1
was independently associated with the risk of recurrence in patients with stage I NSCLC
2
treated with limited resection [9].
3
N1, N12-diacetylspermine (DiAcSpm) is a minor component of urinary polyamines
4
which constitutess less than 0.5% of the total polyamines in healthy human urine [10]. It
5
has been widely accepted that actively proliferating cells tend to excrete more
6
polyamines as a result of the activation of intracellular polyamine metabolism and
7
turnover. Therefore, DiAcSpm is also found to be increased in advanced stage cancer
8
[11]. In recent years, we have reported that urinary DiAcSpm is significantly and
9
frequently elevated in patients with various cancers, including early stage disease [11].
10
Based on this background, we aimed to examine the correlation between preoperative
11
urinary DiAcSpm level and clinicopathological characteristics, especially histological
12
invasiveness, of tumors in patients with clinical stage IA NSCLC.
13
14
15
Patients and methods
Urine samples were obtained before treatment from five hundred sixteen
16
consecutive patients who were diagnosed with operable NSCLC at Tokyo Metropolitan
17
Cancer and Infectious Disease Center, Komagome Hospital during the period from
18
April 2008 to January 2013. Among them, 171 consecutive patients with clinical stage
7
1
IA were consistent with our study cohort. We received prior approval to use urine
2
samples from the ethical committees of the Tokyo Metropolitan Cancer and Infectious
3
Disease Center Komagome Hospital and the Tokyo Metropolitan Institute of Medical
4
Science. Institutional Review Board-approved informed consent was obtained from all
5
the patients.
6
7
8
9
Determination of urinary DiAcSpm by colloid gold aggregation procedure
The urine samples were supplemented with 3mmol/L NaN3 and stored frozen at
-20°C until use, as previously reported [12]. DiAcSpm was measured using a colloidal
10
gold aggregation procedure on a JCM BM-6010 automatic biochemical analyzer (JEOL,
11
Tokyo, Japan). The colloidal gold aggregation procedure is based on the specific
12
binding of a bovine serum albumin(BSA)-acetylspermine conjugate, which is a
13
DiAcSpm mimic, and a stable red-purple solution of colloidal gold-antibody complexes.
14
The BSA-acetylspermine conjugate induces a color change from red-purple to grey by
15
causing the aggregation of colloidal gold particles. When a urinary sample containing
16
DiAcSpm is added to the system, DiAcSpm competitively binds to the colloidal
17
gold-antibody complexes and suppresses the color change caused by aggregation
18
because it is a monovalent antigen that cannot cross-link multiple gold particles. The
8
1
concentration of DiAcSpm in a urine sample can be determined by measuring this color
2
change. The reagent Auto DiAcSpm® (Alfresa Pharma Co., Osaka, Japan), which is
3
intended for use in automatic clinical analyzers, has been developed. The analytical
4
values of DiAcSpm determined by the colloidal gold aggregation procedures have been
5
shown to closely correlate with those determined by mass spectrometric analysis [13].
6
The creatinine levels were measured enzymatically using the NESCAUTO® VLII CRE
7
reagent (Alfresa Pharma Co., Osaka, Japan) on a JCM BM-6010 automatic biochemical
8
analyzer (JEOL, Tokyo, Japan).
9
10
11
Clinicopathological assessment
All of the clinicopathological data were retrieved from medical records.
12
Preoperative evaluation included a physical examination, a blood chemistry analysis,
13
the measurement of tumor markers, and bronchoscopy, chest radiography, and
14
computed tomography (CT), brain MRI, and bone scintigraphy. Integrated positron
15
emission tomography scan and CT scan (PET/CT) was also performed when
16
appropriate. Clinical lymph node metastasis was defined as enlarged lymph nodes
17
measuring > 1cm on the short axis by CT scan and /or hypermetabolic lymph nodes on
18
PET/CT scans. Histological confirmation of lymph node metastasis was made by
9
1
endobronchial ultrasound-guided transbronchial needle aspiration of enlarged lymph
2
nodes. All patients underwent a lobectomy or bilobectomy and systematic lymph node
3
dissection for the resection of the primary lesion.
4
All surgical specimens underwent pathological examination. Each tumor was
5
diagnosed according to current World Health Organization histological classification
6
[14] and was staged according to the seventh edition of the tumor node metastasis
7
classification of the International Union Against Cancer, 7th edition [15]. Vascular
8
invasion, pleural invasion, and lymphatic permeation were evaluated using both of
9
hematoxylin and eosin (HE) staining and Victoria blue van Gieson (VvG) staining
10
11
12
sections.
