Diffuse Malignant Mesothelioma of Pleura

Document Sample
Diffuse Malignant Mesothelioma of Pleura Powered By Docstoc
                                  ~- co~7’right law (Title 17 U,S. Code)

                          Diffuse Malignant Mesothelioma of Pleura
                                          Diagnosis and Survival in 92 Cases

                                                                            JAMES R. JETT, MD,§
                       DUANE M. ILSTRUP, MS,~ AND PHILIP E. BERNATZ, MDI[

                L-]inical, radiographic, surgical, and pathologic findings and survival in 92 patients with diffuse malignant
                mesothelioma (DMM) of the pleura who were examined at the Mayo Clinic between 1950 and 1980,
                were studied retrospectively. With the use of defined criteria and ordinary tissue stains, the 92 cases were
                classified into the following histologic subtypes: purely epithelial, 42 cases; mixed, 29 cases; and sareo-
                matous, 21 cases. Eight of the cases were desmoplastic. Median survivals were 12, 5, and
                3 months for the patients in the epithelial, mixed, and saxcomatous groups, respectively. Survival was
                significantly longer for patients with epithelial DMM. Women survived longer than men but more often
                had epithelial DMM. Early disease manifested as multiple discrete pleural nndules,.predominanfly on
                the parietal pleura. However, nine patients had a dominant mass. Radiographic signs especially suggestive
                of DMM were nodular pleural thickening, irregular thickening of interlobar fissures, a dominant mass,
                er decreased volume of the affected hemithorax.
                                                      Cancer 58:1540-1551, 1986.

 ~ECAUSE DIFFUSE MALIGNANT MESOTHELIOMA better survival in patients with epithelial DMM than i.
 ~ (DMM) has various m.icro~copic appearances and those with other subtypes.6"x!xt~ One institution has
 most pathologists encounter few cases in a lifetime, theported better survival in patients with sareomatous me
 diagnosis is considered difficult. Recently, a study of au-sothelioma,tt’|2"t~ This finding may well be due to th:
 topsy cases convinced us that the problem was not that inclusion of the "giant" pleuraLsarcomas,!6 which are th
DMM was undiagnosable light microscopy but that       malignant form of so-called benign localized fibroc.
 the existing light microscopic descriptions of the disease mesotheliomastT,~s and are not associated with asbest0.
 had not been reduced to the essential practical diagnostic exposure. We concur with Ratzer et aLtt that althoug2’
criteria. ~                                                 these tumors are customarily called mesotheliomas,
   Asis the ca~ with bone tumors, the gross pathologic      might best be considered pleural fibrosarcomas.
features of DMM in the living patient are better assessed      In this study, we (1) applied recently published criteri2
with radiographs than by examination of excised tissue. for the histologic diagnosis of DMM to a large clinic-,!
Likewise, the extent of diseas~ may be better seen in situ series, (2) studied the gross morphologic features ofearb
through the thoracotomy wound, although a tumor-oblit-      DMM as seen surgically and radiographicaIly, and
erated pleural space would negate this premise. Not widely correlated histologic patterns with patient survival.
appreciated are tumors with the microscopic appearances
of DMM that macroscopically are "’pseudolocalized."2"4                         Material and Methods
   The median survival of patients with DMM is ap0rox-
imately 12 m~nths.5-~2 Some institutions have reported        A broad search of the files of the Mayo Clinic Ti~uc
                                                                           Registry from 1950 to 1980 for cases ofall pleural tumor~
                                                                           with a diagnosis other than carcinoma and for carcinoma’
  From the Departments of tPathology, :[:Diagnostic Radiology, §Internal   cross-referenced as mesotheliomas identified 212 sttch
Medicine, ~Medical Statistics and Epidemiology, and ¶Surgery, Mayo         cases. The 48 cases in which microscopic slides from other
Clinic and Mayo Foundation, Roch.ester, Minnesota.
   * Current address: Ol~ce of the Chief Medical Examiner and De-          institulion.s had been reviewed and returned were excludN
partmeot of Pathology, University of Massachusetts Medical Center.         from the study. Microscopic slides from the 164 re~ainio.c
Worcester, Massachusetts.                                                  cases were reviewed. Five cages were excluded from the
  Address for reprints: Krishnan K. Unni, MB, B$, Mayo Clinic, 200
First Street SW, Rochester, MN 55905.                                      study because the tissue consisted of badly crushed needle
  Accepted for publication November 29, 1985.                              biopsy specimens that were thought to be inadequate for


