Postoperative survival for patients with thymoma complicating myasthenia gravis—preliminary retrospective results of the ChART database

Abstract

Background

It is so far not clear that how myasthenia gravis (MG) affected the prognosis of thymoma patients. The aim of this assay is to compare the postoperative survival between patients with thymoma only and those with both thymoma and MG.

Methods

The Chinese Alliance for Research in Thymomas (ChART) registry recruited patients with thymoma from 18 centers over the country on an intention to treat basis from 1992 to 2012. Two groups were formed according to whether the patient complicated MG. Demographic and clinical data were reviewed, patients were followed and their survival status were analyzed.

Results

There were 1,850 patients included in this study, including 421 with and 1,429 without MG. Complete thymectomy were done in 91.2% patients in MG group and 71.0% in non-MG group (P<0.05). There were more percentage of patients with the histology of thymoma AB, B1, or B2 (P<0.05) in MG group, and more percentage of patients with MG were in Masaoka stage I and II. The 5- and 10-year overall survival (OS) rates were both higher in MG group (93% vs. 88%; 83% vs. 81%, P=0.034) respectively. The survival rate was significantly higher in patients with MG when the Masaoka staging was 3/4 (P=0.003). Among patients with advanced stage thymoma (stage 3, 4a, 4b), the constituent ratios of 3, 4a, 4b were similar between MG and non-MG group. Histologically, however, there were significantly more proportion of AB/B1/B2/B3 in the MG group while there were more C in the non-MG group (P=0.000). Univariate analyses for all patients showed that MG, WHO classification, Masaoka stage, surgical approach, chemotherapy and radiotherapy and resectability were significant factors, and multivariate analysis showed WHO classification, Masaoka stage, and resectability were strong independent prognostic indicators.

Conclusions

Although MG is not an independent prognostic factor, the survival of patients with thymoma was superior when MG was present, especially in late Masaoka stage patients. Possible reasons included early diagnosis of the tumor, better histologic types, an overall higher R0 resection and less recurrence.

Introduction

Due to the immunologic feature of the thymus, quite a few thymoma patients are accompanied with autoimmune disorders among which myasthenia gravis (MG) is the most important one with the incidence rate ranging 10–45%. In early times, MG once was reported to be the negative factor (1) because of a lack of experiences to deal with MG related complications after thymothymectomy (2-4). On the other hand, there has been an increasing body of evidences indicating that MG is related to better survival. However, the role of MG in the postoperative outcome for patients with thymoma is so far still not clear. The Chinese Alliance for Research in Thymomas (ChART) registry recruited patients with thymoma from 18 centers over the country in 20 years with the intention to compare the postoperative survival between patients with thymoma only and those with both thymoma and MG so as to have a preliminary understanding of how MG affects the prognosis of patients with thymoma.

Materials and methods

Patients

The ChART registry included patients with thymoma from 18 centers over the country from 1992 to 2012. Demographic and clinical data were reviewed. In this research, two groups were formed according to whether the patient complicated MG—MG group and non-MG group. All the demographic and clinical data were compared between the two groups. Patients were followed before their survival status were compared. Because only de-identified data were used for the study, informed consent was waived by Institutional Review Board (IRB).

The final pathologic staging was based on Masaoka-Coga staging system while the histology types were based on WHO classification. The resectability of thymoma was divided into R0 (no residue tumor), R1 (residue tumor microscopically), R2 (residue tumor macroscopically), and biopsy only according to the pathology of the sample margins and the operation note. The normal thymus was either completely removed (total resection of the thymus) or partly (resection of the tumor and a portion of the thymus) removed according to the surgeon’s preference. Data regarding chemotherapy and radiotherapy were also collected.

Statistical analysis

Statistical analysis was performed by the X test and the unpaired t test with the SPSS 14.0 for Windows. Prognostic factors were analyzed by the Kaplan-Meier method and Cox regression with respect to survival. Comparisons between survival curves were made by the log-rank test.

Results

Patients’ characteristics: data of 2,306 patients were collected, and 1,850 patients entered this study, including 421 with and 1,429 without MG as showed in Table 1, after excluding the following ones: 49 patients only biopsied, 152 lack of surgical information, 124 without WHO classification, and 118 lack of Masaoka staging. There were more proportion of female in MG group than non-MG group (P=0.034) and the average age of MG group was significantly younger (49 vs. 52, P=0.000). The overall survival (OS) for MG group was significantly higher (95.95% vs. 92.29%, P=0.026). The tumor size of MG group was significantly smaller (5.6 vs. 7.2 cm, P=0.000). There were more percentage of patients with the histology of thymoma A and B (P=0.000) in MG group, and more percentage of patients with MG were in Masaoka stage I and II (P=0.004). R0 resection and complete thymectomy were both done significantly more in MG group (90.0% vs. 85.1%, P=0.009; 91.2% vs. 71.0%, P=0.000). Postoperative adjuvant chemotherapy or radiotherapy were given more in patients without MG (8.8% vs. 23.4%, P=0.000; 41.0% vs. 47.1%, P=0.03).