We defined non-invasive tumor as NSCLC showing no vascular invasion,
lymphatic permeation, pleural invasion, or lymph node metastasis.
13
We reviewed the medical records of each patient for clinicopathologic information
14
as follows: age (categorized into two group according to median age , ≤69 years and
15
those >69years), gender, tumor size, preoperative serum CEA level (dichotomized at the
16
normal upper limit of 5mg/dL),
17
invasion, lymphatic permeation, pleural invasion, histologic type, and pathological
18
stage.
pathological lymph node involvement, vascular
10
1
Of 171 NSCLC lesions, 140 adenocarcinomas were also classified according to the
2
newly proposed adenocarcinoma classification proposed by the International
3
Association for the Study of Lung Cancer, American Thoracic Society, and European
4
Respiratory Society (IASLC/ATS/ERS) [16]. The tumors were divided into three
5
groups: (1) adenocarcinoma in situ (AIS), (2) minimally invasive adenocarcinoma
6
(MIA), and (3) invasive adenocarcinoma (I-ADC) and we analyzed correlation between
7
their histological invasiveness defined in this classification and clinicopathological
8
factors as described above.
9
10
Measurement of tumor disappearance rate on chest CT
11
We calculated tumor disappearance rate (TDR) using the following definition of
12
tumor dimensions on chest high-resolution CT [17]: pDmax, the maximum dimension
13
of a tumor on pulmonary window setting images; pDperp, the largest dimension
14
perpendicular to the maximum axis on pulmonary window setting images; mDmax, the
15
maximum dimension of a tumor on mediastinal window setting images; mDperp, the
16
largest dimension perpendicular to the maximum axis on mediastinal window setting
17
images. TDR was calculated by the following formula:
18
TDR = 1 – (mDmax × mDperp) / (pDmax × pDperp)
11
1
We set the threshold of TDR at 0.75 as previously described [18].
2
3
4
Statistical analysis
Two-category comparisons were performed using the Pearson χ2 test and the
5
Fisher exact test for categorical variables, or Mann-Whitney U test were performed for
6
continuous variables. All tests were two-sided, and p-values less than 0.05 were
7
considered statistically significant. Multivariate analysis was performed using logistic
8
regression model to determine the impact of factors that were shown to be significant
9
survival predictors by univariate analysis. The statistical analysis was performed using
10
SPSS software (version 20; SPSS Inc., Chicago, III).
11
12
Results
13
Patient characteristics
14
Our cohort consisted of 84 males and 87 females. Ages ranged from 36 to 89 years
15
with a median of 69 years. Actual tumor size of resected specimens ranged from 0.8 to
16
3.5 cm with a median of 1.8 cm. There were 137 adenocarcinomas, 24 squamous cell
17
carcinomas, 10 of other histology (including 2 adenosquamous carcinomas, 2 large cell
18
carcinomas, 2 non-small cell carcinomas, 2 large cell neuroendocrine carcinomas, 1
19
clear cell carcinoma, and 1 carcinoid). The number of pathological stage IA tumors was
12
1
136, 121 of stage IB, 7 of stage IIA, and 7 of stage IIIA. Thirty-five patients were
2
up-staged after surgery due to lymph node metastasis in 15, actual tumor size > 3cm in
3
eight and pleural invasion in 26 patients. Lymphatic permeation was seen in 24 and
4
vascular invasion in 39 patients. A summary of patient characteristics and pathological
5
characteristics is presented in Table 1. The median urine DiAcSpm was 147.2 (range:
6
3.7-3918.8) nmol/g creatinine in males and 161.8 (range: 66.0-867.6) nmol/g creatinine
7
in females. We set the urine DiAcSpm median as the cut-off value with the genders
8
separated because previous reports demonstrated that the urine DiAcSpm values of
9
healthy women were higher than that of men [12].
10
11
12
Measurement of tumor disappearance rate on chest computed tomography
When we applied 0.75 as the cut-off value of TDR to test correlation with
13
pathological invasiveness, we achieved sensitivity and specificity values of 59.2% and
14
74.6%, respectively with the best predictive accuracy of 70.2%. The urinary DiAcSpm
15
had an accuracy of 60.8%, sensitivity of 69.4%, and specificity of 57.4% for prediction
16
of pathological invasiveness with a cut-off value of 147.2 for males and 161.8 for
17
females. Moreover, the predictive value of non-invasive tumor was 94.9% (37of 39)
18
when the TDR and urinary DiAcSpm were inversely combined.