                          PLEUKAL DIFFUSE MALIGNANT MESOTHI~i-IOMA ddams e~’al.                                                   tS41

 agnosis. The 67 cases of disease processes oiher than              TABLE I. Symptoms at Initial Presentation in 90 E\’altl’abl¢
                                                                           ¯ Cases of Diffuse Malignant Mesothelioma
 MM included 26 cases of metastatic pleurae: carcino-
 atosis, 18 cases of benign localized fibrous mesotheli-               ¯ Symptom                          No. of cases
ha; 10 pleura1 fibrosareomas (malignant fom~." oflocal- Pain                                                  62
:d fibrous mesothelioma), 1 sclerosing bemangioma, 3            Nonpleufitic                                  56
 determinate sarcomas, 2 metastatic sarcoma.5., I malig-         Pleudtic                                       6
 mt lymphoma, 1 indeterminate small cell neoplasm, 1 S.hortness of breath .                                   53
dmtmary blastoma, and 4 benign effusions. On the basis F¢~er, chilis, or sweats                               30                    33
"previou.sly described diagnostic criteria   t and additional W~ikness, fatigue, or malaise
atures (described below), 92 of the 164 cases were iden-      Weight loss                                     22                    24
                                                              Anorexia                                         10                   II
ied as DMM of pleura. These 92 cases form the bases           Sensatinn of heaviness or
this study.                                                        fullness in chest                            6                    7
 The pathologic diagnosis was made on the basis of or- Hoarseness                                               3                    3
amy tissue sections stained with hematoxylin and eosin. Early satiety
 a few cases, mucicarmine and diastase-digested periodic      Others*                                        1 each                   I
.id2Schiff stains were used to evaluate cytoplasmic vac-
                                                                 * Other symptoms included aphonia and dysphagia, abdominal dis-
4es. Clinical and cytologic findings were abstracted from     tention, sensation of pressure in tight upper quadrant, nausea, ktd tasle
 : patients’ histories. Original radiographs, which were in mouth, perceived tachycardia, and headache.
 ailable for cases from 1974 to 1980, were reviewed. Fol-
w-up data were obt.ained in cases in which previous           Diagnostic Evaluation
 urine statistical inquiries had not identified the date of
 ath. Patient survival was estimated with the method of          Radiographic studies for a primary tumor had been
aplan and Meier.t9 Survival curves were.compared with         undertaken in some cases when the results of cytologic
e log-rank test.2°                                            tests or needle biopsy studies were equivocal. Such
                                                            searches focused on the presenting symptoms and signs;
                                                            thus, it was necessary, for instance, to consider mu~cu.
                        Results                             loskeletai disorders or degenerative joint disease of the
                                                            shoulder. Bronchoscopy, performed in 27 eases, revealed
 °inical Data
                                                            signs of extrinsic compression in 17 cases, but the results
  Of the 92 patients, 71 (77%) were men and 21 (23%) were uniformly negative for tumor.
 :re women. The ages of the women ranged from 32 to            Metastatic lesions were seldom detected radiographi.
 I years (mean, 60 years). The ages of the men ranged cally at the time of initial examination. Skeletal radio.
 )m 28 to 77 years (mean, 59 years). The tumor was graphs showed metastatic lesions in only two cases, and
  .ht-sided in 51 patients (55%), left-sided in 38 (41%), liver metastatic lesions were demonstrated in fo,, r eases
 d bilateral in 3 (3%). Although asbestos exposure was      with ~arious methods. A thyroid biopsy demonstrated in-
~cumented in 24 patients (26%), the actual number was filtrating DMM in one patient, and a biopsy of a patho-
 obably higher because no inquiry regarding exposure        logic fracture showed metastatic disease in another patient.
as documented in the records of half the patients and
e study included cases accrued before the association        TABLE 2. Findings at Initial Physical Examination in 87 Evaluable
  mesothelioma" with asbestos inhalation was known.                        Cases of Malignant Mesothelioma
a~ong the 53 cases accrued during the last decade of the              Finding                          No. ofcases                   g,
ady (1971-1980), 16 had documented exposure to.
’stos. Fifty-two patients smoked cigarettes, 36 did not Signs of pleural effusion                            69                      79
                                                            Decreased chest excursion                         !3                      I
rake cigarettes, and 4 had no smoking history recorded.     Palpable lymph nodes                             12                      14
 The frequency of presefiting symptoms (Table 1) None                                                        10                       II
:1 not differ appreciably from the data of previous re- Palpable liver                                         9
                                                            Chest tenderness                                   9
.its. 5-8,10-12.21-25                                      Clubbing                                            5                       6
 Physical examination at presentation usually revealed      Pleural or pericardial rub                         3
;ns of pleural effusion. Other signs were present in a Cachexia or muscle wasting                             2                        2
inority of cases (Table 2). After the initiai symptoms Others*                                                2                        2
                                                                                                           I each
~d l~hysical signs, signs and symptoms indicative ofpr0-
:ssive disease developed ’in some patients. These in-         * Other findings included vocal cord paralysis, cyanosis, rih fi’a,l* I,"
                                                           limited inspiratory ability, tipped head and drooped shoulth.r, lat t:~l
aded signs of diaphragmatic paralysis, vocal cord         .edema, pitting edema of legs, distended neck veins, lupus skin th: g,c~.
lysis, and chest wall nodules.                             and tachycardia.

