Skip Navigation LinksHome > October 2010 - Volume 5 - Issue 10 > Staging System of Thymoma
Journal of Thoracic Oncology:
doi: 10.1097/JTO.0b013e3181f20c05
Malignancies of the Thymus Supplement

Staging System of Thymoma

Masaoka, Akira MD, PhD

Free Access
Article Outline
Collapse Box

Author Information

Emeritus Professor of Surgery, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan.

Disclosure: The author declares no conflicts of interest.

Address correspondence to: Akira Masaoka, MD, PhD; Emeritus Professor of Surgery, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan. E-mail: tenpaku@med.nagoya-cu.ac.jp

Collapse Box

Abstract

Introduction: Thirty years have gone by since the Masaoka staging system of thymoma was proposed in 1981. Although the Masaoka staging system has been accepted by many surgeons and pathologists, some proposals of revision and improvements have been suggested. At this time, I reinvestigated the Masaoka staging system based on the recent follow-up study of the thymomas resected at Nagoya City University.

Methods: Using the follow-up results of 211 thymomas in Nagoya, I analyzed the following aspects: (1) evaluation of the Masaoka staging system as a prognostic factor in the Nagoya series and (2) critical assessment of the proposals of revision to the Masaoka staging system.

Results: (1) Univariate analysis showed that Masaoka stages were significantly prognostic for overall survival (p < 0.0001). (2) The difference of survivals between stage I and II was not significant, but progression-free survival of stage I was 100% for up to 20 years, whereas one tumor death case in stage II was found. (3) Differences of survival between the cases with and without great vessel invasion in stage III were not significant. (4) Prognosis of N+ tumors was yet better defined.

Conclusion: (1) The Masaoka staging system remains a valuable prognostic factor. (2) Combination of stage I with II and separation of stage III into subgroups are not recommended. (3) At the moment, it is better to include N+ tumors in stage IVb.

In the 1960s, thymomas were classified into two categories: noninvasive and invasive. Bernatz et al.1 proposed a histologic classification and described four histologic subtypes: predominantly lymphocytic type, predominantly epithelial type, predominantly mixed type, and predominantly spindle cell type. Tumors belonging to the first three categories consisted of round epithelial cells and lymphocytes with differences in the ratio of the two components determining the category that the tumor belonged to. Bernatz et al. found that the predominantly epithelial subtype of thymoma was the most dominant among invasive thymomas.

To quantify the ratio of epithelial cells and lymphocytes, we2 calculated the numbers of these cells and determined the lymphocytes to epithelial cells ratio (L/E ratio). In our series of 22 cases, L/E ratios ranged from 0.27 to 4.10 with a mean value of 1.93 (Figure 1).

Figure 1
Figure 1
Image Tools

We also found a significant difference in the L/E ratios of invasive and noninvasive thymomas (Figure 2) with invasive thymomas having a lower ratio than noninvasive thymomas. This had also been recognized by Bernatz et al. In our series, we found three cases of recurrent thymoma with L/E ratios that were strikingly lower as compared with tumors at the time of diagnosis (Figure 3). Hence, it seemed that the L/E ratio changed based on the biologic behavior of the tumor: a decrease in the L/E ratio seemed to correlate with a change in the nature of the tumor from encapsulated thymoma to invasive thymoma. This finding led to propose a staging system for thymomas.

Figure 2
Figure 2
Image Tools
Figure 3
Figure 3
Image Tools
Back to Top | Article Outline

MASAOKA STAGING SYSTEM

To establish a new staging system for thymomas, we took into consideration a few basic characteristics of the disease:

1. Thymoma is a very slow growing tumor. In our experience, it could take as long as 30 years for a noninvasive thymoma to become invasive if it was not surgically resected.

2. Some noninvasive thymomas that had been resected completely recurred in the same region.

3. Some patients with completely resected invasive thymomas could survive.

4. Pleural dissemination is a frequent site of progression.

5. Some patients develop lymphogenous or hematogeneous metastases.

Two other staging systems were developed by Bergh et al.3 and Wilkins and Castleman4 (Table 1). The difference between the two systems was limited to one point: stage II Wilkins thymoma involved pleura and pericardial invasion.