13
1
2
3
Correlation between clinicopathological characteristics and urinary DiAcSpm
We looked for a correlation between urinary DiAcSpm and clinicopathological
4
characteristics (Table2). Compared with the low urinary DiAcSpm group, the high
5
urinary DiAcSpm group had significantly more frequent elevated serum CEA (p=0.023),
6
lymph node metastasis (p=0.048), lymphatic permeation (p=0.046), and vascular
7
invasion (p=0.010).
8
Mann-Whitney U test to compare between the absolute value of urinary DiAcSpm
9
and serum CEA level ( ≤5.0 mg/dL v.s. >5.0 mg/dL), lymph node metastasis (N0 v.s.
10
N1-2), vascular invasion (positive v.s. negative), lymphatic permeation (positive v.s.
11
negative), histological type (adenocarcinoma v.s. others), tumor size (≤2.0 cm v.s.
12
>2.0cm), and TDR (≥0.75 v.s. <0.75) gave results consistent with those presented in
13
Table 2 as follows: Urinary DiAcSpm of the normal serum CEA group and N0 group
14
was significantly lower than that of elevated serum CEA group (p=0.044) and N1-2
15
group (p=0.014), respectively. Urinary DiAcSpm of the negative vascular invasion
16
group was significantly lower than that of positive vascular invasion group (p=0.002).
17
Urinary DiAcSpm of the negative lymphatic permeation group was significantly lower
18
than that of positive lymphatic permeation group (p=0.038). In contrast, urinary
14
1
DiAcSpm was not significantly different between adenocarcinoma and others (p=0.585),
2
tumor size ≤2.0 cm and >2.0cm (p=0.249) and TDR (≥0.75 and <0.75 (p=0.489)
3
4
Correlation between pathologically proven cancer invasiveness and clinicopathological
5
factors
6
We investigated the relationship between pathologic invasive factors and
7
clinicopathological factors (Table 3). There were 122 cases of non-invasive tumors and
8
49 cases of invasive tumors. A tumor size >2.0 cm (p=0.001), serum CEA >5.0 mg/dL
9
(p<0.001), high urinary DiAcSpm (p=0.002), and TDR>0.75 (p<0.001) were more
10
frequent in patients with invasive tumors than in those with non-invasive tumors. Age,
11
gender, smoking history, and histological type did not significantly differ regarding
12
pathologically proven tumor invasiveness.
13
14
15
Multivariate analysis
We then performed a multivariate analysis to determine the independent predictors
16
of pathologically proven invasive tumors (Table 4). A tumor size >2.0 cm (Risk ratio
17
(RR) =2.871, 95% confidence interval (CI); 1.347-6.119, p=0.006), high urinary
18
DiAcSpm (RR=3.374, 95%CI; 1.547-7.361, p=0.002), and TDR<0.75 (RR=6.103, 95%
15
1
CI; 1.962-18.981, p<0.001) were independent predictors for invasive tumors. However,
2
serum CEA did not independently correlate with pathologically proven invasive tumors.
3
4
Correlation between histological invasiveness defined in IASLC/ATS/ERS classification
5
and clinicopathological characteristics
6
Of 171 NSCLC lesions, 140 stage IA adenocarcinomas were separately analyzed
7
for the correlation between histological invasiveness defined in IASLC/ATS/ERS
8
classification and clinicopathological factors. There were 87 cases of non-invasive
9
adenocarcinomas and 53 cases of invasive adenocarcinomas. Male gender (p=0.010),
10
Smoker (p=0.033), a tumor size >2.0 cm (p<0.001), serum CEA >5.0 mg/dL (p=0.006),
11
high urinary DiAcSpm (p<0.001), and TDR>0.75 (p=0.023) were more frequent in
12
patients with invasive tumors than in those with non-invasive tumors (Supplementary
13
table 1).
14
1.380-7.650, p=0.007), high urinary DiAcSpm (RR=8.208, 95%CI; 3.470-19.417,
15
p<0.001), and TDR<0.75 (RR=2.783, 95% CI; 1.090-7.108, p=0.032) were independent
16
predictors for invasive tumors. However, gender, smoking history, and serum CEA did
17
not independently correlate with histological invasiveness defined in IASLC/ATS/ERS
18
classification in clinical stage IA adenocarcinomas.