1542                                                       CANCER October 1 1986                                                           Vol. 5

                               Radiographic Hndings by Histologie Type in 42 Patients With Diffuse Malignant Mesothelioma

                                                                         Histologlc type (no. of patleats*)


                  Finding                         Epithelial             Mixed                High-grade            Desmoplaslic       Total
       Nodular pleural thickening                      7                    8                     I                       2             18
       Irregular thickening of fissure                 5                    5                     1                      ¯I             12
       Localized mass                                  3                    3                     o                      .0              6
       Loss of volume of hemithorax                    2                    2                     1                       1              6
       Blunted eostophrealc angle                      7                   10                     3
       Free pleura] fluid "                           12                    6            "’                              0         4   21
       Bone destruction                                0                    1                      I                     2              4
  * Patients oRen had mon: than one radiographic finding.

Radiographic Findings:;.                                      Another characteristic radiographic manifestation
                                                           the diffuse nodular pleural thickening caused by DMk
   The chest radiographs of 42 patients were available for was irregular thickening of the intedobar fissures in l:
review. Findings suggestive of DMM and nonspecific         patients (Fig. 2). This was du~ to involvement of the
findings were described (Table 3).                         ceral pleura by the mesothelioma. Although free pleur:~
   Signs suggestive of DMM: Nodular thickening of the fluid, loculated pleurai fluid, and pleural thickening du,
pleura caused by the DMM was present in 18 patients.       to fibrosis can involve the intedobar fissures, they almos
The pleural thickening usually ranged from 5 to 15 ram, invariably cause smooth thickening of the fissures ratht.
but it occasionally reached a thickness of 20 to 25 ram. than the irregular or nodula~ thickening produced by m:.,
In some cases, this finding was obscured by free pleural   iignant mesoflaelioma.
fluid until some of the fluid was removed with thoracen-      In six patien|s, DMM presented radiographically as
tesis (Figs. 1A and 1B).                                   localized mass (Fig. 3). In three ofthese patients, the mas


   Flog: I A AND lB. (A} Posteroanterior view shows right pleural
effusion and nodular pleural thickening along right upper, chest wall.
(B) Four days later, localized posteroamerior view obtained after
thoracentesis shows nodular pleural thickening (arrows) and irregulnr
thickening of minor fissure (arrowheads).

