Multiple primary malignancies (MPM) are defined by the presence of two or more independent primary malignancies in the same or different organs in an individual patient [1]. One of the earliest systematic studies of MPM was performed by Warren and Gates in 1932 [2]. The development of MPM is not a rare phenomenon. Based on an analysis of several studies, the incidence of MPM was estimated at 0.73–5.2 % in all tumor patients. This wide variation is related to the diverse experiences of doctors and different diagnostic tools used at different hospitals [3–5]. The incidence of MPM has increased dramatically in recent decades [6]. Lung cancer is one of the most commonly diagnosed cancers and causes the highest number of cancer-related deaths [7]. Improvements in diagnostic tools and treatment modalities, including molecularly targeted therapy, have resulted in great advances in lung cancer prognosis. Consequently, patients are surviving long enough to develop subsequent primary malignancies. Although the incidence of MPM has risen in recent decades, research on MPM involving lung cancer remains limited, especially in Chinese patients. This study retrospectively focused on the incidence, clinical features, and prognosis of MPM patients involving lung cancer at the Guangdong Lung Cancer Institute (GLCI).

MPM incidence has increased significantly owing to advances in diagnostic methods and new therapies, such as targeted therapies, which allow more patients to survive long enough to develop subsequent primary tumors [9, 10], screening procedures are useful for the early detection of possible MPM, especially for the patients who are diagnosed in advanced stages [11]. Although there are numerous reports addressing the clinical features of MPM, only a few studies in Taiwan and Japan have investigated MPM involving lung cancer [12, 13]. Unfortunately, these reports have only marginally improved our understanding of the clinical features of patients with this disease. Therefore, this retrospective study was designed to evaluate the clinical characteristics and outcomes of MPM patients involving lung cancer in Chinese patients. The incidence of MPM patients involving lung cancer in our study was 3.4 % (185/5405), similar to the results of a previous study from Turkey, which found an incidence of 3.9 % (40/1038) [14]. The most frequent accompanying malignancies were colon, rectal, and esophageal cancers, which suggests that lung cancer may be closely associated with digestive system neoplasms. The next most frequent accompanying malignancies were thyroid, liver, breast, urinary bladder, and laryngeal cancers, a distribution similar to cancers in the general Chinese population. Rare accompanying malignancies, such as cancers of duodenal, gallbladder, tongue, soft palate, nasal olfactory cell, and osteosarcoma, were also identified. In addition, the observed male predominance, with a sex ratio of 2.5:1 for MPM patients, is consistent with a previous report of male/female ratios ranging from 0.9:1 to 3.5:1 [15]. Moreover, there were 119 non-smokers (68 %) and 56 smokers (32 %), which suggests that unknown factors other than tobacco play an important role in the increased incidence of MPM. It is well established that a high incidence of epidermal growth factor receptor (EGFR) mutations is associated with clinical features, such as adenocarcinoma histology, Asian ethnicity, female sex, and never-smoker status [16–18], and MPM patients with nearly 60 % adenocarcinoma histology and nearly 70 % never-smoker status indicate the high possibility of EGFR mutations, which might explain the increased incidence of MPM patients involving lung cancer. Based on these results, we reviewed the EGFR mutation status of lung cancer specimens, however, we did not have complete EGFR mutation data because our sample set included early records from when EGFR mutation detection was not routine clinical practice. In fact, there were only 84 patients who had EGFR mutation status detected by amplification refractory mutation system-polymerase chain reaction and direct sequencing, and there were only 24 patients with EGFR mutations for an EGFR mutation rate of only 28.6 %. This is similar to the common Asian population, which suggests that EGFR mutations are not responsible for the increased incidence of MPM patients involving lung cancer.