Supplementary table 2 shows that a tumor size >2.0 cm (RR =3.249, 95% CI;
16
1
2
Discussion
3
4
The extent of pulmonary resection for small NSCLC remains a considerable concern
5
to thoracic surgeons currently. The major issue is which subgroup of NSCLC should be
6
considered as a candidate for limited resection. Several investigators have reported that
7
pathologically proven invasive factors were not rare and do not have a negligible
8
recurrence rate, even if a patients’ disease is classified as clinical stage IA NSCLC [1,
9
19]. In fact, some randomized controlled trials are currently ongoing which compare the
10
outcome of lobectomy with that of limited resection in patients with clinical stage IA
11
NSCLC.
12
Previous reports have described that pathological invasive factors of resected
13
NSCLC were strongly correlated with more frequent nodal involvement and poorer
14
outcomes [19, 20]. Based on this background, it has been widely accepted that
15
predicting pathologic non-invasiveness is essential for indicating limited resection. A
16
number of retrospective studies have indicated that tumor non-invasiveness was
17
frequently confirmed in resected lung cancers with a largest diameter of ≤ 2 cm [8, 20],
18
but there are reports of 6% to 12% lymph node metastasis among small-sized NSCLC
19
cases [19]. In the current study, positive lymph node metastasis was observed in 14 of
17
1
171 (8.2%) patients, which is consistent with previous reports. Accordingly, it was
2
suggested that tumor size alone cannot predict the pathologic invasiveness of NSCLC.
3
Several reports have demonstrated that TDR on HRCT was a significant predictor of
4
pathologic invasiveness in small-sized NSCLC [8, 19]. Alternatively it has been
5
reported that the percentage of solid opacity of a tumor on HRCT is a useful predictor of
6
a non-invasive tumor [20-22]. These reports also demonstrated that TDR alone cannot
7
completely predict non-invasive tumors.
8
We have successfully demonstrated that urinary DiAcSpm is one of the useful markers
9
significantly correlating with pathologically proven non-invasive tumor, which may be
10
candidates for limited resection. DiAcSpm is one of the minor polyamine components
11
secreted in human urine. Polyamine excretion is increased by tumor cell proliferation
12
via activation of intracellular polyamine metabolism and turnover [23]. Chen and
13
coauthors demonstrated an increase in DiAcSpm levels associated with the stimulation
14
of oxidative catabolism of polyamines [24]. Moreover, Kuwata et al. recently reported
15
DiAcSpm level to be increased in tumor tissues from both primary site and liver
16
metastasis, suggesting that it might be produced by cancer cells themselves [25].
17
18
We have previously reported that DiAcSpm is frequently elevated in patients with
various cancers such as colorectal cancer, breast cancer, lung cancer, prostate cancer,
18
1
testicular cancer, renal cancer, and renal pelvic cancer with very low false-negative
2
incidence [11, 26]. It should be noted that the urinary DiAcSpm level is more frequently
3
elevated at an earlier stage of colorectal and breast cancer compared to conventional
4
tumor markers such as CEA and CA19-9. In addition, it was reported that poor
5
prognosis of patients with urogenital malignancies was associated with increased
6
urinary DiAcSpm [27]. Hiramatsu et al. also reported that urinary DiAcSpm value
7
revealed strong positive correlation with disease progression [27], which is consistent
8
with the current results that urinary high DiAcSpm was positively correlated with tumor
9
invasiveness (i.e., lymphatic permeation, vascular invasion, and lymph node metastasis).
10
We therefore considered the current result reasonable that urinary DiAcSpm value is
11
significantly associated with pathologically proven tumor invasiveness in stage IA
12
NSCLC.
13
The value for urinary DiAcSpm is usually presented normalized with creatinine, as
14
nmol DiAcSpm/g creatinine. Because DiAcSpm is not reabsorbed from the renal brush
15
border, the glomerular clearance of DiAcSpm is comparable with that of creatinine [28].
16
This fact implies that DiAcSpm which is produced in early stage cancer tissues and
17
excreted into circulation is recovered in the urine without significant loss, which may
18
explain why DiAcSpm serves as a useful tumor marker which is highly sensitive for
19
1
early stage cancers [11]. In one of our previous studies, we reported that the value for
2
urinary DiAcSpm in a healthy male significantly differed from that of a healthy female
3
[12]. We therefore employed separately the cut-off value for urinary DiAcSpm by
4
gender.