Table 1
Table 1
Image Tools

In our opinion, there were some major inadequacies in the staging systems of Bergh et al. and Wilkins and Castleman: (1) the actual sites of intrathoracic metastasis were not clear, (2) the description of invasion was inadequate, (3) stage III includes too broad spectrum of disease, and (4) hematogenous and lymphogenous metastases were not well represented.

We proposed a surgical and pathologic staging system based on the extent of disease in 1981 (Table 2).5 We defined tumor extent as follows: (1) stage I: completely encapsulated, (2) stage II: invasion by the second layer surrounding the tumor (defined below), (3) stage III: direct invasion beyond the second layer, and (4) discontinuous progression.

Table 2
Table 2
Image Tools

The anatomic layers surrounding the tumor are depicted in Figure 4. Invasion into the pleural cavity is preceded by breaching of anatomic barriers in the following sequence: capsule, mediastinal pleura, visceral pleura, and finally lung. On the other hand, when the tumor invades the mediastinum, it invades through the capsule, mediastinal adipose tissue, pericardium or great vessels, and finally heart. Accordingly, the second layer surrounding the tumor is either the mediastinal pleura or mediastinal adipose tissue.

Figure 4
Figure 4
Image Tools

There are two points that need to be stressed. First, stage II disease involves invasion into the mediastinal pleura at perioperative inspection, even if histologic invasion could not be proved. We believe that this finding precedes histologic invasion.

Second, stage IV disease is divided into two categories IVa and IVb. Stage IVa consists of serosal dissemination, i.e., involvement of the pleura and pericardium. Stage IVb consists of metastasis via lymphogenous and hematogenous routes. Accordingly, stage IVb includes cases with lymph node metastasis at any station.

In 1981, we reviewed 93 thymoma cases at Osaka University.5 The stage of disease is shown in Table 3. All three cases of stage IVb disease had hematogenous metastasis involving the liver, lung, and rib. Survival curves associated with these four stages are shown in Figure 5.

Table 3
Table 3
Image Tools
Figure 5
Figure 5
Image Tools
Back to Top | Article Outline

CHANGE IN THE OPERATIVE PROCEDURE FOR THYMOMA

In 1970s, an important problem for thoracic surgeons was the operative route in thymectomy for myasthenia gravis. Whether transsternal or transcervical, the target of resection was the intracapsular thymus. I call it simple thymectomy.

However, we became aware that the results of transcervical thymectomies were inferior to those of transsternal thymectomies. We suspected that parts of the thymic tissues might be left in the transcervical thymectomies. So, we investigated the presence and distribution of thymic tissues in the anterior mediastinum systematically.6 As a result of this study, we found many foci of thymic tissues in the extrathymic anterior mediastinal adipose tissues in most cases.

I thought that to eliminate the thymic tissues, en bloc resection of the anterior mediastinal adipose tissues including the thymus gland is necessary. So, I advocated to apply this procedure in the treatment of myasthenia gravis, and named it “extended thymectomy”.7 It could improve the operative results for myasthenia gravis.

On the other hand, at that time, another problem concerning thymomectomy was pointed out. It was the occurrence of myasthenia gravis after thymomectomy for nonmyasthenic patients.8 In this era, resection of only the thymoma was the standard operation for thymoma. I call it simple thymomectomy. I thought, to prevent the occurrence of postthymomectomy myasthenia gravis, surgeons should perform the extended thymectomy, instead of simple thymomectomy for the thymoma patients, too.5

The aim of this procedure was prevention of the postthymomectomy myasthenia gravis, and its conduct in all thymoma patients was supported by finding of lack of immunologic deficiency in the myasthenic patients, who had undergone these operations.

Table 4 shows transition of operative procedures in Nagoya City University occurring every 5 years. In early 1980s, simple thymomectomy was dominant, but later the extended thymectomies were performed in most patients. After the extended thymectomy became a routine operation for thymoma, we encountered lymphnode metastases, which were found in the anterior mediastinal adipose tissues resected by this operation. This let us propose our tumor, node, metastasis (TNM) classification system.

Table 4
Table 4
Image Tools
Back to Top | Article Outline

TNM CLASSIFICATION SYSTEM

In 1991, we proposed a TNM classification system for thymoma (Table 5).9 T factors mimicked the staging criteria. N1 corresponds to anterior mediastinal, N2 other mediastinal, and N3 extrathoracic lymphnode metastasis. M1 means hematogenous metastasis in any organ.

