Epithelial Malignant Pleural Mesothelioma After ...

Cardiopulmonar y Imaging • Original Research Gill et al. CT After Extrapleural Pneumonectomy

Downloaded from www.ajronline.org by 100.0.197.5 on 08/14/17 from IP address 100.0.197.5. Copyright ARRS. For personal use only; all rights reserved

Cardiopulmonary Imaging Original Research

Epithelial Malignant Pleural Mesothelioma After Extrapleural Pneumonectomy: Stratification of Survival With CT-Derived Tumor Volume Ritu R. Gill1 William G. Richards 2 Beow Y. Yeap 3 Shin Matsuoka1 Andrea S. Wolf 2 Victor H. Gerbaudo1 Raphael Bueno 2 David J. Sugarbaker 2 Hiroto Hatabu1 Gill RR, Richards WG, Yeap BY, et al.

Keywords: CT, imaging biomarkers, mesothelioma, pleura, pneumonectomy, survival, tumor volume DOI:10.2214/AJR.11.7015 Received April 6, 2011; accepted after revision July 13, 2011. 1 Department of Radiology, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St, Boston, MA 02115. Address correspondence to R. R. Gill ([email protected]). 2 Division of Thoracic Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA. 3 Department of Medicine, Massachusetts General Hospital, Boston, MA.

AJR 2012; 198:359–363 0361–803X/12/1982–359 © American Roentgen Ray Society

OBJECTIVE. The purpose of this study was to assess the usefulness of CT-derived tumor volume, with control for other prognostic factors, for stratifying survival after surgery-based multimodality treatment of a large cohort of patients with epithelial malignant pleural mesothelioma. MATERIALS AND METHODS. We retrospectively reviewed 338 patients with mesothelioma who underwent extrapleural pneumonectomy between 2001 and 2007. The study cohort comprised 88 patients with epithelial subtype tumors, DICOM-format CT scans, and data regarding neoadjuvant and adjuvant therapy. Tumor volume was calculated, and KaplanMeier survival and Cox regression analyses were performed to compare the estimated survival functions of patient subgroups based on volume and other covariates related to outcome (sex, age, preoperative platelet count, hemoglobin concentration, WBC count, clinical and pathologic TNM category, and administration of neoadjuvant and adjuvant therapy). A multivariate regression model was derived on the basis of the most significant univariate predictors. RESULTS. The median estimated tumor volume was 319 cm3 (range, 4–3256 cm3). In univariate analysis, tumor volume, hemoglobin concentration, platelet count, pathologic TNM category, and administration of adjuvant chemotherapy or radiation therapy met the criteria for inclusion in the reverse stepwise regression analysis. In the final model, tumor volume, hemoglobin concentration, and administration of adjuvant chemotherapy or radiotherapy were identified as independently associated with overall survival. CONCLUSION. With control of prognostic covariates, CT-derived tumor volume can be used to stratify survival of patients with epithelial mesothelioma after extrapleural pneumonectomy and should be included in prognostic evaluation of patients for whom resection is being considered.

M

alignant pleural mesothelioma (MPM) is an asbestos-related neoplasm that originates in the pleural mesothelial cells and progresses primarily by local extension to encase the lung and mediastinal structures. MPM is refractory to most of the conventional therapies [1]. The median patient survival period ranges from 7 months with palliative care to 13 months with chemotherapy [2, 3]. Patients with disease confined to the involved hemithorax and who undergo multimodality therapy based on surgical resection to remove all gross tumor and radiotherapy, chemotherapy, or both to control micrometastatic disease have had longer survival times [4, 5]. However, few tools are available to predict which patients are most likely to have sufficient survival benefit from this approach to offset the morbidity and mortality associated with aggressive treatment.

