Observational Study Medicine ®

OPEN

Clinical analysis and prognosis of synchronous

and metachronous multiple primary malignant
tumors
Meng Lv, PhDa, Xiao Zhang, MDb, Yanwei Shen, PhDa, Fan Wang, PhDa, Jiao Yang, MDa, Biyuan Wang, PhDa,
∗
Zheling Chen, PhDa, Pan Li, PhDa, Xiaoman Zhang, PhDa, Shuting Li, PhDa, Jin Yang, MDa,

Abstract
The aim of this study was to determine the clinical features, treatment factors, and prognosis of patients with multiple primary
malignant tumors (MPMTs). In total, 161 patients with MPMTs at our hospital (The First Affiliated Hospital of Xi’an Jiaotong University,
Xi’an, Shaanxi, China) were analyzed in this study. We found that among 161 patients with MPMTs, 78 (48.4%) patients had
synchronous tumors and 83 (51.6%) patients had metachronous tumors. Most clinical and pathological features were similar in both
groups. Most patients with MPMTs were men and older patients (>50 years old), and adenocarcinoma was the most frequent
pathology type. The most frequent location of all MPMTs was the digestive system. The leading tumor association was between
digestive–digestive tumors, also. However, patients with synchronous tumors and MPMTs of the digestive system showed a shorter
survival time. In the metachronous cancer group, the median interval time was 60 months, and a short interval time (60 months) was
associated with a shorter survival time. In addition, survival time was increased in the younger age group (50 years old) and in
patients who accepted surgery-based comprehensive therapy. However, only interval time (60 months) was an independent
prognostic factor associated with survival for the metachronous cancer group. Therefore, careful surveillance and follow-up are
especially important in these patients.
Abbreviations: GIST = gastrointestinal stromal tumor, MPMT = multiple primary malignant tumor, OS = overall survival,
STS = soft tissue sarcoma.
Keywords: clinical characteristics, metachronous, multiple primary malignant tumors, prognosis, synchronous

1. Introduction patients with cancer to survive long enough to develop multiple

primary malignant tumors (MPMTs).[3]
In the 21st century, cancer has become a serious hazard to human
MPMTs were first described by Billroth[4] in 1889 and
health. It is reported that there were about 4292,000 newly
reported in a detailed study by Warren and Gates[5] in 1932.
diagnosed cancer cases in 2015 in China, corresponding to
Based on criteria proposed by Warren and Gates, diagnosis of
almost 12,000 new cancer diagnoses on average each day.[1] In
MPMTs was dependent on each tumor must have clear evidence
the United States, a total of 1658,370 new cancer cases are
of malignancy on histologic examination, each tumor must be
projected to occur in 2015.[2] However, with the development of
geographically separate and distinct, and the possibility of a
modern diagnostic procedures and chemotherapy/radiation/
metastatic lesion having spread from a prior cancer must be
target therapy, the survival rate is increasing. This allows more
excluded. International rules for MPMTs are more detailed, and
tumors arising in an organ or a pair of organs or a tissue are
usually considered to be 1 tumor. However, there are 2
Editor: Simona Gurzu. exceptions to this rule: systemic cancers potentially involving
Author contributions: ML and XZ have contributed equally to the article. many different organs should only be counted once in any
The authors have no funding and conflicts of interest to disclose. individual, and cancers with different histology should be
a
Department of Oncology, The First Affiliated Hospital of Xian Jiaotong University, regarded as multiple cancers, even if they are diagnosed
Xi’an, b Department of Medical Oncology, Xianyang Center Hospital, Xianyang, simultaneously at the same site.[6,7]
Shaanxi, PR China. The incidence of MPMTs has been reported to range from
∗
Correspondence: Prof Jin Yang, Department of Oncology, The First Affiliated 0.52% to 11.7% in various studies from different countries.[8–10]
Hospital of Xian Jiaotong University, Xi’an, Shaanxi 710061, PR China
In different geographical regions, the incidence, characteristics,
(e-mail: [email protected])
and survival rates associated with MPMTs have been found to be
Copyright © 2017 the Author(s). Published by Wolters Kluwer Health, Inc.
This is an open access article distributed under the Creative Commons
diverse. Further studies are needed to help understand this
Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and disease.
reproduction in any medium, provided the original work is properly cited. In this study, we analyzed the clinical characteristics and
Medicine (2017) 96:17(e6799) prognosis of patients with MPMTs in our hospital in northwest
Received: 4 December 2016 / Received in final form: 5 April 2017 / Accepted: 9 China from January 2008 to February 2015. We also examined
April 2017 the risk factors associated with poor prognosis for patients with
http://dx.doi.org/10.1097/MD.0000000000006799 MPMTs.