Table 1. Patients’ characteristics.

Characteristics	MG, n=421 (%)	Non-MG, n=1,429 (%)	P
Sex			0.034
Male	206 (20.9)	782 (79.1)
Female	215 (24.9)	647 (75.1)
Age (average), years	49	52	0.000
Overall survival (%)	95.95	92.29	0.026
WHO classification			0.000
A	18 (4.3)	89 (6.2)
AB	80 (19.0)	356 (24.9)
B1	68 (16.2)	164 (11.5)
B2	128 (30.4)	169 (11.8)
B3	107 (25.4)	256 (17.9)
C	19 (4.5)	351 (24.6)
Carcinoid	1 (0.2)	44 (3.1)
Histology			0.000
Type A thymoma	18 (4.3)	89 (6.2)
Type B (including AB) thymoma	383 (91.0)	945 (66.1)
Thymic cancer	20 (4.8)	395 (27.6)
Thymectomy			0.000
Incomplete	37 (8.8)	412 (29.0)
Complete	382 (91.2)	1,008 (71.0)
Tumor size (cm)	5.6	7.2	0.000
Masaoka staging			0.004
1	196 (46.6)	547 (38.3)
2	83 (19.7)	277 (19.4)
3	116 (27.6)	458 (32.1)
4	26 (6.2)	147 (10.3)
Chemotherapy (given)	36 (8.8)	317 (23.4)	0.000
Radiotherapy (given)	167 (41.0)	636 (47.1)	0.031
Resectability (rate of R0 resection)	379 (90.0)	1,212 (85.1)	0.009

MG, myasthenia gravis.

The 5-year and 10-year OS rates were both higher in MG group (93% vs. 88%, 83% vs. 81%, P=0.034, respectively) (Figures 1,2,3,4). However, the survival rate was significantly higher in non-MG group when the Masaoka staging was 1 (P=0.000), and the result was opposite when the Masaoka staging was 3/4 (P=0.003).

Figure 1.

Comparison of overall survival between MG group and non-MG group. MG, myasthenia gravis.

Figure 2.

Comparison of overall survival between MG group and non-MG group in Masaoka stage I. MG, myasthenia gravis.

Figure 3.

Comparison of overall survival between MG group and non-MG group in Masaoka stage II. MG, myasthenia gravis.

Figure 4.

Comparison of overall survival between MG group and non-MG group in Masaoka stage III and IV. MG, myasthenia gravis.

Among patients with advanced stage thymoma (stage 3, 4a, 4b) (Table 2), the constituent ratios of 3 and 4 were similar between MG and non-MG group. Histologically, however, there were significantly more proportion of AB/B1/B2/B3 in the MG group while there were more C in the non-MG group (P=0.000). The tumor size of MG group was also significantly smaller (6.4 vs. 7.9 cm, P=0.000). Total thymectomy had been done in more proportion of patients in MG group (86.5% vs. 72.7%, P=0.001), but the two groups were comparable in resectability.

Table 2. The specific analysis of late Masaoka stage patients between two groups.

Groups	MG, n=142 (%)	Non-MG, n=605 (%)	P
Masaoka stage			0.128
3	116 (81.7)	458 (75.7)
4	26 (18.3)	147 (24.3)
WHO classification			0.000
A	2 (1.4)	14 (2.3)
AB	15 (10.6)	38 (6.3)
B1	19 (13.4)	29 (4.8)
B2	39 (27.5)	71 (11.7)
B3	54 (38.0)	154 (25.5)
C	13 (9.2)	273 (45.1)
Carcinoid	0 (0.0)	26 (4.3)
Complete thymectomy	122 (86.5)	434 (72.7)	0.001
Tumor size (cm)	6.4	7.9	0.000
Resectability	104 (73.2)	400 (66.4)	0.119
Recurrence	16 (15.7)	137 (31.7)	0.001

MG, myasthenia gravis.

Univariate analyses showed that MG, WHO classification, Masaoka stage, approach of operation, chemotherapy and radiotherapy and resectability were significant factors for survival (Table 3). Whereas in multivariate analyses of Cox regression model (Table 4), WHO classification, Masaoka stage, and resectability were strong independent prognostic indicators.

Table 3. Univariate analyses in all patients.