The circumferential morphologic features of MPM present challenges to clinical staging with CT, MRI, and PET. These modalities are reliable for determining extrathoracic extension of disease but lack sensitivity and specificity in predicting pathologic T and N status [6]. Moreover, even when patients are classified based on pathologic analysis, current staging systems do not effectively stratify prognosis [7]. Threedimensional display of axial images may more accurately portray overall disease extent, given the complex morphologic features of MPM. Pass et al. [8] reported that tumor volume estimated from CT scans was associated with survival of surgically treated MPM patients. However, the study cohort was not large enough to stratify or account for known prognostic factors, such as adjuvant therapy, tumor histology, and extent of resection, that may have influenced patient

AJR:198, February 2012 359


Gill et al. outcome independently of tumor volume. More recent studies showing the same association have had similar limitations [9–12]. The current study was designed to minimize or account for potential influences of surgical procedure, extent of multimodality treatment, and histologic nature of the tumor on survival. All tumors were staged pathologically. Thus we retrospectively selected for this study only patients who had undergone extrapleural pneumonectomy (EPP) as part of intended trimodality therapy, were found to have epithelial subtype tumors, and had documentation of adjuvant treatment. We investigated the relative prognostic significance of CT-derived preoperative tumor volume among other clinical and pathologic predictors of survival.

MDCT scans (Sensation 64, Siemens Healthcare) at 120 kVp, 100–200 mA (based on patient weight), 0.6-mm collimation, high-speed mode, pitch equivalent of 0.75, slice interval of 5 mm, and slice thickness of 5 mm. All scans were obtained after IV contrast administration. Seventy-five milliliters of contrast material (iopromide, Ultravist 370, Bayer HealthCare) was administered IV at 4 mL/s with an automated power injector. Total scanning time was approximately 6.3 seconds.

CT Of the 88 patients, 36 patients underwent PET/ CT scans (Discovery ST system, GE Healthcare). Patients were imaged in the supine position with their arms above their heads when possible in an attempt to reduce beam-hardening artifacts and performing shallow breathing. Images were obtained at 140 kVp and 75–120 mA (varying according to patient weight), 0.8 second per rotation, pitch of 1.675:1, reconstructed slice thickness of 3.75 mm, and total scanning time of 42.4 seconds. No IV or oral contrast material was administered. Fifty-two patients underwent 64-

360

Clinical Stage

Pathologic Stage

The CT images were analyzed by a thoracic radiologist with 10 years of experience. The

I

II

III

IV

Total

I

1

0

1

0

2

II

7

3

6

0

16

III

12

5

25

1

43

IV

10

5

10

2

27

Total

30

13

42

3

88

Analysis of CT Images

TABLE 2: Results of Univariate Overall Survival Analysis for Putative Prognostic Variables Survival Rate (%)

Materials and Methods Subjects With approval from the institutional review board, we retrospectively reviewed data from the International Mesothelioma Program Patient Data Registry. We audited the medical charts of 338 patients with MPM who underwent cytoreductive surgery by EPP between 2001 and 2007. The indications for primary resection at our institution include tissue biopsy confirming mesothelioma, unilateral disease, and a negative mediastinoscopy. We identified a homogeneous cohort who had epithelial histologic features at final pathologic examination; 112 of these patients had DICOMformat preoperative CT scans available. Among the 112, 88 patients had complete data regarding whether adjuvant chemotherapy or radiotherapy was administered. These 88 patients constituted the final study cohort, for whom CT data, clinical and pathologic stage, demographic and preoperative laboratory factors related to prognosis (sex, age, platelet count, hemoglobin concentration, and WBC count [13–15]), and vital status (known to be alive, known to be dead, lost to follow-up) had been obtained. The study was compliant with HIPAA [16].