Table 5
Table 5
Image Tools

However, the staging system was not revised. All N-positive and M-positive cases were included in stage IVb.

Back to Top | Article Outline

EVALUATION OF THE MASAOKA STAGING SYSTEM AS A PROGNOSTIC FACTOR

Many authors investigated the relationship between the Masaoka stage and survival. Major articles on these investigations since 1990 are listed in Table 6.10–26 Almost all authors except Wilkins et al.13 recognized a significant correlation between staging and survival.

Table 6
Table 6
Image Tools
Back to Top | Article Outline

REINVESTIGATION OF THE MASAOKA STAGING SYSTEM

In 1980, I was transferred to Nagoya City University from Osaka University. In 2010, we performed a follow-up study of thymomas in Nagoya City University and reinvestigated the relationship between stages and survival after an interval of 30 years from the Osaka series.27

The series consists of 211 thymomas treated in Nagoya City University Hospital from 1971 to 2008. Of these, 201 patients underwent operations: exploratory thoracotomy in 7, subtotal resection of tumor in 30, simple thymomectomy in 33, and extended thymectomy in 131 patients. Stage distribution is as follows: stage I 76 (36.0%), stage II 61 (28.9%), stage III 31 (14.7%), stage IVa 33 (15.6%), and stage IVb 10 (4.7%). Two survival parameters were adopted: overall survival and progression-free survival.

The overall survivals of each stage are shown in Figure 6 and Table 7. The curves are arranged stepwise according to the stages. Survival of IVb seems somewhat superior to that of IVa. Significant differences are found in I versus III, I versus IVa, I versus IVb, II versus III, II versus IVa, II versus IVb.

Figure 6
Figure 6
Image Tools
Table 7
Table 7
Image Tools

The progression-free survivals are shown in Figure 7 and Table 8. The survival curves of I and II become much closer and have high percentage of survival (stage I: 100%, stage II: 94.1%). The survival curves of stage IVa and IVb are nearly superimposed. Overall survival curves seem to fall into three groups; stage I and II, stage III, and stage IVa and IVb.

Figure 7
Figure 7
Image Tools
Table 8
Table 8
Image Tools

The reason why the progression-free survival rates of stage I and II lie higher than the overall survival curves is that most of the deaths in stage I and II are due to the associated disease or complication of therapies, but not the tumor. Indeed, only one patient with stage II died due to tumor recurrence. In contrast, the progression-free survival rates of stage IVb were lower than the overall survival rates. This finding was due to the presence of persistent tumor in surviving patients with stage IVb.

Univariate analysis proved that the relationship between overall survival and stage is significant (p < 0.0001). However, we could not show a relation between progression-free survival and stage, because the progression-free survival in stage I remained 100% up to 20-year follow-up.

Back to Top | Article Outline

COMPARISON OF THE RESULTS OF THE OSAKA SERIES AND THE NAGOYA SERIES

I compared the results of the Osaka series (n = 93) and the Nagoya series (n = 211). Both were investigated with 30 years follow-up. Differences are as follows; (1) the overall survivals in stage I and II of the Nagoya series are superior to those of Osaka series (10-year survival rates of stage I: 87.1% versus 66.7%, those of stage II: 80.6% versus 60.0%), (2) the overall survivals in stage IV of Nagoya series are superior to that of Osaka series (10-year survival rates of stage IV: IVa 20.0%, IVb 51.9% versus IV 0%).

These differences might be caused by the change in the operative procedures. In the Osaka series, a majority of patients underwent simple thymomectomies, whereas a majority of patients in the Nagoya series underwent extended thymectomies. The extended thymectomies could reduce tumor recurrence, particularly in stage I and II, and increase incidence of N+ cases classified in stage IVb.

Indeed, in the Osaka series, recurrences occurred in two of 37 of stage I and three of 13 of stage II. On the other hand, in the Nagoya series, recurrence occurred only in one of 61 of stage II, who had undergone simple thymomectomy.