5
We considered the TDR to be able to complement the urinary DiAcSpm value by its
6
high specificity for predicting non-invasive tumors. They were proven to be
7
independently correlated with pathological tumor invasiveness. Thus, we have shown
8
that the combination of TDR and urinary DiAcSpm was strongly correlated with
9
pathological invasiveness. Our further analysis on the correlation between
10
clinicopathological characteristics and histological invasiveness defined in
11
IASLC/ATS/ERS classification in adenocarcinomas (Supplementary table 1, 2),
12
strongly supported the results of analysis on the original patient cohort (table 2-4). In
13
fact, the latter analysis revealed a clearer correlation between histological invasiveness
14
and urinary DiAcSpm.
15
The mechanisms underlying the increase in urinary DiAcSpm value in cancer
16
patients have not been fully understood, although a considerable amount of literature
17
has reported the usefulness of urinary DiAcSpm as a novel tumor marker. The current
20
1
study and further investigations may contribute in clarifying the mechanism and clinical
2
significance of DiAcSpm.
3
Several limitations of the current study may exist because of its design including
4
the retrospective data collection in small cohort of patients and possibly indwelling bias
5
for selecting clinical stage IA disease. Although HRCT were retrospectively reviewed
6
by two experienced observers who were blind to patient identification, image evaluation
7
is essentially subjective and our evaluation may lack reproducibility. In addition, we did
8
not definitively prove indication for limited resection because our data included no
9
survival data. We therefore should follow up with the patients in the current cohort in
10
the future.
11
12
13
Conclusion
In conclusion, we have successfully shown the significant correlation between
14
urinary DiAcSpm and pathological tumor invasiveness in patients with clinical stage IA
15
NSCLC. Further investigations should elucidate the oncological significance and
16
clinical usefulness of DiAcSpm.
17
18
Conflicts of Interest
21
1
2
The authors have reported that no potential conflicts of interest exist with any
companies/organizations whose products or services may be discussed to in this article.
3
4
5
6
7
Acknowledgments
The authors thank Dr. Keith Kretzmer, Medical Communication Advisor of
Department of General Thoracic Surgery of Teikyo University School of Medicine, for
their editorial review of this manuscript.
8
9
10
11
12
13
14
15
16
17
18
19
20
Author contributions: Dr. Takahashi had full access to all of the data in this study. Dr.
Horio takes responsibility for accuracy of the data analysis and Dr. Sakaguchi takes
responsibility for the integrisity of the data.
Dr. Takahashi: contributed to the design and coordination of the study, prepared the
manuscript, read and approved the final manuscript, and served as principle author.
Dr. Horio: contributed to preparing the manuscript and read and approved the final
manuscript.
Dr. Sakaguchi: contributed to the design and coordination of the study, revised the
article for important intellectual content, read and approved the final manuscript.
Dr. Hiramatsu: contributed to performing biochemistry analysis and read and approved
the final manuscript.
21
Dr. Kawakita: contributed to design of the study, preparing the manuscript, and read and
approved the final manuscript.
22
Reference
23
1. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited
24
resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann
25
Thorac Surg 1995, 60:615-623.
26
2. Yankelevitz DF, Reeves AP, Kostis WJ, Zhao B, Henschke CI. Small pulmonary
27
nodules: volumetrically determined growth rates based on CT evaluation.
28
Radiology 2000, 217:251-256.
22
1
3.
Tsubota N, Ayabe K, Doi O, Mori T, Namikawa S, Taki T, Watanabe Y. Ongoing
2
prospective study of segmentectomy for small lung tumors. Study Group of
3
Extended Segementectomy for Small Lung Tumor. Ann Thorac Surg 1998,
4
66:1787-1790.
5
4. Yoshida J, Nagai K, Yokose T, Nishimura M, Kakinuma R, Ohmatsu H, Nishiwaki Y.
6
Limited resection trial for pulmonary ground-glass opacity nodules: fifty-case
7
experience. J Thorac Cardiovasc Surg 2005, 129:991-996.
8
5. Nakao M, Yoshida J, Goto K, Ishii G, Kawase A, Aokage K, Hishida T, Nishimura
9
M, Nagai K. Long-term outcomes of 50 cases of limited-resection trial for
10
11
pulmonary ground-glass opacity nodules. J Thorac Oncol 2012, 7:1563-1566.
6. Ichinose Y, Yano T, Tokoyama H, Inoue T, Asoh H, Katsuda Y. The correlation
12
between tumor size and lymphatic vessel invasion in resected peripheral stage I
13
non-small-cell lung cancer. A potential risk of limited resection. J Thorac
14
Cardiovasc Surg 1994, 108:684-686.