                               PLEURAL DIFFUSE MALIGNANT MESOTHELIOMA                            Admns eta/.                            : 1543

                                                                           FIG. 3. Posteroanterior view s~ow~ localized pleural mass,(arrow) as-
                                                                        sociated with nodular pleural thickening (arrowhead) and blunting of
                                                                        left costophrenic angle,

 F~G. 2. Laleral dew shows nodular pleural thickening along anterior
~¢st wall {arrow). irregular thickening of major ~ssurr (arrowheads),
   blunting of right posterior costophrcnic angle.                      brosis. Two ofthe patients had uncalcified pleural’plaques;
                                                                        one of these two patients also had calcified pleural plaques.
                                                                           Findings suggestive of metastatic carcinoma: Radio-
as associated with loculated fluid or pleural thickening                graphic signs that indicate metastatic adenocarcinoma
, the costophrenic angle. In two patients, the mass was                 rather than malignant mesothefioma are hilar adenopathy,
¯ sociated with noduIar pleural thickening; two patients                nodular puhnonary metastatic disease, and bilateral dis-
,fl~ a localized mass also had free pleural fluid.
 Diffuse thickening of the pleura produced decreased
,lume of the affected hemithorax in six patients (Fig. 4).
 ~casement of the lung prevented full expansion of the
 ng on inspiration. The decreased volume was not due
  bronchial obstruction caused by medial extension of
e tumor in any of the cases.
 Occasionally, thoracentesis resulted in a persistent hy-
opneumothorax. The ¢ncasing tumor prevented reex-
 nsion of the involved lung and thus produced a "trapped
¯ ~g." This occurred in three patients. The nodular thick-
 lug of the visceral or parietal pleura or both was well
 rnonstrated by the pneumothorax (Fig. 5).
 Nonspecific signs of DMM: Blunted costophrenic angles
 ticative of thickened pleura or loculated fluid were
 :znt in 25 patients (Fig. 3). Free pleural fluid was pres-
 : in 21 patients.
 !nvasion ofthe ribs was visible radiographically in four
 :ients. The invasion was best seen on localized views
 the ribs or on computed tomograms, but occasionally
 ~’as visible on the chest radiographs.
 ~igns of asbestos exposure: Radiographic evidence of
 estos exposure was very unusual. None of the 42 pa-        [~IG. 4. P0stcroanterior v~ew shows decreased volume of left hemitho-
 ~ts had radiographic evidence of diffuse pulmonary fi- tax, enca.semcnt of left lung, and nodular pleural thickening (arrowheads).

 1544                                                 CANCER October 1 1986

                                                                      remaining patients, malignant disease had been diagnosed
                                                                      in 48%, atypical cells had been reported in 28%, and a
                                                                      benign diagnosis had been rendered in 24%. An unequiv-
                                                                      ocal diagnosis of mesothelioma had been made in onl.~
                                                                      10%. Inflammation and fibrosis were frequently men-
                                                                      tioned in the patients with benign and atypical diagnoses.