1
Lv et al. Medicine (2017) 96:17 Medicine

Table 1
Clinical characteristics.
Variable Synchronous tumors (n [%]) Metachronous tumors (n [%]) Total (n [%]) P
No. of patients 78 (48.4) 83 (51.6) 161
Median age, y 64 57 (for primary tumor)
63 (for second tumor)
Median interval, mo 60
Age at diagnosis of the primary tumor, y 0.058
50 12 (15.4) 24 (28.9) 36 (22.4)
>50 66 (84.6) 59 (71.1) 125 (77.6)
Gender 0.038
Female 24 (30.8) 39 (47.0) 63 (39.1)
Male 54 (69.2) 44 (53.0) 98 (60.9)
Family history 0.80
No 71 (91.0) 74 (89.2) 145 (90)
Yes 7 (9.0) 9 (10.8) 16 (10)

2. Patients and methods these, 161 (1.0%) patients were diagnosed with MPMTs. Of
these 161 patients, 78 (48.4%) had 2 synchronous tumors, and
2.1. Data collection
83 (51.6%) patients had 2 metachronous tumors (Table 1).
A total of 161 patients at our hospital (The First Affiliated In the synchronous cancer group, the median age was 64 years.
Hospital of Xi’an Jiaotong University, Xi’an, Shaanxi, China) In the metachronous cancer group, the median age was 57 years
with a diagnosis of MPMTs between January 2008 and February at the time of diagnosis of the first primary cancer and 63 years at
2015 were reviewed according to the criteria proposed by the time of diagnosis of the second primary cancer. The interval
Warren and Gates.[5] time (the time between the date of diagnosis of the first primary
MPMTs may be defined as synchronous or metachronous cancer and the date of diagnosis of the second primary cancer)
tumors. “Synchronous” tumors refer to cases in which the second was evaluated only for metachronous tumors. The median
primary cancer is diagnosed within 6 months of the primary interval for metachronous cancers was 60 months (range, 7–360
cancer; “metachronous” tumors refer to cases in which the months, Table 1). Our results showed an interval of within 60
second primary cancer is diagnosed more than 6 months after the months for 57.8% (48/83) of patients with metachronous
diagnosis of the first primary cancer. cancers. Breast cancer and urogenital system cancer were the
All pathological results were confirmed by the Pathology most common first primary cancers in patients showing a long
Department of our hospital. Records containing uncertain data, interval time (≥120 months). In both the synchronous and
such as indecisive pathologic reports or absent registration of metachronous cancer groups, most patients were over 50 years
former tumors, were excluded. Six tumors with ≥3 primary old (84.6% and 71.7%, respectively). However, there were more
tumors were also excluded from the study. patients in the metachronous cancer group of less than 50 years of
This study was reviewed and approved by the Human Ethics age than in the synchronous cancer group (28.9% vs 15.4%),
Committee of The First Affiliated Hospital, College of Medicine indicating that patients with metachronous primary cancer were
of Xi’an Jiaotong University. generally younger.
In total, 63 (39.1%) patients with MPMTs were females and
2.2. Treatment and overall survival 98 (60.9%) were males. In both the synchronous and
Patients with MPMTs were treated with surgery, chemotherapy, metachronous cancer groups, men were more frequent, and
radiotherapy, and comprehensive treatment according to clinical there was a statistical difference in the distribution of synchro-
guidelines. Overall survival (OS) was calculated as the number of nous and metachronous cancer cases between gender groups
months between the date of diagnosis and the date of death or the (P = .038; Table 1).
date of the end of the follow-up (January 2016). The percentage of patients who had a family history of cancer
was similar between the synchronous and metachronous cancer
2.3. Statistical analysis groups (9.0% vs 10.8%, Table 1).

For statistical analysis, the SPSS (version 17.0; SPSS Inc.,

Chicago, IL) program was used. Differences between groups 3.2. Pathological features of MPMTs
were evaluated by the chi-square or fisher exact test. Survival Among all of the MPMT cases analyzed, the frequent pathology
probabilities were estimated using the Kaplan–Meier method. types were adenocarcinomas (49.3%), squamous carcinoma
Cox proportional hazard multivariate analysis was performed to (26.1%), hematopoietic and lymphoid tissues (8.1%), transi-
identify independent factors associated with death. All P values tional cell carcinoma (6.2%), and sarcomas and soft tissue
<0.05 were considered statistically significant. tumors (0.9%, Table 2). In addition, other specific carcinomas
(9.3%) in Table 2 include neuroendocrine tumors, gastrointesti-
3. Results nal stromal tumors (GISTs), embryonal carcinoma, malignant
melanoma, seminomas, renal clear cell carcinoma, and so on.
3.1. Clinical features of MPMT patients
Among synchronous tumors specifically, the most frequent
In total, 15,683 patients were diagnosed with malignant tumors pathology groups were adenocarcinomas (55.1%) and squamous
in our hospital between January 2008 and February 2015. Of carcinomas (23.1%). In the first metachronous tumor and the