Furthermore, in the Nagoya series, nine N+ cases were detected, whereas no N+ cases were observed in the Osaka series. There were five cases of T3N1M0. Although the mean follow-up period was only 4.3 years, two T3N1M0 cases had recurrences. Prognosis of N+ thymoma is a significant problem.

Back to Top | Article Outline

PROPOSALS OF REVISION OF THE MASAOKA STAGING SYSTEM

Although the Masaoka staging system was in general recognized as a valuable prognostic indicator, some proposals for its revision have been suggested.

Back to Top | Article Outline
Combination of Stage I and II

Gupta et al.28 reviewed 21 papers published in the 21st century and performed a metaanalysis of the results of stage I and II cases. Because they did not find significant differences between them, they proposed to combine stage I and II, and collapse the staging system into three stages.

I had read these 21 articles in detail and found two issues that are important and need to be mentioned. The first is divergence of the definition of stage II among reports. Some include the cases with histologic invasion to mediastinal pleura into stage II, whereas others include them into stage III. Such confusion might originate from the insufficient description of stage II that I wrote myself. The paragraph in the definition of stage II 1 (Table 2) should be read as follows: with or without histologic invasion.

The second issue to be mentioned is the lack of description of operative procedures in many of these 21 reports. As mentioned earlier, the operative procedure might affect the results. Whether extended thymectomy or simple thymomectomy was used is an important issue. However, many authors did not describe it in their reports. The results of stage I and II should be reinvestigated after addressing these two issues.

Furthermore, there is a trend to use thoracoscopic surgery for stage I and II thymoma, which does not allow to perform an extended thymectomy. The prevalence of thoracoscopic surgery could affect the survival of stage I and II thymomas.

However, some reports recognized a significant difference between the results of stage I and II, as described by Haniuda et al.,29 Mineo et al.,24 Maggi et al.,30 and Lardinois et al.14

I did not observe significant differences between stage I and II in overall survival. However, progression-free survival of stage I remained at 100% for a long time in our series. In such circumstance, the presence of a tumor death case in stage II might be meaningful, even if the number of cases is very small.

Based on the above mentioned reasons, I think, it is better to keep the separation of stage I and II at the moment.

Back to Top | Article Outline
Separation of Subgroups in Stage III

Okumura et al.31 suggested to divide stage III into two subclasses, i.e., IIIa: without great vessel invasion and IIIb: with great vessel invasion, based on his own data.

However, we did not find significant differences in survival between the group with great vessel invasion and that without great vessel invasion in our Nagoya series.

There are scarce reports concerning this subdivision.

Back to Top | Article Outline
Introduction of Other Factors Into the Staging System

Asamura et al.32 proposed to add tumor size, based on his own data, considering that large tumors (>10 cm in diameter) have a worse prognosis. However, tumor extent, as proposed in the Masaoka staging system seems to be more relevant than dimension. Furthermore, the fact that some types of thymomas can be dramatically reduced in size rapidly after administration of steroids argues against this proposal for change.33

There have been proposals to include completeness of operation in the staging system.11,34 Needless to say, complete resection is the best prognostic factor. However, I think, that a staging system should offer a guideline to select the therapeutic modality. Completeness of resection could be designated by a separate parameter. I propose the following criteria for results of surgery:

R0a = complete resection by extended thymectomy,

R0b = complete resection by simple thymomectomy,

R1 = microscopically incomplete resection,

R2 = macroscopically incomplete resection.

The subdivision of R0 cases should be useful to analyze the results.

Back to Top | Article Outline

TNM CLASSIFICATION AND STAGE

In 2004, the World Health Organization (WHO) proposed a system of TNM classification and staging (Table 9).35 This system is similar except for two points: (1) histologic invasion into mediastinal pleura is categorized into T3, (2) T1N1M0, T2N1M0, and T3N1M0 belong to stage III.

Table 9
Table 9
Image Tools

In 2005, Bedini et al.26 proposed a new system named Istituto Nazionale Tumori TNM-based staging system (Table 10). I think, the system by Bedini et al. is too complicated and limitation of the involved area by dissemination might cause confusion in this staging system.

Table 10
Table 10
Image Tools

A major difference between the WHO system and mine focuses on the allocation of N1. In the WHO staging system, N1 is included in stage III, whereas it is included in stage IVb in my classification. My system respects the concept of invasion by contiguity (stage III) versus discontinuous progression (stage IV). On the contrary, the WHO staging system respects localization of the involved area, giving priority to the results of operation.