                                                                      Open Pleural Biopsy
                                                                        Open pleural biopsy was done in 75 ofthe 92 patients.
      4                                                              At the time ofopen pleural biopsy, a pathologic diagnosis
                                                                      of mesothelioma had been unequivocally made in 77~
                                                                     of the patients. A malignant disease had been diagno~xl
                                                                     in 15% Of the patients, and mesothelioma had been s’ug-
                                                                     gested but not diagnosed unequivocally. The original di-
                                                                     agnoses for five mesotheliomas ~vere benign fibrosis (two
                                                                     cases), atypical inflammatory process (one case), and
                                                                     malignant fibrous histiocytom~i (one case). For two meso-
                                                                     theliomas, the original diagriosis was es~. ntially mesothe.
                                                                     iioma versus adenocareinoma. Sarcomatous DMM, orig-
   FIG. 5. Localized view shows hydropneumothorax ("trapped iung")
 and nodular pleura! thickening (arrows) after thoracentvsis.        inally diagnosed as a fibrosar’coma, developed in a ’60.
                                                                     year-old woman 13 years after: irradiation of ipsilaleral
                                                                     breast carcinoma. She died 4 months after biopsy of the
 crete pleural masses. None of these findings occurred in            mesothelioma.
 our patients.
    On the basis of the above considerations, the findings           Gross Description at Open Biopsy
 in 20 of the 42 patients with available ?adiographs sug-          Of the 75 patients who had had open biopsies, 5 had
gested a i:liffuge malignant mesothelioma r~ther than me-       the tumor sampled through a limited thoracotomy inci-
tastasis to the pleura. In the 22 other patients, the radio-    sion. In these patients, the chest cavity was not inspected.
graphic appearance was considered indeterminate (that           often because of obliteration of.the pleural space. Seven
is, nonspecific for either malignant mesothelioma or            biopsies wire done at other institutions, and no operative
pleural metastatic disease). In none of the patients was        notes were available. Thus, surgical descriptions were
the radiographic appearance considered specific for pleural     available in 63 patients.
metastasis.                                                        In 46 patients, multiple pleural nodules or masses were
   Correlation with histologic type: The radiographic find-     described. Parietal and visceral pleurae were separ.alel)
ings in the 42 patients with DMM were tabulated ac-             described in 20 of these patients; in 19, the nodules were
cording to histologic typ~ (Table 3). Analysis with a chi-      larger or more numerous on parietal pleura. Most striking
square test showed no statistically significant difference     wa.~ the report of a dominant mass surrounded by nu-
in the distribution of the radiographic findings among the      merous scattered smaller nodules in 8 of the 46 patients
histologic types of DMM.                                       (Fig. 6). The largest such mass was estimated to be 12 to
                                                                15 cm in diameter. In one patient, a pedunculated 4-cn~
Cytologic Study of Pleural Fluid
                                                               mass surrounded the phrenic nerve as it entered the dia-
   Cytologic repoi-ts were available for 69 patients. The phragm. Satellite nodules were present on the nearby dia-
results were negative for malignant cells in 38 patients phragmatic pleura within a loculated effusion. Elsewhere.
(55%), aiypical in 9 (13%) (mesothelioma was suggested the pleura was normal. In other patients, the dominanl
in 4 of tfiese patients), and po.sitivb for malignant cells in mass was located in the major fissure (one patidnt) and
22 (32%.). Among these 22 patients, in only 7 had me- in the pericardiodiaphragmatic sulcus (one patient). The
sothelioma been suggested by the pathologist, and in 4         remainder of the case records de~ribed pleural thickening
adenoc~rcinoma cells had been reported.                        or pleural fusion without elaboration.

Needle Biopay                                                        Treatment
   The results of needle biopsy were available in 52 pa-               Ofthe 92 patients, 33 (36%) received radiation therapY.
tients: in 2, only skeletal muscle was present; in the 50            30 (33%) received irradiation and chemotherapy, and IS

~O..7                       PLEURAL DIFFUSE MALIGNANT MESOTHELIOMA                      Adam~’et al.                              1545

20%) received chemotherapy only. Among patients,
¯ who received irradiation, 5 who received irradiation and
¯ hemotherapy, and 6 who received only chemotherapy
tad limited resection of tumor in addition to olSen biopsy.
.his procedure usually partial decortieation. Eleven
.afients received no chemotherapy or radiation therapy.
"wo ofthese had some debulking of tumor. In the 1950s
nd 1960s the mo~ frequent treatment was intrapleural
 itrogen mustard and-irradiation of the diseased hemi-
mrax. By the late 1970s, radiation was being used less
-equently. Multiple-agent intravenous chemotherapy liad
.ecome the most frequent treatment near the end of the
970s. More recently, pleural pneumonectomy has been
 sed for selected patients, but the limited numbers do not
 ermit valid conclusions. Morbidity and early mortality
1 flits small group have been discouraging.