In the univariate analysis, we demonstrated no OS difference between male and female patients, which is different from one reported study showing better OS in female patients [10]. The discrepancy may be the result of the different types of tumors and heterogeneity of the study populations. Our study suggests that time of occurrence and stage are independent factors for OS in MPM patients. Metachronous MPM patients had a better prognosis than synchronous MPM patients. This is consistent with previous studies of MPM involving lung cancer [12] and MPM involving other cancers, such as gastric cancer and hepatocellular carcinoma [19–21], and it is because of detection of the second primary cancer at a curable stage with periodic medical check-ups for patients with a history of cancer. However, there were 70 patients (63 %) in the metachronous group with stage III or IV lung cancer and only 34 patients (53.2 %) in the synchronous group. We believe that the following reasons can explain this contradiction. First, time of occurrence other than stage is the most important determining factor of MPM patient prognosis, which was shown in the multivariate analysis. Second, there were 18 LCF MPM patients and 93 OCF MPM patients in the metachronous group. More OCF MPM patients were observed in the metachronous group, which suggests a better prognosis.

There are several studies focusing on the difference between synchronous and metachronous groups, such as MPM involving lung cancer and MPM involving gastric cancer and hepatocellular carcinoma, however, to the best of our knowledge, only one study has focused on the difference between LCF and OCF groups [12], which did not show a significant difference either calculated time from the diagnosis of the first cancer or from the second cancer. In this study, the prognosis did not show a significant difference between LCF and OCF MPM patients on multivariate analysis, which is consistent with the previous study. Based on the order of occurrence, we split MPM patients to compare the OS between the metachronous and synchronous patients either in LCF MPM or in OCF MPM patients. Our results showed that metachronous MPM patients have better OS than synchronous MPM patients in the OCF MPM group, but it did not show a significant difference in the LCF MPM group, which suggests that lung cancer is more aggressive than other malignancies. In addition, Zeng et al. reported no difference in OS between patients with MPM involving hepatocellular carcinoma (HCC) and patients with HCC alone [22], indicating that extrahepatic primary malignancies had no effect on the survival of HCC patients. For these reasons, MPM patients do not always have a poor prognosis, especially patients with metachronous MPM.

Although the underlying mechanisms of MPM have not been fully elucidated, inherited predisposition is thought to be an important factor [23]. Such predispositions include Lynch syndrome, an autosomal dominant-inherited disorder of colorectal cancer, and susceptibility to other tumors caused by germline mutations in DNA mismatch repair genes [24]. The immune system of patients was suggested to be another important factor [25], and intensive exposure to carcinogens including chemotherapy and/or radiotherapy used in the treatment of tumors [26, 27] and field cancerization in organs exposed to carcinogens, leading to the proliferation of numerous primary tumors, have also been suggested to be responsible for MPM [28, 29].

Recently, “ALKoma” was proposed as a cancer subtype with a shared target as an essential growth driver [30]. More and more investigators suggest that cancers of different organs of origin, but with the same molecular targets, should be managed together because the common molecular targets observed in diverse tumors determine clinical practice better than organ-based classification [31]. Consequently, further genetic and molecular studies using next-generation sequencing explored whether MPM involving lung cancer share common molecular targets, such as EGFR, Kirsten rat sarcoma viral oncogene homolog (KRAS) and anaplastic lymphoma kinase (ALK) to improve therapeutic outcome. Further genetic and molecular investigations will focus on understanding the pathogenesis of MPM involving lung cancer and improving therapeutic outcomes.

There are two new features in the present study that differ from previous reports. First, this study had the largest number of MPM cases involving lung cancer in mainland People’s Republic of China. Second, it is the first study to explore the impact of both time of tumor occurrence (synchronous MPM vs. metachronous MPM) and order of tumor occurrence (LCF MPM vs OCF MPM) on the survival of MPM patients. Our results show that metachronous MPM patients have a better prognosis compared with synchronous MPM patients, 6 months intervals may partially explained why metachronous MPM patients have a better prognosis than synchronous MPM patients.

Although our study revealed some unique results, there are some potential shortcomings and limitations. First, the study was retrospective and was conducted at a single institution. Second, the study did not explore the impact of extrapulmonary primary malignancies on lung cancer survival. Third, some information was deficient because of the long follow-up duration. Nonetheless, the consistency with reported studies confirms the validity of our conclusions.

Colorectal cancer, esophageal cancer, and thyroid cancer were the tumors most frequently accompanying lung cancer. Time of occurrence and stage were independent factors for OS, which suggests that MPM patients presenting with metachronous cancers and early stage have a better prognosis.