Staging systems of cancers of other organs use progression of metastasis to lymphnode stations as an important part of stage definition. However, lymphnode metastasis is a rare situation in thymoma, and its progression stages are poorly defined. Kondo and Monden36 collected follow-up data of 1064 thymomas in Japan and there were only 19 (1.8%) cases of N+ thymomas. This is a low incidence compared with the cancers of other organs. Also in our series, the incidence of N+ tumor was low: nine of 211 (4.3%). The behavior of lymphogenous metastasis of thymoma warrants further investigation. Until more information is obtained, it might be better to include the N+ thymomas in IVb stage.

Back to Top | Article Outline

EPILOGUE

The Masaoka staging system still remains a valuable and reproducible prognostic factor of thymoma. However, some proposals of revision of the staging system have been offered, to identify significant differences in survival between each identified stage. In my opinion, the staging system should obey the following principles:

1. It should be logically justified.

2. It should be simple to use.

3. Frequent revisions should be avoided.

Since modalities of diagnosis and treatment are progressing day by day, a staging system should be as comprehensive as yet as simple as possible.

Back to Top | Article Outline

REFERENCES

1.Bernatz PE, Harrison EG, Clagett OT. Thymoma: a clinicopathologic study. J Thorac Cardiovasc Surg 1961;42:424–444.

2.Masaoka A, Nagaoka Y, Maeda M, et al. Study on the ratio of lymphocytes to epithelial cells in thymoma. Cancer 1977;40:1222–1228.

3.Bergh NP, Gatzinsky P, Larsson S, et al. Tumors of the thymus and thymic region. I. Clinicopathological studies on thymomas. Ann Thorac Surg 1978;25:91–98.

4.Wilkins EW Jr, Castleman B. Thymoma: a continuing survey at the Massachusetts General Hospital. Ann Thorac Surg 1979;28:252–256.

5.Masaoka A, Monden Y, Nakahara K, et al. Follow-up study of thymomas with special reference to their clinical stages. Cancer 1981;48:2485–2492.

6.Masaoka A, Nagaoka Y, Kotake Y. Distribution of thymic tissue at the anterior mediastinum. Current procedures in thymectomy. J Thorac Cardiovasc Surg 1975;70:747–754.

7.Masaoka A, Monden Y. Comparison of the results of transsternal simple, transcervical simple, and extended thymectomy. Ann N Y Acad Sci 1981;377:755–765.

8.Kimura J, Van Allen MW. Post-thymomectomy myasthenia gravis. Report of a case of ocular myasthenia gravis after total removal of a thymoma and review of literature. Neurology 1967;17:413–420.

9.Yamakawa Y, Masaoka A, Hashimoto T, et al. A tentative tumor-node-metastasis classification of thymoma. Cancer 1991;68:1984–1987.

10.Pescarmona E, Rendina EA, Venuta F, et al. Analysis of prognostic factors and clinicopathological staging of thymoma. Ann Thorac Surg 1990;50:534–538.

11.Regnard JF, Magdeleinat P, Dromer C, et al. Prognostic factors and long-term results after thymoma resection: a series of 307 patients. J Thorac Cardiovasc Surg 1996;112:376–384.

12.Gripp S, Hilgers K, Wurm R, et al. Thymoma: prognostic factors and treatment outcomes. Cancer 1998;83:1495–1503.

13.Wilkins KB, Sheikh E, Green R, et al. Clinical and pathologic predictors of survival in patients with thymoma. Ann Surg 1999;230:562–574.

14.Lardinois D, Rechsteiner R, Läng RH, et al. Prognostic relevance of Masaoka and Müller-Hermelink classification in patients with thymic tumors. Ann Thorac Surg 2000;69:1550–1555.

15.Ogawa K, Uno T, Toita T, et al. Postoperative radiotherapy for patients with completely resected thymoma: a multi-institutional, retrospective review of 103 patients. Cancer 2002;94:1405–1413.

16.Okumura M, Ohta M, Tateyama H, et al. The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer 2002;94:624–632.

