Background

Glioblastoma complicated with other non-CNS neoplasms is a challenging clinical problem, and the managing clinical risks of which has not been fully explored. Several case series observed an increasing part of GBM patients have been diagnosed with non-CNS tumors previously, which were classified as multiple primary malignant neoplasm (MPMN) [1]. Previous studies showed that the patients survived from cancer are at increased risk of developing the second or even the third primary tumors [2, 3]. On the other hand, it remains inconclusive whether other neoplasms affect the prognosis of patients with primary isocitrate dehydrogenase (IDH) wild-type glioblastoma. Considering such an inconclusive factor may influence the accuracy of clinical trials and cohort studies of gliomas, and it is important to uncover the clinical and pathological characteristics of these patients. In this study, we retrospectively investigated the clinical and molecular pathological characteristics of these patients based on the glioblastoma patient cohort from Beijing Tiantan Hospital and the Chinese Glioma Genome Atlas (CGGA).

Methods

Patients

Clinical data of 117 patients who were diagnosed with primary glioblastoma plus non-CNS tumors were collected from Beijing Tiantan Hospital from January 2019 to February 2022. According to the fifth edition of the WHO classification of tumors of the central nervous system, 45 patients who were pathologically diagnosed with IDH-wild-type were finally enrolled in this study [4]. And the data of patients diagnosed with GBM only was collected from CGGA 325 database as the matched group.

Collection of data

The specific data included age at diagnosis of GBM, gender, date of receiving surgical operation of GBM, date of death or the last following-up, O6-methylguanine-DNA methyltransferase (MGMT) promoter status, telomerase reverse transcriptase (TERT) promoter status, and postoperative radiotherapy and chemotherapy status. The following-up ended on April 9, 2022. For comparison, the paralleled data of the patients who were diagnosed with GBM only between January 2019 and February 2022 was collected from the CGGA database. The primary endpoint was OS, defined as the time interval between the day of surgical operation for GBM and the patient’s death or last following-up.

Statistical analysis

All statistical analyses were conducted with IBM SPSS 25. For normally distributed data or non-normally distributed data, it will be expressed as the mean ± SD or median, respectively. For the two groups, Pearson’s chi-squared (χ2) test was employed to analyze the categorical variables. And student t test and Mann–Whitney U test were utilized to evaluate the continuous variables. Survival analyses were performed through the Kaplan–Meier method and the differences in survival rates between the two groups were compared by using the log-rank test. Univariate COX proportional hazard regression model analyses were applied to determine the factors that affect the OS. A P value < 0.05 was considered statistically significant.

Results

General characteristics of patients

For the group of GBM patients plus primary tumors of other sites, 45 cases were finally enrolled. And for the group of patients diagnosed with GBM only, 112 cases that met the criteria of selection were picked (Fig. 1).

Fig. 1
figure 1

Flowchart of the enrollment of the glioma patients in the two groups. GBM, glioblastoma

Full size image

Distribution of non-CNS tumors in the patients

Based on our study, it seemed that female patients are more likely to develop multiple primary tumors (n = 35, 77.8%). And the most common non-CNS primary tumor was the hysteromyoma of the female patients (n = 17, 37.8%). Meanwhile, the second common no-CNS tumor was the breast tumor (n = 8, 17.8%), including breast cancer (n = 5) and breast adenoma (n = 3). For the male patients, urinary system tumors were the most common (n = 4, 8.9%), including bladder tumors (n = 3) and clear cell renal carcinoma (n = 1). When classified according to the system, reproductive system tumors of the female were the most part in our study (n = 30, 66.7%), which was comprised of hysteromyoma (n = 17), breast tumors (n = 8), endometrial carcinoma (n = 3), and ovarian tumors (n = 2). Urinary system tumors were the second most common (n = 6, 13.3%), which were composed of renal tumors (n = 3) and bladder tumors (n = 3). And the incidence of the digestive system tumors was the next (n = 4, 8.9%), including colorectal cancer (n = 2), gastric cancer (n = 1), and ampulla carcinoma (n = 1). The rest multiple primary tumors included thyroid cancer (n = 3), based cell carcinoma (n = 2), lumbar tumor (n = 1), and nasal lymphosarcoma (n = 1). Moreover, there were three GBM patients diagnosed with two non-CNS primary tumors.

Comparison of clinical characteristics and survival trends between two groups

As exhibited in Table 1, the group of GBM patients with non-CNS tumors got an older age at diagnosis for GBM when compared to the group of patients with GBM only (median age 56.00 compared to 53.50, p = 0.002). And for the aspect of gender, it seems that the female patients diagnosed with GBM tend to have more possibilities to develop multiple primary tumors than the male (77.80% of female patients in the group of GBM plus non-CNS tumors compared to 36.60% of female patients in the group of GBM only, p < 0.0001). For the mutation status of the TERT promoter, the group of GBM patients with non-CNS tumors consisted of a higher proportion of the status of TERT promoter mutation than the other group (68.40% TERT promoter mutation status in the former group compared to 47.00% in the latter group, p = 0.034). However, there was no significant difference in the status of MGMT promoter methylation status between the two groups (50% MGMT promoter methylation in the 34 patients from the group of GBM plus non-CNS tumors compared to 38.2% in the 68 patients from the group of GBM only, p = 0.257). More importantly, no statistically significant difference in OS was observed between the two groups (79.47 of mean rank for the group of GBM plus non-CNS tumors compared to 69.33 for the other group, p = 0.177).