TABLE 1: Comparison of Clinical and Pathologic TNM Staging Systems for Malignant Pleural Mesothelioma

Characteristic

n

Median

1y

3y

5y

Sex

p 0.2456

Women

22

22.1

77

39

26

Men

66

18.0

66

29

14

≤ 60 y

41

18.1

73

33

24

> 60 y

47

19.5

66

31

12

Within normal limits

75

19.5

72

33

16

Leukocytosis

13

12.9

54

23

0

Age at surgery

0.2999

WBC count

0.5989

Platelet count

0.0417


70

21.7

75

37

18

Thrombocytosis

18

11.9

44

11

11


52

31.9

79

44

27

Anemia

36

13.4

56

22

0

≤ 500 cm3

59

24.4

79

44

24

> 500 cm3

29

12.0

48

7

0

Hemoglobin concentration

0.0006

Tumor volume

< 0.0001

Neoadjuvant chemotherapy

0.3418

Administered

20

18.1

80

25

0

Not administered

68

20.5

66

34

21

Administered

60

25.9

83

43

23

Not administered

28

8.3

37

8

0

I–II

43

24.4

76

37

16

III–IV

45

15.9

62

27

18

Adjuvant chemotherapy or radiotherapy

< 0.0001

Clinical stage

0.4774

Pathologic stage

0.1897

I–II

18

32.2

83

44

25

III–IV

70

17.7

65

28

16

AJR:198, February 2012

CT After Extrapleural Pneumonectomy 25

1.0

Proportion Surviving

No. of Patients


20

15

10

0.8 0.6 Tumor Volume ≤ 500 cm3

0.4 0.2

5

0.0

0 0

500

1000

1500 2000 2500 Tumor Volume (cm3)

3000

3500

Fig. 1—Graph shows distribution of tumor volume for epithelial subtype malignant pleural mesothelioma.

tumor was identified and selectively segmented with the 3D volume feature of the software (Vitrea Enterprise suite 6.0, Vital Images). Pleural effusion, atelectasis, adjacent solid organs, and chest wall musculature were excluded from the tumor. Extrapleural sites of tumor and discontinuous regions of involvement were identified and manually added to total tumor volume. The same radiologist analyzed all the CT scans for clinical staging using the American Joint Committee on Cancer/International Union Against Cancer TNM system. Because the study was retrospective and all patients were surgical candidates, the M status of all tumors was known to be M0.

Statistical Analysis Dichotomous categoric transforms of continuous variables were derived by datadriven approaches or based on clinical criteria. For tumor volume, an optimal cutpoint was based on maximizing the observed hazard ratio in univariate analysis. Survival functions were computed with volume dichotomized at cutpoint intervals of 100 cm 3. Patient age at surgery was dichotomized close to the median value. Preoperative laboratory values were transformed according to World Health Organization criteria (anemia, hemoglobin concentration less than 12 g/dL for women and 13 g/dL for men) [17] or institutionally defined normal limits (thrombocytosis, platelet count greater than 450,000/µL; leukocytosis, WBC count greater than 10,000/µL). TNM stages I and II were pooled and compared with stages III and IV.

Tumor Volume > 500 cm3

0

10

20

30 40 50 60 70 Time After Surgery (mo)

80

90

Fig. 2—Graph shows Kaplan-Meier survival estimates for epithelial tumors with volume of 500 cm 3 or less versus those with a volume greater than 500 cm 3.

Survival duration was defined as the time from the date of surgery to the date of death or last follow-up examination. Survival duration was censored at date of last contact for patients who were alive at the end of the study or were lost-to-follow-up. Kaplan-Meier analysis was used to estimate the survival functions for patient subgroups. Cox regression was used to estimate the hazard ratio and to determine the level of significance associated with each categoric variable in univariate analysis of overall survival. Effects yielding p < 0.20 by univariate analysis were included in a backward stepwise procedure in which Cox regression was used to derive a final multivariate model comprising the first set of variables encountered with all adjusted p < 0.05. Pairwise associations between categoric variables were assessed with the Fisher exact test. StatView version 4.5 software (Abacus Concepts) was used for all statistical calculations. All p values were two sided.