2
Lv et al. Medicine (2017) 96:17 www.md-journal.com

Table 2
Pathologic characteristics.
Metachronous tumors (n [%])
Histology type Synchronous tumors (n [%]) First tumor Second tumor Total (n [%])
Squamous carcinoma 36 (23.1) 27 (32.5) 21 (25.3) 84 (26.1)
Transitional cell carcinoma 9 (5.8) 4 (4.8) 7 (8.4) 20 (6.2)
Adenocarcinomas 86 (55.1) 35 (42.2) 38 (45.8) 159 (49.3)
Sarcomas and soft tissue tumors 2 (1.3) 0 (0) 1 (1.2) 3 (0.9)
Hematologic and lymphoid malignancy 10 (6.4) 7 (8.4) 9 (10.8) 26 (8.1)
Other specific carcinomas 13 (8.3) 10 (12.0) 7 (8.4) 30 (9.3)
Total 156 (100) 83 (100) 83 (100) 322 (100)

second metachronous tumor groups, the most frequent patholo- (11.8%, Table 3). Specifically, the most frequent tumor
gy types were also adenocarcinomas (42.2% and 45.8%, localization of the synchronous tumors was stomach (17.3%),
respectively) and squamous carcinomas (2.5% and 25.3%, whereas the leading localization of the first and second
respectively, Table 2). Adenocarcinomas were therefore the most metachronous tumor was breast (15.6%) and lung (22.9%),
common pathological type in both the synchronous and respectively (Table 3).
metachronous cancer groups. Among males, tumors were most frequently located in the
digestive (48.5%), respiratory (20.9%), and urogenital (20.9%)
system. Among females, the urogenital system (30.2%), digestive
3.3. Distribution of MPMTs
system (24.6%), and breast tissue (15.9%) were the predominant
Tumors of the digestive system were the most common MPMTs sites for tumor localization (Table 4).
(39.1%), followed by urogenital system (24.5%) and respiratory
system (17.1%) tumors (Table 3). In the synchronous tumor
3.4. The associations between different systems with
group, the top 3 systems were the digestive (48.7%), urogenital
(21.8%), and respiratory (15.4%) systems. In the first meta- regard to MPMTs
chronous tumor group, the most commonly affected area was the In the synchronous tumor group (Fig. 1A), the most common
urogenital system (32.5%), followed by the digestive system associations were observed between the digestive and digestive
(27.7%) and breast tissue (15.6%), whereas in the second system tumors (n = 28), followed by urogenital–urogenital system
metachronous tumor group, the digestive system (32.5%), tumors (n = 11), and digestive–respiratory system tumors (n =
respiratory system (25.3%), and urogenital system (21.7%) 10). In the metachronous tumors group (Fig. 1B), the most
were most commonly affected (Table 3). frequent associations were between the digestive–digestive
The most frequent sites for tumor localization in all cases of system tumors (n = 9), digestive–respiratory system tumors
MPMTs were the stomach (13%), lung (12%), and esophagus (n = 9), and urogenital–urogenital system tumors (n = 9).

Table 3
The distribution of synchronous and metachronous tumors.
Metachronous tumors (n [%])
Location Synchronous tumors (n [%]) First tumor Second tumor Total (n [%])
Digestive systems 76 (48.7) 23 (27.7) 27 (32.5) 126 (39.1)
Lip, oral, cavity 1 (0.6) 0 (0) 3 (3.6) 4 (1.2)
Esophagus 23 (14.7) 10 (12.0) 5 (6.0) 38 (11.8)
Stomach 27 (17.3) 4 (4.8) 11 (13.3) 42 (13.0)
Colorectal, anus 14 (9) 4 (4.8) 4 (4.8) 22 (6.8)
Liver, pancreas, gallbladder 11 (7.1) 5 (6.0) 4 (4.8) 20 (6.2)
Respiratory systems 24 (15.4) 10 (12.0) 21 (25.3) 55 (17.1)
Nasal cavity, larynx 3 (1.9) 8 (9.6) 2 (2.4) 9 (2.8)
Lung, bronchus 21 (13.5) 2 (2.4) 19 (22.9) 39 (12.1)
Hematopoietic and lymphatic system 10 (6.4) 7 (8.4) 9 (10.8) 26 (8.1)
Breast 5 (3.2) 13 (15.6) 2 (2.4) 20 (6.2)
Urogenital systems 34 (21.8) 27 (32.5) 18 (21.7) 79 (24.5)
Cervix uteri 2 (1.3) 8 (9.6) 2 (2.4) 12 (3.7)
Corpus uteri 3 (1.9) 4 (4.8) 2 (2.4) 9 (2.8)
Ovary 2 (1.3) 1 (0) 3 (3.6) 6 (1.9)
Prostate 9 (5.8) 5 (6.0) 2 (2.4) 16 (5.0)
Male genital system 1 (0) 2 (2.4) 4 (4.8) 2 (0.6)
Kidney 8 (5.1) 3 (3.6) 0 (0) 11 (3.4)
Bladder 9 (5.8) 4 (4.8) 5 (6.0) 18 (5.6)
Thyroid 5 (3.2) 1 (1.2) 2 (2.4) 8 (2.5)
Others 2 (2.4) 2 (2.4) 3 (3.6) 7 (2.2)
Total 156 (48.4) 83 (25.8) 83 (25.8) 322 (100)