15
7. Suzuki K, Asamura H, Kusumoto M, Kondo H, Tsuchiya R. “Early” peripheral
16
lung cancer: prognostic significance of ground glass opacity on thin-section
17
computed tomographic scan. Ann Thorac Surg 2002, 74:1635-1639.
23
1
8.
Shimada Y, Yoshida J, Hishida T, Nishimura M, Ishii G, Nagai K. Predictive
2
factors of pathologically proven noninvasive tumor characteristics in T1aN0M0
3
peripheral non-small cell lung cancer. Chest 2012, 141:1003-1009.
4
9.
Nitadori J, Bograd AJ, Kadota K, Sima CS, Rizk NP, Morales EA, Rusch VW,
5
Travis WD, Adusumilli PS. Impact of micropapillary histologic subtype in
6
selecting limited resection vs lobectomy for lung adenocarcinoma of 2cm or
7
smaller. J Natl Cancer Inst 2013, 105:1212-1220.
8
9
10. Kawakita M, Hiramatsu K. Diacetylated derivatives of spermine and spermidine
as novel promising tumor markers. J Biochem 2006, 139:315-322.
10
11. Hiramatsu K, TakahashiK, Yamaguchi T, Matsumoto H, Miyamoto H, Tanaka S,
11
Tanaka C, Tamamori Y, Imajo M, Kawaguchi M, Toi M, Mori T, Kawakita M. N1,
12
N12-diacetylspermine as a sensitive and specific novel tumor marker for early-
13
and late- stage colorectal and breast cancers. Clin Can Res 2005, 11:2986-2990.
14
12. Hiramatsu K, Sakaguchi K, Fujie N, Saitoh F, Takahama E, Moriya SS, et al.
15
Excretion of N1, N12-diacetylspermine in the urine of healthy individuals. Ann
16
Clin Biochem 2013, 51:459-467..
17
13. Samejima K, Hiramatsu K, Takahashi K, Kawakita M, Kobayashi M, Tsumoto H,
18
Kohda K. Identification and determination of urinaryacetylpolyamines in
24
1
cancer patients by electrospray ionization and time-of-flight mass spectrometry.
2
Anal Biochem 2010, 401:22-29.
3
14. Travis WD BE, Mueller-Hermelink HK, Harris CC. World Health Organization
4
classification of tumours: pathology and genetics of tumours of the lung, pleura,
5
thymus and heart. 3rd edition. Lyon: IARC Press; 2004.
6
15. Goldstraw P, Crowley J, Chansky K, Giroux DJ, Groome PA, Rami-Porta R,
7
Postmus PE, Rusch V, Sobin L; International Association for the Study of Lung
8
Cancer International Staging Committee; Cancer Research and Biostatistics;
9
Observers to the Committee; Participating Institutions. The IASLC Lung Cancer
10
Staging Project: proposals for the revision of the TNM stage groupings in the
11
forthcoming (seventh) edition of the TNM Classification of malignant tumours.
12
J Thorac Oncol 2007, 2:706-714.
13
16. Travis WD, Brambilla E, Noguchi M, Nicholson AG, Geisinger KR, Yatabe Y, Beer
14
DG, Powell CA, Riely GJ, Van Schil PE, Garg K, Austin JH, Asamura H, Rusch VW,
15
Hirsch FR, Scagliotti G, Mitsudomi T, Huber RM, Ishikawa Y, Jett J,
16
Sanchez-Cespedes M, Sculier JP, Takahashi T, Tsuboi M, Vansteenkiste J, Wistuba I,
17
Yang PC, Aberle D, Brambilla C, Flieder D, Franklin W, Gazdar A, Gould M,
18
Hasleton P, Henderson D, Johnson B, Johnson D, Kerr K, Kuriyama K, Lee JS,
25
1
Miller VA, Petersen I, Roggli V, Rosell R, Saijo N, Thunnissen E, Tsao M,
2
Yankelewitz D. International association for the study of lung cancer/american
3
thoracic society/european respiratory society international multidisciplinary
4
classification of lung adenocarcinoma. J Thorac Oncol 2011, 6:244-285.
5
17. Takahashi Y, Ishii G, Aokage K, Hishida T, Yoshida J, Nagai K. Distinctive
6
histopathological features of lepidic growth predominant node-negative
7
adenocarcinomas 3-5 cm in size. Lung Cancer 2013, 79:118-124.