Results Eighty-eight patients (66 [75%] men, 22 [25%] women; median age at surgery, 61 years; range, 31–78 years) had epithelial subtype tumors and complete data required for this study, including available preoperative DICOM CT data for volumetric analysis and documentation of whether neoadjuvant or adjuvant chemotherapy or adjuvant radiation therapy had been administered. Median preoperative laboratory values were as follows: WBC count, 8.1 × 103/µL (range, 3.3–15.3 × 103/µL); platelet count, 309 × 103/µL (range,

150–898 × 103/µL); hemoglobin concentration (men), 13.9 g/dL (range, 8.6–16.5 g/dL); hemoglobin concentration (women), 11.7 g/dL (range, 8.7–14.4 g/dL). Clinical and pathologic stage distributions for the cohort are presented in Table 1. The clinical stage was predictive of the pathologic stage in 31 patients (35%), was lower in 49 patients (56%), and was higher in eight patients (9%). Twenty patients (23%) underwent neoadjuvant chemotherapy before EPP. Thirty-eight patients (43%) underwent adjuvant chemotherapy postoperatively. Fifty patients (57%) underwent adjuvant radiotherapy. Twenty-eight patients (32%) underwent both adjuvant modalities. The median overall survival period for the entire cohort was 18.7 months. The results of univariate survival analysis of all putative prognostic factors are presented in Table 2. The distribution of estimated tumor volumes was skewed (Fig. 1). The median tumor volume was 319 cm3 (range, 4–3256 cm3). Comparisons of survival functions with volume dichotomized at intervals of 100 cm3 yielded the largest observed difference at a cutpoint of 500 cm3. The 59 patients with a tumor volume of 500 cm3 or less had a significantly longer survival duration than the 29 patients with a tumor volume greater than 500 cm3 (24.4 versus 12 months; p < 0.0001) (Fig. 2). Tumor volume, hemoglobin concentration, platelet count, pathologic TNM stage, and provision of adjuvant chemotherapy or radiation therapy met the criterion for inclusion in the multivariate analysis. Backward stepwise


Gill et al. TABLE 3: Backward Stepwise Cox Regression Model Covariate

Hazard Ratio

95% CI

p

2.02

1.18–3.47

0.0109


Tumor volume > 500 cm3 Anemia

1.99

1.19–3.33

0.0089

Adjuvant chemotherapy or radiotherapy

0.30

0.18–0.52

< 0.0001

Cox regression resulted in a model comprising tumor volume greater than 500 cm3 (hazard ratio, 2.02), anemia (hazard ratio, 1.99), and provision of adjuvant chemotherapy or radiotherapy (hazard ratio, 0.30) as covariates independently associated with survival duration (Table 3). This model was confirmed with a forward stepwise procedure. Discussion The results of our study established that preoperative CT-derived tumor volume and hemoglobin concentration are strongly and independently associated with survival after EPP in patients with epithelial MPM. These findings support the recommendation that tumor volume estimation be incorporated as part of the routine preoperative assessment of patients with epithelial MPM for whom EPP is being considered. Pass et al. [8] first suggested that preoperative tumor volume might serve as a clinical indicator of T category of MPM. That study showed a robust association between volume and overall and progression-free survival, despite including a relatively small number of patients (n = 48) who had a mixture of histologic subtypes and underwent different surgical procedures. Controlling for other prognostic factors, we confirmed that this association remains significant in a larger cohort of patients with tumors with purely epithelial histologic features who underwent a uniform surgical procedure. The clinical TNM system is currently the standard metric used for prognosis in the care of MPM patients for whom cytoreductive surgery is being considered. Accurate clinical staging of MPM, particularly determination of T status, continues to be challenging owing to the complex morphologic features of this tumor. CT is the primary method for preoperative evaluation of patients with MPM. The addition of MRI improves detection of diaphragmatic and chest wall invasion [18]. Use of PET/CT has improved overall TNM staging, particularly through detection of occult metastasis [19, 20], but accurate prediction of nodal status continues to be limited for all imaging modalities [21–23].