linical Spread of Disease
 Documentation of clinical metastasis after the time of
agnosis must be regarded as incomplete in this study.            F~G. 6, Diffuse malignant mesothelloma forming paradoxically soli~
2any patients were not examined at the Mayo Clinic             parietal pleuml mass. Fibrous adhesions connected mass to visceral pleura.
                                                               Nodule in lung is actually a tongue of invasive tumor.
"ter the initial diagnosis, and details of follow-up care
h~r than the date of death are la~king. However, the
llowing findings were documented. Tumor growth in a          poor intercellular adhesion with a characteristic "faliing-
oracotomy incision developed in four patients. Growth        apart" pattern was very helpful for recognizing epithelial
 a needle biopsy track occurred in one patient. Abdom-       patterns, especially in early cases or when the nuclear cy-
al metastatic lesions developed in three patients. Brain     toplasmic ratio was somewhat high. Nuclear cytoplasmic
eta.static disease was clinically diagnosed in two patients. ratios were higher when even minimal crush artifact was
~.lpable involvement of the brachial plexus, chest wall,     present. Diagnostically useful features of epithelial DMM
ine, axilla, and groin wa~ present in one patient each.      not illustrated in our prior autopsy study are shown in
aopsies                                                      Figures 7 through 9. Cytoplasmic vacuoles stained with
                                                             mucicarrnine and diastase-digested periodic acid-Schiff
 Details of autopsies were available in 16 cases. Ten of     were evaluated in a few cases in which the tissue consisted
ese cases were included in a prior study) Slides and         ofsmall needle biopsy specimens. Neutral mucopolysac-
otocols from five of the six remaining autopsies were charides were not demonstrated in any case.
dewed. With the exception of a solitary brain metaslatie        Twenty-one tumors showed only sarcomatous differ-
,ion in a case of sarcomatous DMM, the findings con- entiation: 9 ofthese were low grade (that is, desmoplastic),
7ned to those of the prior study.                           and the 12 others were obviously malignant by cytologic
                                                            criteria. Of the nine desmoplastic tumors, two were not
;stologic Patterns                                          diagnosed as being malignant until autopsy.
 On review, tubulopapillary, epithelioid, and myxoid           The 29 tumors in the mixed category had both sarco-
~es of epithelial differentiation were encountered, as      matous and epithelial components, including epithelioid,
¯ re high-grade and desmoplastic types of sarcomatous       tubu’lopapillary~ myxoid, high-grade sarcomatous, and
 tErentiation. In many cases ofepithelial differentiation, desmoplastic areas in various combinations.
: lumor was at an early stage and could not be consis-         In addition to mesothelioma, fibrous pleural plaques
.tly labeled as tubulopapillary or epithelioid. Therefore, were identified in the tissue sections of five cases. DMM
  tabulation of the data, all types of epithelial differen- had grown over the surfaces of the plaques and infiltrated
:ion were listed only as epithelial.                        some of then].
Of the 92 tumors. 42 manifestedpurely epithelial dif-
-~ntiation. In most cases, the tumor had been sampled       Sltrt,iva[
’.he stage of multiple small nodules. In addition to the       Follow-up was virtually complete. Of the 92 patients,
.viously published diagnostic criteria,~ the presence of    88 were dead, 2 were alive at 5 and 7 years after diagnosis,

1546                                                           CANCER October          1 1986                                                 VoL 5

                                                                                    and 2 were lost to follow-up. The median survival tim,
                                                                                    for all 92 patients was 242 days; the 3-year survival prob
                                                                                    ability was 10%, and the 5-year survival probability wa
                                                                                    3% (Fig. 10). The differences in survival among patient
                                                                                    with purely epithelial DMM and those with each of
                                                                                    other histologic categories of DMM werp statistically sig
                                                                                    nificant (P < 0.05). Survival differences between patient:
                                                                                    with mixed DMM and those with sarcomatous DMM
                                                                                    were not significant. There was essentially no differenct
                                                                                    in survival between patients with high-grade sarcomatou~
                                                                                    DMM and those with desmoplastic DMM (Table 4).
                                                                                       Patients with epithelial DMM had the best survival.
                                                                                    The median survival probability was 1.1 years for patients
                                                                                    with epithefial DMM, 0.7 year for those with mixed
                                                                                    DMM, and 0.3 year for those with sareomatous DMM.
                                                                                    The l-year survival probability was 51% for patients with
                                                                                    epithelial DMM, 27% for those .with mixed DMM, and
                                                                                    19% for those with sarcomatous DMM. The most stalking
                                                                                   difference was in the 2-year survival probability--it was
                                                                                    32% for patients with epithelial DMM, 8% for those wilh
                                                                                   mixed DMM, and 5% for those with sareomatous DMM
                                                                                   (Table 5) (Fig. 11).
                                                                                       When the data were analyzed by gender, women were
                                                                                   found to survive longer than men (P < 0.03) (Table 6).
                                                                                   Further analysis demonstrated that this difference could
                                                                                   be explained by the greater percentage of women.with the
   FIG. 7. Myxoid diffuse malignant mesothel.ioma with microcystic ap-
pearance. Archipelagic pools of noncollagenous matrix possess incomplete            favorable epithelial type: 76% of the women but only 3-7°Z
lining of tumor cells. This feature is rarely seen with carcinoma. "Papillae"      of the men had epithelial DMM (Table 7). Furthermore,
with fibrous cores seem to be created passively by expansion of pools              the survival of women with epithelial DMM was not sig-
into areas of least resistance. Outside the pools, tingle tumor cells infiltrate
preexisting connective tissue and dilute it with liquid matrix (H & E              nificantly different from that of men with the same his-
x~).                                                                               tologie t)q:~z-s of disease (P = 0.38) (Table 8).