3
Lv et al. Medicine (2017) 96:17 Medicine

Table 4
The distribution of MPMTs in male and female patients.
Male (n [%]) Female (n [%])
Location First Second Total First Second Total
Digestive system 49 (50.0) 46 (46.9) 95 (48.5) 15 (23.8) 16 (25.4) 31 (24.6)
Respiratory system 17 (17.3) 24 (24.5) 41 (20.9) 2 (3.2) 11 (17.5) 13 (10.3)
Hematologic and lymphatic system 6 (6.1) 8 (8.2) 14 (7.1) 7 (11.1) 5 (7.9) 12 (9.5)
Breast 0 (0) 0 (0) 0 (0) 16 (25.4) 4 (6.3) 20 (15.9)
Urogenital system 24 (24.5) 17 (17.3) 41 (20.9) 19 (30.2) 19 (30.2) 38 (30.2)
Others 2 (2.0) 3 (3.1) 5 (2.6) 4 (6.3) 8 (12.7) 12 (9.5)
Total 98 (100) 98 (100) 196 (100) 63 (100) 63 (100) 126 (100)

More specifically, in the synchronous tumor group, the most accepted the surgery therapy including the surgery-alone and
common association was found between esophageal cancer and surgery-based comprehensive therapy after the synchronous tumor
gastric cancer (n = 16). This was followed by associations was diagnosed, only 9 (11.5%) patients did not accept any therapy.
between gastric and lung cancers (n = 5), and bladder and 3.6. Survival outcome
prostate cancers (n = 5). In the metachronous tumor group,
esophageal and lung cancers (n = 5) held the first place, and Among the 161 patients, 138 were followed up with a median
esophageal and gastric cancers (n = 4) held the second place. follow-up period of 36 months from diagnosis of the first primary
tumor, and 78 patients died. The median survival time in the
synchronous group was 12 months (range 1–120 months). The
3.5. Treatment of MPMTs median survival time of the metachronous group was 96 months
Among all of the MPMT patients, 148 (92%) patients accepted (range 12–408 months) from diagnosis of the first primary cancer
therapy after the first primary cancer was diagnosed, and 145 and 12 months (range 2–96 months) from diagnosis of the second
(89.5%) patients accepted therapy after the second primary cancer primary cancer. In summary, the patients with synchronous
was diagnosed, including surgery, chemotherapy, and radiothera- tumors showed a shorter survival time than the patients with
py. In the metachronous tumor group, 24 (28.9%) patients metachronous tumors (P < 10 3) from diagnosis of the first
accepted therapy in terms of surgery alone, 23 (27.7%) patients primary cancer (Fig. 2A). In addition, for both first primary and
accepted surgery-based comprehensive therapy (surgery combined second primary tumors, the patients with MPMTs of the digestive
with chemotherapy or radiotherapy), and 34 patients (41%) system showed a shorter survival time than patients with MPMTs
accepted chemotherapy or radiation therapy alone after the first of other systems (Fig. 2B and C).
primary cancer was diagnosed. In addition, 23 (27.8%) patients In the synchronous group, the 1-year survival rate was 56.9%,
accepted surgery-based therapy and 39 (47%) patients did not and the 3-year survival rate was 20.9%. In the metachronous
accept surgery therapy after the second primary cancer was group, the survival rates differed between first and second
diagnosed. In the synchronous tumor group, 39 (50%) patients primary cancer diagnoses; the 1-year survival rates were 95.7%
and 58.0%, respectively, and the 3-year survival rates were
83.7% and 41.2%, respectively.
In the metachronous group, the survival time of younger
people (50 years old) was longer than that of older people (>50
years old), and a short interval time (60 months) was associated
with a shorter survival time (Fig. 2D and E). With regard to
treatment, the OS time of patients differed significantly between
the comprehensive therapy group (including surgery-based
therapy combined with chemotherapy or radiotherapy) and
the surgical therapy-alone group (P = .048, Fig. 2F). However, in
cases of second primary cancer, there was no statistical difference
in OS between the surgery-based therapy and the no surgical
therapy groups (P = .315, Fig. 2G).
In addition, a short interval time (60 months) was found to be
an independent poor prognostic factor of survival in the
metachronous cancer group based on Cox regression hazard
model (Table 5). Specifically, a short interval (60 months) was
associated with a shorter survival time. However, in the
synchronous group, there were no detectable differences
regarding age, gender, or treatment types based on either
Figure 1. Associations among synchronous and metachronous tumors. (A) Kaplan–Meier or Cox proportional hazard multivariate analyses.
Number of patients with synchronous tumors derived from the relative
systems. (B) Number of patients with metachronous tumors derived from the
relative systems. 1—Digestive system, 2—respiratory system, 3—hemato- 4. Discussion
poietic and lymphoid system, 4—breast tissue, 5—urogenital system, and 6—
others. MPMT is a special phenomenon in tumorigenesis that is
beginning to become better understood due to a number of