362

All patients in the current cohort underwent EPP and had the pathologic stage accurately determined. In agreement with a previous report [24], clinical staging had low accuracy in prediction of pathologic stage. As a univariate predictor of survival, clinical stage did not meet the criterion for inclusion in the multivariate analysis. The relation between pathologic stage and prognosis also was relatively weak, as previously documented [7], and was not significant after correction for other prognostic variables. The current study showed that tumor volume and hemoglobin concentration are preoperative predictors of survival that are independent of clinical stage, at least for patients with resectable epithelial MPM. One additional postoperative factor (ability to undergo intended adjuvant therapy) conferred additional independent prognostic information. Interestingly, the combination of large tumor volume and anemia was significantly associated with patients’ lack of fitness or lack of willingness to undergo adjuvant treatment. Eleven of 20 patients (55%) with both tumor volume greater than 500 cm3 and anemia did not undergo adjuvant therapy, compared with only 17 of the other 68 patients (25%) (p = 0.0154). Although the details of this interaction await further investigation in a larger dataset, the observation further supports the postulate that radiographic tumor volume and hemoglobin concentration are the basis for improved preoperative prognostic assessment. Additional predictors of fitness to undergo adjuvant therapy and minimally invasive pathologic or molecular markers of survival [25] could be incorporated to enhance prognostic accuracy. At publication of the findings of Pass et al. [8], estimation of tumor volume was a time- and labor-intensive process requiring specialized apparatus. However, current availability of sophisticated U.S. Food and Drug Administration–approved software and its hybridization with PACS workstations enables incorporation of tumor volume estimation into routine radiographic assessment and workflow for epithelial MPM. This addition would improve patient care by allowing more reliable prognos-

tic estimates for individual patients as they consider the risks and benefits of surgerybased multimodality treatment. The chief limitations of this study were that it was retrospective and that the patient cohort was selected on the basis of the availability of imaging and postoperative data, including complete surgical resection by EPP, pathologic finding of epithelial histologic features, and administration of adjuvant therapy. Future work will explore the generalizability of these findings to nonepithelial disease and to other cytoreductive procedures. Future work also will be performed to assess the value of CT-derived tumor volume and anemia in preoperative screening of surgical candidates. References 1. Robinson BW, Lake RA. Advances in malignant mesothelioma. N Engl J Med 2005; 353:1591– 1603 2. Merritt N, Blewett CJ, Miller JD, et al. Survival after conservative (palliative) management of pleural malignant mesothelioma. J Surg Oncol 2001; 78:171–174 3. Vogelzang NJ, Rusthoven JJ, Symanowski J, et al. Phase III study of pemetrexed in combination with cisplatin versus cisplatin alone in patients with malignant pleural mesothelioma. J Clin Oncol 2003; 21:2636–2644 4. Sugarbaker DJ, Flores RM, Jaklitsch MT, et al. Resection margins, extrapleural nodal status, and cell type determine postoperative long-term survival in trimodality therapy of malignant pleural mesothelioma: results in 183 patients. J Thorac Cardiovasc Surg 1999; 117:54–63 5. Flores RM. Induction chemotherapy, extrapleural pneumonectomy, and radiotherapy in the treatment of malignant pleural mesothelioma: the Memorial Sloan-Kettering experience. Lung Cancer 2005; 49(suppl 1):S71–S74 6. Flores RM, Akhurst T, Gonen M, et al. Positron emission tomography defines metastatic disease but not locoregional disease in patients with malignant pleural mesothelioma. J Thorac Cardiovasc Surg 2003; 126:11–16 7. Richards WG, Godleski JJ, Yeap BY, et al. Proposed adjustments to pathologic staging of epithelial malignant pleural mesothelioma based on analysis of 354 cases. Cancer 2010; 116:1510– 1517 [Erratum in Cancer 2010; 116:2503] 8. Pass HI, Temeck BK, Kranda K, et al. Preoperative tumor volume is associated with outcome in malignant pleural mesothelioma. J Thorac Cardiovasc Surg 1998; 115:310–317 9. Liu F, Zhao B, Krug LM, et al. Assessment of therapy responses and prediction of survival in malig-