                                                                                                             FtG. g. Epithelial diffuse malignant me-
                                                                                                          sothelioma, ~ady lesion. Noncohesive tu-
                                                                                                          mor cells have proliferated under s~rosa.
                                                                                                          Much of surface me,sothdium is rather flat.
                                                                                                          but some surface cell~ are plump and appear
                                                                                                          identical to underlying tumor cells. This is
                                                                                                          an important feature. Notice characteristic
                                                                                                          loose "’falling-apart" appearance generated
                                                                                                          by noncollagenous tumor matrix dissecting
                                                                                                          between cells (H & E Xl00).

                               PLEURAL DIFFUSE MALIGNANT MESOTHELIOMA                ¯ Adams a aL                                1547

    FtG. 9. Epithelial diffhse malignant me-
  ,th¢lioma, tubulopapilla~" patteCn, early
 :~tce lesion. Notice papillary sud’ace and
Jcntity of infiltrating tumor cells with sur-
 ...-~ cells on adjacent serosa. "Ribhohing
 attem is vaguely apparent within this nod-
.ie because of characteristically loose tumor
 :Its between tubalopapillary configurations
!~ & E ×64).

  Analysis by pleural laterality did not demonstrate any        to demonstrate invasion of the chest wall, ~’ertel~i’ae, and
 arvival difference between patients with left-sided DMM        mediastinum.26
 nd those with right-sided DMM. Two of the three pa-               Before biopsy, radiographic appearances can be quite
.ents with bilateral pleural DMM survived more than 2           helpful in the clinical distinction of DMM from metastatic
 ears, but analysis was not possible. Nine patients survived    adenocarcinoma. Some of the findings produced by DMM
 years or more: seven had purely epithelial types of DMM        are nonspecific, such as ple.ural fluid (either free or loe-
 me of whom had DMM manifesting as a solitary mass),            ulated), a localized mass, and chest wall invasion, but
 ne had mixed DMM, and one had sarcomatous DMM.                 other findings are very suggestive of DMM, such as diffuse
-hese nine patients were examined throughout the 3 de-          nodular pleural thickening, irregular thickening of an in-
ades ofthe study. There was no obvious correlation with         terlobar fissure, and decrease in the volume of the involved
 ~e method of treatment in this retrospective review.           thorax.2~ Although diffuse pleural metastasis from carci-
  The patient alive at 5 years was a 32-year-old woman          nomas can occasionally produce diffuse nodular pleural
’ith systemic lupus erythematosus and C2 deficiency. The        thickening, it usually manifests as bilateral discrete pleural
atient alive 7 years after diagnosis was a 68-year-old          masses. More importantly, metastasis rarely catises irreg-
¯ ’oman. Both patients had epithelial types of DMM.

~adiographic b~terpretation
  The diagnosis of DMM is sometimes suggested by the
~nstellation of clinical and radiographic findings. The-                     \                ".
iagnosis, however, is made by microscopic examination
f tissue. With this in mind, the radiographic findings
set’ul in the workup are discussed.
  Radiographic abnormalities produced by DMM are
sually well seen on posteroanterior and lateral views of
~e chest. Lateral decubitus views and localized views of
~e ribs are.helpful to detect free pleural fluid and invasion            0              2              4              6
1" the ribs, respectively.                                                                          Years
 Computed tomography is helpful to demonstrate en-                FIG. 10. Overall survival probability curve of patients with diffuse
asement of the lung by DMM and it is especially useful          malignant mesothelioma.