17.Nakagawa K, Asamura H, Matsuno Y, et al. Thymoma: a clinicopathologic study based on the new World Health Organization classification. J Thorac Cardiovasc Surg 2003;126:1134–1140.

18.Kondo K, Yoshizawa K, Tsuyuguchi M, et al. WHO histologic classification is a prognostic indicator in thymoma. Ann Thorac Surg 2004;77:1183–1188.

19.Park MS, Chung KY, Kim KD, et al. Prognosis of thymic epithelial tumors according to the new World Health Organization histologic classification. Ann Thorac Surg 2004;78:992–998.

20.Rea F, Marulli G, Girardi R, et al. Long-term survival and prognostic factors in thymic epithelial tumours. Eur J Cardiothorac Surg 2004;26:412–418.

21.Zhu G, He S, Fu X, et al. Radiotherapy and prognostic factors for thymoma: a retrospective study of 175 patients. Int J Radiat Oncol Biol Phys 2004;60:1113–1119.

22.Kim DJ, Yang WI, Choi SS, et al. Prognostic and clinical relevance of the World Health Organization schema for the classification of thymic epithelial tumors: a clinicopathologic study of 108 patients and literature review. Chest 2005;12:755–761.

23.Rena O, Papalia E, Maggi G, et al. World Health Organization histologic classification: an independent prognostic factor in resected thymomas. Lung Cancer 2005;50:59–66.

24.Mineo TC, Ambrogi V, Mineo D, et al. Long-term disease-free survival of patients with radically resected thymomas: relevance of cell-cycle protein expression. Cancer 2005;104:2063–2071.

25.Wright CD, Wain JC, Wong DR, et al. Predictors of recurrence in thymic tumors: importance of invasion, World Health Organization histology, and size. J Thorac Cardiovasc Surg 2005;130:1413–1421.

26.Bedini AV, Andreani SM, Tavecchio L, et al. Proposal of a novel system for the staging of thymic epithelial tumors. Ann Thorac Surg 2005;80:1994–2000.

27.Masaoka A, Yano M, Eimoto T, et al. Follow-up study of thymomas and reinvestigation of the Msaoka staging system. Cancer. In press.

28.Gupta R, Marchevsky AM, McKenna RJ, et al. Evidence-based pathology and the pathologic evaluation of thymomas: transcapsular invasion is not a significant prognostic feature. Arch Pathol Lab Med 2008;132:926–930.

29.Haniuda M, Kondo R, Numanami H, et al. Recurrence of thymoma: clinicopathological features, re-operation, and outcome. J Surg Oncol 2001;78:183–188.

30.Maggi G, Casadio C, Cavallo A, et al. Thymoma: results of 241 operated cases. Ann Thorac Surg 1991;51:152–156.

31.Okumura M, Miyoshi S, Takeuchi Y, et al. Results of surgical treatment of thymomas with special reference to the involved organs. J Thorac Cardiovasc Surg 1999;117:605–613.

32.Asamura H, Nakagawa K, Matsuno Y, et al. Thymoma needs a new staging system. Interact Cardiovasc Thorac Surg 2004;3:163–167.

33.Tateyama H, Takahashi E, Saito Y, et al. Histopathologic changes of thymoma preoperatively treated with corticosteroids. Virchows Arch 2001;438:238–247.

34.Gamondès JP, Balawi A, Greenland T, et al. Seventeen years of surgical treatment of thymoma: factors influencing survival. Eur J Cardiothorac Surg 1991;5:124–131.

35.International academy of pathology (IAP). TNM classification of malignant thymic epithelial tumours. In WD Travis, E Brambilla, HK Müller-Hermelink, CC Harris (Eds.), Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon: IARC Press, 2004. Pp. 147.

36.Kondo K, Monden Y. Lymphogenous and hematogenous metastasis of thymic epithelial tumors. Ann Thorac Surg 2003;76:1859–1865.

Keywords:

Thymoma; Staging system; TNM classification; Follow-up study

© 2010International Association for the Study of Lung Cancer

Login

Article Tools

Images

Share

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Other Ways to Connect

Twitter
twitter.com/JTOonline

 



Visit JTO.org on your smartphone. Scan this code (QR reader app required) with your phone and be taken directly to the site.

 For additional oncology content, visit LWW Oncology Journals on Facebook.