4
Lv et al. Medicine (2017) 96:17 www.md-journal.com

Figure 2. Survival times associated with different multiple primary malignant tumor groups. (A) The survival times for the synchronous or metachronous tumor
groups from diagnosis of the first primary cancer. Synch: synchronous tumors, Metach: metachronous tumors. (B) The survival times for patients with digestive
system tumors and tumors in other systems from diagnosis of the first primary cancer. (C) The survival times for the digestive system tumor and other systems tumor
groups from diagnosis of the second primary cancer. Digestive: digestive system tumor group. Other: other systems tumor group, including the respiratory,
hematopoietic, lymphoid and urogenital system, and breast. (D) The survival times of younger people (50 years old) and older people (>50 years old) from
diagnosis of the first primary cancer in the metachronous tumor group. (E) The survival times for different interval times (60 vs >60 months) for the metachronous
tumor group. (F) The survival times for different therapies for the metachronous tumor group from diagnosis of the first primary cancer. S-based: surgery-based
comprehensive therapy, S-alone: surgical therapy alone. (G) The survival times for different therapies in the metachronous tumor group from diagnosis of the
second primary cancer. S: surgical therapy; NS: no surgery.

studies worldwide. Studies have reported that the incidence of MPMTs can occur at any age. However, earlier studies
MPMTs was 0.52% to 3.66% in China compared with 0.73% to reported that patients with MPMTs tend to be older than those
11.7% in other countries.[8–10] In our hospital in Shaanxi, China, with a single primary cancer, and in most reports, more than 75%
the incidence of MPMTs was 1.0%, which is similar to the rates of patients with MPMTs were more than 50 years of age.[13–15]
of prevalence previously reported in China but less than those Our results were consistent with these earlier studies, with 77.6%
reported in other countries. Many factors may contribute to the
diverse incidence of MPMTs in different geographical regions, Table 5
such as genetic factors, environmental factors, diagnostic Analysis of prognostic factors contributing to mortality of
methods, and follow-up information. metachronous MPMTs.
The prevalence of synchronous MPMTs has differed between
Variable Hazard Ratio (95% CI) P
previous studies (range 30%–55%).[3,11,12] In our study, 48.4% of
patients had synchronous type and 52.6% of the patients had Age (>50 years old vs 50 years old) 1.850 (0.422–8.109) .415
metachronous type tumors. This might be a result of the different Gender (male vs female) 1.142 (0.410–3.185) .799
Interval (>60 months vs 60 months) 0.132 (0.038–0.454) .001
population characteristics, different diagnostic tools, and different
Treatment for the first primary cancer 1.099 (0.302–3.997) .886
rules of cancer registry between hospitals. In our study, the (surgery alone vs surgery-based
prevalence of synchronous and metachronous MPMTs was similar, comprehensive therapy)
emphasizing the importance of an accurate synchronous diagnosis.