Table 1 Clinical and pathological characteristics of the enrolled GBM patients at baseline
Full size table

For the analyses of survival, there was no significant difference in the comparison of OS between the group of GBM patients with non-CNS tumors and patients with GBM only (median OS 27.8 months for the former group compared to 23.9 months for the latter, p = 0.701) (Fig. 2). Considering that the primary non-CNS tumors include the benign or malignant, which could be an important factor affecting the prognosis, we conducted a further comparison on the OS in the group of GBM patients with non-CNS tumors. And it was observed that there was significant difference in OS between the two groups (median OS 32.06 months for GBM patients with primary benign tumors and 22.83 months for GBM patients with primary malignant tumors, p = 0.026) (Fig. 3).

Fig. 2
figure 2

Kaplan–Meier estimates of overall survival (OS) time for patients with IDH-wild-type GBM only and patients with IDH-wild-type GBM plus other tumors. Note: In the group of IDH-wild-type GBM patients, the overall survival data of 13 patients are not available. Hence, there are 99 patients left in the end. Note: GBM glioblastoma

Full size image
Fig. 3
figure 3

Kaplan–Meier estimates of overall survival (OS) time for IDH-wild-type GBM patients with benign tumors and malignant tumors. Note: For this group of 45 GBM patients with other tumors, it is hard to tell the malignancy of the complicated tumor of one patient according to the medical record. Therefore, there are 44 patients in all left in this group. Note: GBM glioblastoma

Full size image

Univariable analysis of prognostic factors

Univariable Cox proportional hazard regression analysis revealed that the status of postoperative adjuvant therapy was the only factor affecting the survival time of patients (p < 0.001). Furthermore, receiving the chemotherapy and receiving both chemotherapy and radiotherapy were the strongest prognostic factors for the OS of patients (p < 0.001), while receiving the radiotherapy alone meant little (Table 2).

Table 2 Univariate COX proportional hazard regression model analysis of factors associated with the survival of IDH-wild-type GBM patients
Full size table

Discussion

The first research concerning the multiple primary tumors could trace back to 1921, reporting the incidence of 4.7% in 3000 patients diagnosed with malignant tumors [5]. Continued studies were conducted to find out the possible connections and impact [6, 7]. And the results tend to be similar and limited in various cancers [8, 9]. Almost no significant difference was observed when the survival time of patients with multiple tumors was compared to the patients with a single tumor in the past studies [10,11,12].

In recent years, with the great progress of diagnosis and treatment for tumors, the overall survival of cancer patients has been positively changed, which, however, led to the increased number of patients with multiple primary tumors [7, 13]. A similar phenomenon in GBM patients attracted our attention that usually led to a confusing situation for doctors to develop management plans for them or for researchers to recruit them into clinical trials. Meanwhile, there were few studies on GBM patients with multiple primary tumors in the past 5 years. And the fifth WHO guideline for the classification of CNS tumors identified the GBM as the IDH-wild-type, revolutionizing the understanding and clinical practice. It is worth conducting the study under this new situation.

In the beginning, we collect the data of all patients diagnosed with gliomas and other primary tumors, trying to figure out the difference of clinical and pathological characteristics and the prognosis compared to the control group which consisted of patients with gliomas only. From January 2019 to February 2022, 117 glioma patients were diagnosed with multiple primary tumors. However, considering the relatively better prognosis of patients with WHO II and WHO III gliomas, we finally determined to pick the 45 patients who were diagnosed with IDH-wildtype GBM to conduct the analyses.

We chose the GBM patients diagnosed with IDH-wild-type in the past 3 years, trying to reveal the clinical outcome and characteristics of the GBM patients with multiple primary tumors in the current situation to provide some possible guide for clinical practice. It was observed that the patients with multiple tumors usually got older age when they were diagnosed with GBM. And female patients who were frequently diagnosed with tumors of the reproductive system accounted for a large part in the special group based on our data. The distribution of non-CNS tumors was different from the past studies [14]. Since TERT promoter status and MGMT promoter status had been identified as the factors that strongly influence the prognosis of GBM patients, we choose them as the target to conduct the analysis of pathological characteristics [15,16,17]. Patients with multiple tumors tend to get the mutated TERT promoter, which implies that they may get a worse clinical prognosis [18]. And there was no significant difference in the MGMT promoter status between the two groups. For the last part, the postoperative adjuvant therapy was the only factor that affects the prognosis which was consistent with relevant clinical studies [19, 20].

For the analysis of overall survival, there was no significant difference between the two groups, which was consistent with the past studies [11, 14]. And the further analysis of OS between the GBM patients with benign tumors and GBM patients with malignant tumors showed significant difference, which meant that patients with benign tumors tend to have a longer median OS. According to our following-up results, GBM recurrence or progression is the main cause of death for the patients with primary non-CNS tumors. The malignant tumors diagnosed before, we presume, may have made an impact on the patients’ survival status.

For the limitation of our study, the lower number of patients may negatively and largely affect our analysis. And the number is possibly lower than the real due to the negligence in the process of history taking and recording. Meanwhile, limited data on molecular pathology could not support us to do further analysis on the possible connection between the multiple tumors. In addition, based on the history recording, the information about the complicated tumors is limited. Further exploration could not be conducted. How much of a role dose the complicated tumor play in the clinical outcome remains a question to this study.

To sum up, it is unexpected to find out that there was no significant difference in OS between the two groups. Based on our results, patients with malignant tumors before got a poorer survival outcome probably owing to the damage of the previous tumors or the relatively conservative treatment for them that could drive the progression or recurrence of GBM. Further studies including more patients and data on molecular pathology are urgently required to be conducted to better understand the clinical and pathological characteristics of this special group to provide more convincing guidance for clinical practice and trials. What is more, that may be able to discover some innate mechanism of multiple primary tumors and provide a support for the treatment.