AJR:198, February 2012


CT After Extrapleural Pneumonectomy nant pleural mesothelioma through computer-aided volumetric measurement on computed tomography scans. J Thorac Oncol 2010; 5:879–884 10. Lee HY, Hyun SH, Lee KS, et al. Volume-based parameter of (18) F-FDG PET/CT in malignant pleural mesothelioma: prediction of therapeutic response and prognostic implications. Ann Surg Oncol 2010; 17:2787–2794 11. Nowak AK, Francis RJ, Phillips MJ, et al. A novel prognostic model for malignant mesothelioma incorporating quantitative FDG-PET imaging with clinical parameters. Clin Cancer Res 2010; 16:2409–2417 12. Plathow C, Klopp M, Thieke C, et al. Therapy response in malignant pleural mesothelioma: role of MRI using RECIST modified RECIST and volumetric approaches in comparison with CT. Eur Radiol 2008; 18:1635–1643 13. Herndon JE, Green MR, Chahinian AP, Corson JM, Suzuki Y, Vogelzang NJ. Factors predictive of survival among 337 patients with mesothelioma treated between 1984 and 1994 by the Cancer and Leukemia Group B. Chest 1998; 113:723–731 14. Curran D, Sahmoud T, Therasse P, et al. Prognostic factors in patients with pleural mesothelioma: the European Organization for Research and

Treatment of Cancer experience. J Clin Oncol 1998; 16:145–152 15. Edwards JG, Abrams KR, Leverment JN, Spyt TJ, Waller DA, O’Byrne KJ. Prognostic factors for malignant mesothelioma in 142 patients: validation of CALGB and EORTC prognostic scoring systems. Thorax 2000; 55:731–735 16. Chen X, Zhang J, Wu D, et al. HIPAA’s compliant auditing system for medical imaging system. Conf Proc IEEE Eng Med Biol Soc 2005; 1:562–563 17. [No authors listed]. Nutritional anemias: report of a WHO scientific group. World Health Organization Tech Rep Ser 1968; 405:5–37 18. Heelan RT, Rusch VW, Begg CB, Panicek DM, Caravelli JF, Eisen AC. Staging of malignant pleural mesothelioma: comparison of CT and MR imaging. AJR 1999; 172:1039–1047 19. Truong MT, Erasmus JJ, Marom EM, Munden RF. Imaging evaluation in the diagnosis and staging of malignant pleural mesothelioma. Semin Roentgenol 2004; 39:386–396 20. Truong MT, Marom EM, Erasmus JJ. Preoperative evaluation of patients with malignant pleural mesothelioma: role of integrated CT-PET imaging. J Thorac Imaging 2006; 21:146–153 21. Pilling J, Dartnell JA, Lang-Lazdunski L. Inte-

grated positron emission tomography-computed tomography does not accurately stage intrathoracic disease of patients undergoing trimodality therapy for malignant pleural mesothelioma. Thorac Cardiovasc Surg 2010; 58:215–219 22. Wilcox BE, Subramaniam RM, Peller PJ, et al. Utility of integrated computed tomography-positron emission tomography for selection of operable malignant pleural mesothelioma. Clin Lung Cancer 2009; 10:244–248 23. Sørensen JB, Ravn J, Loft A, Brenøe J, Berthelsen AK; Nordic Mesothelioma Group. Preoperative staging of mesothelioma by 18F-fluoro-2-deoxyD-glucose positron emission tomography/computed tomography fused imaging and mediastinoscopy compared to pathological findings after extrapleural pneumonectomy. Eur J Cardiothorac Surg 2008; 34:1090–1096 24. Nakas A, Black E, Entwisle J, Muller S, Waller DA. Surgical assessment of malignant pleural mesothelioma: have we reached a critical stage? Eur J Cardiothorac Surg 2010; 37:1457–1463 25. Gordon GJ, Dong L, Yeap BY, et al. Four-gene expression ratio test for survival in patients undergoing surgery for mesothelioma. J Natl Cancer Inst 2009; 101:678–686