5
Lv et al. Medicine (2017) 96:17 Medicine

of patients being over 50 years old. However, the median age of breast carcinoma and genitourinary malignancies. Physicians
diagnosis of the first primary cancer was lower in the should be aware of patients with primary malignant fibrous
metachronous group than in the synchronous group, indicating histiocytoma who demonstrate a risk for developing a renal cell
that young people are more likely to have metachronous tumors carcinoma.[30–32] Of the GIST patients, 10% to 20% developed
and should be closely monitored for a longer time. MPMTs. The most common association with GIST was the
As reported by some previous studies,[16,17] our results showed stomach, the prostate, the breast, the esophagus, and the kidney
a short interval time (60 months) was associated with a shorter cancer. The report indicated that the patients with GIST with 2 or
survival time. This may indicate that the first primary cancer had more other cancers had a poor prognosis.[33,34] So these patients
been cured over a longer interval time. The interval for more than with GIST and other cancers should be noted. In addition to the
50% of patients was within 60 months, indicating that screening frequent MPMTs, there are rare cases, such as giant cavernous
for second primary cancers should be performed within 5 years hepatic hemangioma with endometrial adenocarcinoma, syn-
emphatically. However, our results showed that among the 22 chronous adrenocortical carcinoma with GIST, synchronous
patients with an interval time of more than 120 months, the first jejunal carcinoid tumor with colorectal polyps, and bilateral
primary cancers of 12 of these patients were breast and urogenital synchronous sporadic renal cell carcinoma.[35–38] Because of the
system cancers. This may reflect the long survival time of breast rarity of the incidence of these MPMTs, we need more data to
and urogenital system cancers, and these patients should explore their association.
therefore be monitored beyond 60 months. The high risk for MPMT patients who did not have a family
Consistent with other reports,[18,19] men were more frequent history of malignancy might be in part due to the environmental
than women among both the synchronous and metachronous factors and/or genetic factors that are shared between the 2
groups in our study. In addition, more than 60% of patients with malignancies. The specific mechanism of MPMTs is not unclear until
MPMTs exhibited primary tumors of the digestive and now. Recently, some studies reported that many genes, including
respiratory systems in men. The main reasons for the high BRCA2, ATM, POLD1, PABL2, SMAD4, and so on, played an
incidence of tumors of the digestive and respiratory systems were important role in the occurrence and pathogenesis of MPMTs.[39–41]
tobacco and alcohol in men.[20,21] This indicated that these 2 sites Specifically, the germline mutations of BRCA1/BRCA2 were
should be closely monitored by screening, especially in men. Men associated with increased risk of breast, ovarian, stomach, color-
should also be advised of the risk between smoking and alcohol ectum, uterus, and pancreas cancers. ATM truncations were also
consumption and developing MPMTs or a single primary cancer. detected in many cancer types, mostly in lung, stomach, and prostate
In our study, the most frequent sites of localization and cancers.[39]PABL2 gene variation was associated with increased risk
pathology of tumors were the digestive system and adenocarci- of ovarian and stomach carcinoma.[39] The POLD1 mutation was
nomas, respectively. Within the digestive system, the stomach and also associated with colorectal cancer and endometrial cancer
the esophagus were the most frequent sites of tumor localization. predisposition.[41] More and more studies reported the common
This was not in agreement with previous reports.[3,9,22–25] We gene variations in different types of cancers.
think that the main reason for this difference may be regional In addition to the gene list, significant associations have been
differences. In developed countries, the most frequent site of previously noted between the microsatellite instability (MSI)
tumor localization was the colorectum.[22] In Japan, the incidence phenotype and multiple primary malignancies. Genetic instability
of gastric cancer with colorectal cancer was reported to be may play an important role in the development of second primary
high.[23] In Turkey, the most frequent site of MPMT localization tumors. Therefore, testing for MSI in the primary cancer might
was the skin.[3] In Guang Dong province, China, nasopharyngeal help detect those patients who are at high risk for developing
carcinoma was the most frequent reported for MPMTs.[9] In double primary malignancies.[42–44]
Beijing, the most frequent site for MPMTs was breast tissue.[24] The high risk of MPMTs is also associated with the ways and
However, in Shaan Xi province, the most frequent site of tumor effects of treatment.[45] In the synchronous tumor group, 50%
localization was the upper gastrointestinal tract, including gastric patients accepted the surgery therapy after the synchronous tumor
and esophageal cancers.[25] This demonstrated that in less- was diagnosed. But the treatment strategies for synchronous and
developed regions, MPMTs were focused predominantly within single tumors are different. With the example of colorectal cancer,
the upper gastrointestinal tract of patients, likely related to some authors have suggested that total or subtotal colectomy
economic factors such as poor eating habits. should be performed.[46,47] Passman et al[48] recommended a more
When analyzed in more detail, MPMTs were most frequently extensive resection for lesions in adjacent segments. Lee et al[49]
associated with the stomach and the esophagus (synchronous suggested that 2 regional resections are preferable through the
group, n = 16; metachronous group, n = 4) in our study. There are comparison between the 2 regional resections and extensive
a number of possible explanations for this. First, studies have resection approaches. Therefore, there has been little agreement
reported many similar genetic changes between gastric and among surgeons regarding the appropriate surgical treatment for
esophageal cancers. Hence, these 2 types of cancer potentially synchronous cancers located in separate segments. This need more
share similar molecular mechanisms for pathogenesis.[26,27] study to answer the question.
Second, the stomach and the esophagus are both part of the Moreover, in our study, the patients who accepted surgery-
digestive system and would therefore potentially be exposed to based comprehensive therapy (surgery combined with chemo-
the same pathogenic factors.[28,29] Third, gastric and esophageal therapy or radiotherapy) had a longer survival time than the
cancers can be detected simultaneously by endoscopy. In less- patients who accepted surgery alone. These results indicate that
developed regions, it may therefore be prudent to focus on the doctors should carefully design treatment strategies to include
detection of gastric and esophageal cancers by endoscopy at chemotherapy or radiation therapy according to current guide-
follow-up. lines. In addition, there was no statistically significant difference
Some studies reported about the MPMTs with soft tissue in the OS time from diagnosis of a second primary cancer
sarcoma (STS) and GIST. Of STS patients, 6% to 8% developed between the surgery-based therapy and the no surgery groups
MPMTs, which were most frequently associated with STS and after the second primary group was diagnosed. Therefore,