Factors	P
Sex (male/female)	0.088
Age (≥50/<50)	0.289
MG (yes/no)	0.038
Tumor size (≥5 cm/<5 cm)	0.459
WHO classification (A/AB or B1 or B2 or B3/C + NETT)	0.000
Masaoka’s staging (1/2/3/4)	0.000
Approach of operation (thorascope/open)	0.043
Thymectomy (incompletely/completely)	0.041
Radiotherapy (no/yes)	0.000
Chemotherapy(no/yes)	0.000
Resectability (R0/R1 + R2)	0.000

MG, myasthenia gravis; NETT, neuroendocrine thymic tumor.

Table 4. Multivariate analyses in all patients.

Factors	P	OR (95% CI)
MG (no/yes)	0.967	1.016 (0.479, 2.157)
Sex (female/male)	0.738	0.924 (0.580, 1.470)
WHO classification	0.012
AB + B1 + B2 + B3	0.847	3,335.39 (0, 2.521E38)
C + NETT	0.827	7,582.361 (0, 5.732E38)
Masaoka stage (IV/III/II/I)	0.001
II	0.050	3.046 (1.002, 9.254)
III	0.000	6.423 (2.489, 16.577)
IV	0.000	7.034 (2.416, 20.474)
Radiotherapy (yes/no)	0.138	0.656 (0.376, 1.145)
Chemotherapy (yes/no)	0.046	1.723 (1.009, 2.942)
Approach of operation (thorascope/open)	0.655	1.401 (0.319, 6.147)
Tumor size (≥5 cm/<5 cm)	0.448	0.780 (0.411, 1.481)
Resectability (R0/R1 + R2)	0.003	0.457 (0.273, 0.766)
Thymectomy (completely/incompletely)	0.593	1.148 (0.692, 1.906)

MG, myasthenia gravis; NETT, neuroendocrine thymic tumor.

In patients with advanced stage thymoma, the postoperative recurrence was much higher in non-MG group than MG group (Figures 5,6).

Figure 5.

Comparison of tumor recurrence between MG group and non-MG group in advanced stage. MG, myasthenia gravis.

Figure 6.

Comparison of tumor recurrence between MG group and non-MG group. MG, myasthenia gravis.

Discussion

The appearance of MG in patients with thymoma makes them stand out from the population not only because of the fact the perioperative management has become more complicated, but the postoperative outcome is somewhat different. Results of earlier studies on postoperative prognosis of thymoma are quite different from those of recent studies (2,3,5). Anyway, MG does have an impact on patients with thymoma, as is shown in this multicenter study. Patients in MG group were at earlier stage of the disease and there were more proportion of type AB and B in this group, while patients with thymoma only were at quite advanced stage and their pathologic classification tended to be worse. The onset of MG is a good reason for the timely diagnosis and attention for the thymic tumor. The overall R0 resection rate, an independent prognostic factor according to the multivariate analysis of this study, was also higher in patients with MG. All these might account for the positive influence of MG on the long term outcome of subjects with thymoma.

The inter-relations among MG, tumor histology and Masaoka staging have been under the study of quite a few physicians (6,7). Ruffini et al. proposed that two models, early Masaoka stage A/AB WHO type and high Masaoka stage/B WHO type, were most frequently seen in MG patients. He noted the influence of MG on Masaoka staging and histology, and his study showed only Masaoka stage had a prognostic significance on OS and disease free survival (DFS) (6). His conclusions could be well applied to the results of this multicenter study except that in the ChART database the independent prognostic factors included histology, tumor resectability besides Masaoka stage, and MG seemed to have a positive impact on all three elements.

MG affects patients with thymoma in quite an interesting way. Although multivariate analysis indicated that MG was not an independent prognostic factor for a superior survival status, univariate study showed that on the whole patients with MG actually lived longer. Further investigation found MG had a different influence on the patients’ survival in different Masaoka stages. Theoretically MG is usually not fatal and adds no mortality to thymoma providing adequate medical care is given. However, perioperative death from myasthenic crisis or pulmonary complications is possible if importance is not fully attached. In this study the worse survival status of MG group in early stages might result from the increased death contributed by MG related complications. Dramatically, the survival advantage of MG group became extremely obvious when thymoma was in quite advanced stage. This might be explained by the better pathologic classification and lower recurrence rate of thymoma in the MG group. Although this advantage was to some extent offset by the increased mortality of MG group in early stages, the positive effect of MG on the OS of thymoma patients still came into sight as shown by the single variant analysis. Hopefully, as the understanding and intervention for MG has developed rapidly, which minimizes death caused by MG, and multidisciplinary collaboration for thymoma patients with MG has become much tighter, an even better prognosis of thymoma patients with MG will be expected in future studies.

As a retrospective study, loss of patients’ information in the ChART database caused quite a lot of cases to be excluded from the analysis. Bias hence produced seemed inevitable. In addition, patients included in the data base spanned 20 years during which medical progression was rapid, especially for MG. Therefore medical care received by these patients was not homogeneous. A prospective study may be needed to prove the results of this analysis.