8
18. Takahashi M, Shigematsu Y, Ohta M, Tokumasu H, Matsukura T, Hirai T. Tumor
9
invasiveness as defined by the newly proposed IASLC/ATS/ERS classification
10
has prognostic significance for pathologic stage IA lung adenocarcinoma and
11
can be predicted by radiologic parameters. J Thorac Cardiovasc Surg 2014,
12
147:54-59.
13
19. Okada M, Nishio W, Sakamoto T, Uchino K, Tsubota N. Discrepancy of computed
14
tomographic image between lung and mediastinal windows as a prognostic
15
implication in small lung adenocarcinoma. Ann Thorac Surg 2003, 76:1828-1832.
16
20. Yokose T, Suzuki K, Nagai K, Nishiwaki Y, Sasaki S, Ochiai A. Favorable and
17
unfavorable morphological prognostic factors in peripheral adenocarcinoma of
18
the lung 3cm or less in diameter. Lung Cancer 2000, 29:179-188.
26
1
21. Yoshida J, Nagai K, Yokose T, Takahashi K, Nishimura M, Goto K, Nishiwaki Y.
2
Primary peripheral lung carcinoma smaller than 1cm in diameter. Chest 1998,
3
114:710-712.
4
22. Matsuguma H, Oki I, Nakahara R, Suzuki H, Kasai T, Kamiyama Y, Igarashi S,
5
Mori K, Endo S, Yokoi K. Comparison of three measurements on computed
6
tomography for the prediction of less invasiveness in patients with clinical stage
7
I non-small cell lung cancer. Ann Thorac Surg 2013, 95:1878-1884.
8
9
10
23. Russell DH, Levy CC. Polyamine accumulation and biosynthesis in a mouse
L1210 leukemia. Cancer Res 1971, 31:248-251.
24. Chen Y, Kramer DL, Li F, Poter CW. Loss of inhibitor of apoptosis proteins as a
11
determinant of polyamine analog-induced apoptoses in human melanoma cells.
12
Oncogene 2003, 22:4964-4972.
13
25. Kuwata G, Hiramatsu K, Samejima K, Iwasaki K, Takahashi K, Koizumi K,
14
Horiguchi S, Moriya SS, Kobayashi M, Kawakita M. Increase of N1,
15
N12-diacetylspermine in tissue from colorectal cancer and its liver metastasis. J
16
Cancer Res Clin Oncol 2013, 139:925-932.
17
26. Sugimoto M, Hiramatsu K, Kamei S, Kinoshita K, Hoshino M, Iwasaki K,
18
Kawakita M. Significance of urinary N1, N8-diacetylspermidine and N1,
27
1
N12-diacetylspermine as indicator of neoplastic diseases. J Cancer Res Clin
2
Oncol 1995, 121:317-319.
3
27. Hiramatsu K, Sugimoto M, Kamei S, Hoshino M, Kinoshita K, Iwasaki K,
4
Kawakita M. Diagnosis and prognostic usefulness of N1, N8-diacetylspermidine
5
and N1, N12-diacetylspermine in urine as novel tumor markers of malignancy. J
6
Cancer Res Clin Oncol 1997, 123:539-545.
7
28. Miki T, Hiramatsu K, Kawakita M. Interaction of N1, N12-diacetylspermine with
8
polyamine transport systems of polarized porcine renal cell line LLC-PK1. J
9
Biochem 2005, 30:1943-1946.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
Tables 1 Baseline characteristics of initial study cohort (n=171)
Age
Median (range)
69 (36-89)
Gender
male
84
female
87
Smoking history
Never-smoker
58
Smoker
113
Synchronous multiple lung cancer
yes
4
no
167
Histological type
adenocarcinoma
137
squamous cell carcinoma
24
large cell carcinoma
2
large cell neuroendocrine carcinoma
2
adenosquamous caricnoma
2
carcinoid
1
clear cell carcinoma
1
NSCLC, NOS
2
Lymphatic permeation
negative
147
positive
24
Vascular invasion
negative
132
positive
39
Pleural invasion
negative
145
positive
26
Pathological stage
IA
136
IB
21
IIA
7
29
IIB
0
IIIA
7
1
2
Table 2 Correlation between urine DiAcSpm and clinicopathological factors
Factors
Urine DiAcSpm level (nmol/g creatinine)
Low
High
≤69
47
40
>69
38
46
Never-smoker
32
26
Smoker
53
60
≤2.0
55
49
>2.0
30
37
≤5.0
79
69
>5.0
6
17
25
24
60
62
adenocarcinoma
68
69
non-adenocarcinoma
17
17
82
75
3
11
negative
78
69
positive
7
17
negative
73
59
positive
12
27
p-value *
Age (years)
0.286
Smoking history
0.335
Tumor size (cm)
0.348
Serum CEA level (mg/dL)
0.023
TDR
≥0.75
<0.75
0.867
Histological type
0.970
Lymph node metastasis
N0
N1-2
0.048
Lymphatic permeation
0.046
Vascular invasion
30
0.010
Pleural invasion
1
pl0
75
70
pl1-2
10
16
0.287
*: Fisher’s exact test, DiAcSpm = diacetylspermine, CEA = carcinoembryonic antigen level.