6
Lv et al. Medicine (2017) 96:17 www.md-journal.com

doctors should perform a careful preoperative evaluation to [9] Li W, Zhan Y, Li G. Double cancers: a clinical analysis of 156 cases.
Zhonghua Zhong Liu Za Zhi 1996;18:296–8.
determine whether there is a need for surgery.
[10] Demandante CG, Troyer DA, Miles TP. Multiple primary malignant
Our findings also showed that patients with synchronous neoplasms: case report and a comprehensive review of the literature. Am
tumors displayed shorter survival times than patients with J Clin Oncol 2003;26:79–83.
metachronous tumors (P  .000) from diagnosis of the first [11] Aydiner A, Karadeniz A, Uygun K, et al. Multiple primary neoplasms at a
primary cancer. In all MPMT cases, either for the first primary or single institution: differences between synchronous and metachronous
neoplasms. Am J Clin Oncol 2000;23:364–70.
the second primary tumors, patients with MPMTs of the digestive [12] Liu Z, Liu C, Guo W, et al. Clinical analysis of 152 cases of multiple
system showed shorter survival times than patients with MPMTs primary malignant tumors in 15,398 patients with malignant tumors.
of other systems. This difference may be due to the characteristics PLoS One 2015;10:e0125754.
of metachronous and digestive system tumors. In addition, in the [13] Hajdu SI, Hajdu EO. Multiple primary malignant tumors. J Am Geriatr
Soc 1968;16:16–26.
metachronous group, the survival rate of younger people (50
[14] Berge T, Cederqvist L, Schonebeck J. Multiple primary malignant
years old) was higher than that of older people (>50 years old). tumours. An autopsy study of a circumscribed population. Acta Pathol
The reasons for this may include the following: first, older Microbiol Scand 1969;76:171–83.
patients have an increased incidence of cardiac and cerebral [15] Haddow AJ, Boyd JF. Multiple primary neoplasms in the Western
vascular problems; second, older patients have had more time to Hospital Region, Scotland: a survey based on cancer registration data.
Scott Med J 1972;17:143–52.
potentially accumulate pathogenic genes; third, the life expec- [16] Enblad P, Adami HO, Glimelius B, et al. The risk of subsequent primary
tancy of young people is longer; and lastly, treatment strategies in malignant diseases after cancers of the colon and rectum. A nationwide
different ages of patients are different. cohort study. Cancer 1990;65:2091–100.
However, our study has some limitations. Comprehensive [17] Heron DE, Komarnicky LT, Hyslop T, et al. Bilateral breast carcinoma:
risk factors and outcomes for patients with synchronous and
analysis of MPMTs requires the observation of a large, well
metachronous disease. Cancer 2000;88:2739–50.
defined population for ≥10 years. We studied only partial data [18] Kaneko S, Yamaguchi N. Epidemiological analysis of site relationships of
from different departments within a single hospital, and the synchronous and metachronous multiple primary cancers in the National
median follow-up period was only 36 months. Furthermore, we Cancer Center, Japan, 1962–1996. Jpn J Clin Oncol 1999;29:96–105.
did not consider the staging of tumors, environmental factors [19] Kagei K, Hosokawa M, Shirato H, et al. Efficacy of intense screening and
treatment for synchronous second primary cancers in patients with
such as smoking and diet, radiation fields, or other complications esophageal cancer. Jpn J Clin Oncol 2002;32:120–7.
such as cardiac and cerebral vascular incidents. [20] Moy KA, Fan Y, Wang R, et al. Alcohol and tobacco use in relation to
gastric cancer: a prospective study of men in Shanghai, China. Cancer
5. Conclusion Epidemiol Biomarkers Prev 2010;19:2287–97.
[21] He Y, Jiang B, Li LS, et al. Changes in smoking behavior and subsequent
Our study showed that patients with digestive, urogenital, and mortality risk during a 35-year follow-up of a cohort in Xi’an, China. Am
respiratory tumors were more likely to develop MPMTs. In J Epidemiol 2014;179:1060–70.
[22] Moertel CG, Bargen JA, Dockerty MB. Multiple carcinomas of the large
particular, older people (>50 years old) and men were high-risk intestine: a review of the literature and a study of 261 cases.
populations. Short interval times (60 months) were associated Gastroenterology 1958;34:85–98.
with a poor prognosis for patients with metachronous cancer. [23] Kato T, Suzuki K, Muto Y, et al. Multiple primary malignancies
Therefore, careful surveillance and follow-up are especially involving primary sporadic colorectal cancer in Japan: incidence of
gastric cancer with colorectal cancer patients may be higher than
important in these patients.
previously recognized. World J Surg Oncol 2015;13:23.
[24] Hao C, Wang Y, Huang XF. Clinical analysis of multiple primary
Acknowledgments cancers. J Clin Oncol 1999;26:505–10.
[25] Wang AP, Yu ZX. An analysis of 15903 inpatients with cancer in Shanxi
Statistical analysis involved in this article was reviewed by Province. Zhonghua Zhong Liu Za Zhi 2006;15:307–8.
statistician Min Yi, Department of Breast Surgical Oncology, The [26] Lv XD, Wang J, Wang LD. Expressions of mdm2, Bcl-2, bax and p53 in
University of Texas MD Anderson Cancer Center, Houston, TX. tissues from patients with esophageal carcinoma and carcinoma of
gastric cardia concurrently. Chin J Cancer Prev Treat 2006;13:1285–8.
The authors thank the patients and family members for their [27] Chen H, Wang LD, Guo M, et al. Alterations of p53 and PCNA in cancer
participation in this study. and adjacent tissues from concurrent carcinomas of the esophagus and
gastric cardia in the same patient in Linzhou, a high incidence area for
esophageal cancer in northern China. World J Gastroenterol
References 2003;9:16–21.
[1] Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA [28] Moertel CG, Dockerty MB, Baggenstoss AH. Multiple primary
Cancer J Clin 2016;66:115–32. malignant neoplasms. II. Tumors of different tissues or organs. Cancer
[2] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 1961;14:231–7.
2015;65:5–29. [29] Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral
[3] Kilciksiz S, Gokce T, Baloglu A, et al. Characteristics of synchronous- stratified squamous epithelium; clinical implications of multicentric
and metachronous-type multiple primary neoplasms: a study of hospital- origin. Cancer 1953;6:963–8.
based cancer registry in Turkey. Clin Genitourin Cancer 2007;5:438–45. [30] Merimsky O, Kollender Y, Issakov J, et al. Multiple primary
[4] Billroth T, Reimer G. Die allgemeine chirurgische pathologie und malignancies in association with soft tissue sarcomas. Cancer
therapie. 51 Vorlesungen-Ein Handbuch fur Studierende und Artze, 14 2001;91:1363–71.
Auflage, Berlin:1889;908. [31] Kozawa E, Sugiura H, Tsukushi S, et al. Multiple primary malignancies
[5] Warren S, Gates O. Multiple primary malignant tumors: a survey of the in elderly patients with high-grade soft tissue sarcoma. Int J Clin Oncol
literature and a statistical study. Am J Cancer 1932;16:1358–414. 2014;19:384–90.
[6] Hayat MJ, Howlader N, Reichman ME, et al. Cancer statistics, trends, [32] Maretty-Nielsen K. Prognostic factors in soft tissue sarcoma. Dan Med J
and multiple primary cancer analyses from the Surveillance, Epidemiol- 2014;61:B4957.
ogy, and End Results (SEER) program. Oncologist 2007;12:20–37. [33] Pandurengan RK, Dumont AG, Araujo DM, et al. Survival of patients
[7] Multiple Primary and Histology Coding Rules. Available at: http://seer. with multiple primary malignancies: a study of 783 patients with
cancer.gov/tools/mphrules/. Accessed April 9, 2015. gastrointestinal stromal tumor. Ann Oncol 2010;21:2107–11.
[8] Liu FS, Qin DX, Wang QL. The clinical pathology analysis of 172 cases [34] Sevinc A, Seker M, Bilici A, et al. Co-existence of gastrointestinal stromal
of multiple primary malignant tumors. Zhonghua Zhong Liu Za Zhi tumors with other primary neoplasms. Hepatogastroenterology
1979;1:113–13. 2011;58:824–30.