2
3
4
Table 3 Relationship between pathologic invasive factors and clinicopathological
factors
Factors
Invasive tumor
Non-invasive tumor p-value *
Age (years)
≤69
>69
19
30
68
54
0.062
Gender
male
female
30
19
54
68
0.062
Smoking history
Never-smoker
Smoker
11
38
47
75
0.051
Tumor size (cm)
≤2.0
>2.0
20
29
84
38
0.001
Serum CEA level (mg/dL)
≤5.0
>5.0
35
14
113
9
<0.001
Histological type
adenocarcinoma
non-adenocarcinoma
35
14
102
20
0.090
(nmol/g creatinine)
low
high
15
34
70
520
0.002
TDR
≥0.75
<0.75
4
45
45
77
<0.001
Urine DiAcSpm level
5
*: Fisher’s exact test, DiAcSpm = diacetylspermine, CEA = carcinoembryonic antigen level, TDR =
6
tumor disappearance rate,
31
1
2
Table 4 Multivariate analysis for prediction of non-invasive NSCLC among the clinical
Stage IA patients
Variables
Risk factors
Risk ratio for
invasive tumor
95% CI
p-value*
Tumor size (cm)
>2.0
2.871
1.347-6.119
0.006
Urine DiAcSpm level
(nmol/g creatinine)
high
3.374
2.736-7.361
0.002
Serum CEA (mg/dL)
≥5.0
2.316
0.841-6.377
0.104
<0.75
6.103
1.962-18.98
<0.001
TDR
3
*: Logistic regression analysis, CI = confidence interval, DiAcSpm = diacetylspermine, CEA = serum
4
carcinoembryonic antigen level, TDR = tumor disappearance rate.
5
6
7
8
Supplementary table 1 Relationship between histological invasiveness in
IASLC/ATS/ERS
classification
and
clinicopathological
factors
in adenocarcinoma cases
Factors
AIS or MIA
Invasive adenocarcinoma p-value *
Age (years)
≤69
>69
52
35
26
27
0.225
Gender
male
female
37
50
23
30
0.010
Smoking history
Never-smoker
Smoker
41
46
15
38
0.033
Tumor size (cm)
≤2.0
>2.0
65
22
26
27
<0.001
Serum CEA level (mg/dL)
≤5.0
>5.0
82
5
42
12
0.006
32
Urine DiAcSpm level
(nmol/g creatinine)
Low
High
TDR
≥0.75
<0.75
59
28
11
42
<0.001
33
54
10
43
0.023
1
*: Fisher’s exact test, DiAcSpm = diacetylspermine, CEA = carcinoembryonic antigen level; TDR: tumor
2
disappearance rate; AIS: adenocarcinoma in situ;
3
Supplementary table 2 Multivariate analysis for prediction of invasive adenocarcinoma
among the clinical Stage IA patients
4
Variables
Risk factors
MIA: minimally invasive adenocarcinoma.
Risk ratio for
invasive tumor
95% CI
p-value*
male
1.843
0.683-4.970
0.227
Smoking history
smoker
2.174
0.878-5.405
0.093
Tumor size (cm)
>2.0
3.249
1.380-7.650
0.007
Urine DiAcSpm level
(nmol/g creatinine)
high
8.208
3.470-19.42
<0.001
Serum CEA (mg/dL)
≥5.0
1.143
0.328-3.978
0.834
<0.75
2.783
1.090-7.108
0.032
Gender
TDR
5
*: Logistic regression analysis, CI = confidence interval, DiAcSpm = diacetylspermine, CEA = serum
6
carcinoembryonic antigen level, TDR = tumor disappearance rate.
7
33