7
Lv et al. Medicine (2017) 96:17 Medicine

[35] Bara TJr, Gurzu S, Jung I, et al. Giant cavernous hepatic hemangioma [42] Kim HS, Cho NB, Yoo JH, et al. Microsatellite instability in double
diagnosed incidentally in a perimenopausal obese female with endometrial primary cancers of the colorectum and stomach. Mod Pathol
adenocarcinoma: a case report. Anticancer Res 2016;36:769–72. 2001;14:543–8.
[36] Kovecsi A, Jung , Bara T, et al. First case report of a sporadic [43] Yun HR, Yi LJ, Cho YK, et al. Double primary malignancy in colorectal
adrenocortical carcinoma with gastric metastasis and a synchronous cancer patients: MSI is the useful marker for predicting double primary
gastrointestinal stromal tumor of the stomach. Medicine (Baltimore) tumors. Int J Colorectal Dis 2009;24:369–75.
2015;94:1549. [44] Ohtani H, Yashiro M, Onoda N, et al. Synchronous multiple primary
[37] Gurzu S, Bara TJr, Bara T, et al. Synchronous intestinal tumors: gastrointestinal cancer exhibits frequent microsatellite instability. Int J
aggressive jejunal carcinoid and sigmoid malignant polyp. Rom J Cancer 2000;86:678–83.
Morphol Embryol 2012;53:193–6. [45] Hemminki K, Boffetta P. Multiple primary cancers as clues to
[38] Hu XY, Xu L, Guo JM, et al. Surgical strategy of bilateral synchronous environmental and heritable causes of cancer and mechanisms of
sporadic renal cell carcinoma-experience of a Chinese university hospital. carcinogenesis. IARC Sci Publ 2004;157:289–97.
World J Surg Oncol 2017;15:53. [46] Enker WE. Multiple carcinomas of the large bowel: a natural experiment
[39] Lu C, Xie M, Wendl MC, et al. Patterns and functional implications of in etiology and pathogenesis. Proc Inst Med Chic 1977;31:178–9.
rare germline variants across 12 cancer types. Nat Commun [47] Wang HZ, Huang XF, Wang Y, et al. Clinical features, diagnosis,
2015;6:10086. treatment and prognosis of multiple primary colorectal carcinoma.
[40] Susswein LR, Marshall ML. Pathogenic and likely pathogenic variant World J Gastroenterol 2004;10:2136–9.
prevalence among the first 10,000 patients referred for next-generation [48] Passman MA, Pommier RF, Vetto JT. Synchronous colon primaries have
cancer panel testing. Genet Med 2016;18:823–32. the same prognosis as solitary colon cancers. Dis Colon Rectum
[41] Palles C, Cazier JB, Howarth KM, et al. Germline mutations affecting the 1996;39:329–34.
proofreading domains of POLE and POLD1 predispose to colorectal [49] Lee BC, Yu CS, Kim J, et al. Clinicopathological features and surgical options
adenomas and carcinomas. Nat Genet 2013;45:136–44. for synchronous colorectal cancer. Medicine (Baltimore) 2017;96:6224.