                                                            CANCER October 1 1986                                                           Vol. 5

   TABLE 4. Statistical Significance of Differences in Survival Among            10 .~t
           Patients With Diffuse Malignant Mesothelioma
                      According to Histologic Ty.~e
                                                                                                                        Epithehal 1n=42|
                                                         Difference in
                                                       survival of patlent   ~                               ~          Mixed and Sarcornatous
          Histologlc types compared                        (P value)         ¯ ~ ~o       ’;                                  (n=so~
  Epithelial vs. sar~omatous and mixed                      0.0001
  Epithelial vs. ~arcomatous                                0.0001
  Epithelial vs. mixed                                      0.0029
  Mixed vs. sareomatous                                     0.2147
  Sareomatou~, high-grade vs. d~moplasti¢.                  0.8087
                                                                                  0                    ~ ~ ..........
                                                                                      0          2                 4
 ular thickening of a fissure or decreased volume of a lung                                                   Years
 due~o encasement of the lung by metastatic neoplasm.
                                                                              FIG. l ]. Su~ival pro~Si]ity cu~= of ~tlen~ ~
    ExtenSive involvement ofthe pleura in one hemithorax                     m~i~aat m~the,oma a~ of th~ ~th ~l other
 while the other hemithorax is uninvolved is another clin-
 ical finding that indicates DMM rather than metastatic
 adenocarcinoma. Alternatively, because bilateral in-                     influence the pathologist. As pointed out by Wright ~
 volvement of the left mid right.pleurae by DMM is unusual                aL,32 a positive history may wrongly bias the pathologis
 (3 of 92.cases in this sedes), extensive involvement of the              toward a diagnosis of DMM. The diagnosis should &
 pleura in both hemithoraees favors the diagnosis of met-                 based on histologic features.I"
 astatic adenoearcinoma. The distinction between pleural                     Desmoplastic DMM versus benign inflarnmatory fibro
 extension of an occult primary lung tumor and pleural                    sis of pleura and ordinary pleural plaque: With this tyo
 mesothelioma must be made microscopically.                               of diagnostic problem, the cells ofdesmoplastie DMM d,
                                                                          not appear malignant. However, some clues can be use.,
  Microscopic Interpretation                                             to guide the surgeon. If the surgeon describes the tissu
                                                                         as a fibrous peel, the pathologist generally will find benigl
     Most. of our microscopic observations are not new.
                                                                         fibrosis rather than desmoplastie DMM. Desmoplasti,
  However, categorization ofthese microscopic observations DMM invades the loose areolar tissue ofthe endothoraci,
  into diagnostically useful criteria and diagnostically useless         fascia and infiltrates the visceral pleura so that a surgiea
  findings has not been previously emphasized. The diag-                 cleavage plane is extremely difficult or impossible 1,
  nostic problems encountered by pathologists in regard to               achieve.33
 DMM are generally one of four types.
                                                                            Once desmoplastic DMM is suspected, the surgeon am
    Adenocarcinoma versus DMM: Two points should be                      the pathologist preparing frozen sections should look fo
 kept in mind. First, adenocarcinoma and DiM have dif- tumor invasion of adipose tissue in the endothoraeie fas
 ferent histologic appearances.. Second, asbestos exposure               cia, intercostal muscles, or lung. Microscopically, invasim
 is associated with an increased incidence of carcinoma, is required to make an unequivocal diagnosis ofdesmo
 principally of the lung.28m S.elikoff et aL,~° in theil: study plastic DM_M.t Fibrosis caused by unrecognized asbestosi
 of 17,800 insulation workers exposed to asbestos, de-
                                                                         may lead to dense attachment to the chest wall but doe
scribed 486 cases oflung cancer and 175 cases of DMM.
                                                                         not invade muscle or lung.
Unlike Whitaker and Shilkin,31 we believe that neither a                    Other features which are helpful but not always reliabh
positive nor a negative history of asbestos exposure should              are the tendency for desmoplastic DMM to be avascular
                                                                         focally necrotic, and relatively free ofinflammation. Thes~
       TABLE .5. Survival by Histologic Type in 92 Patients With         findings are in contrast to the characteristic hypervascu
                   Diffuse Malignant Mesothelioma                       larity, edema, and inflammatory infiltrate of reactive fi
                                                          probability       Isolated benign plaques should not be a diagn0sti
                                          Median                        problem because they are commonly seen and easily rec
  Histolog~¢        No.                  surviva!,
     type         patients      %           yr                          ognized by the surgeon. However, mesothelioma cat
                                                        l-yr       2-yr
                                                                        spread over and focally infiltrate pleural plaques. In a cas:
Epithelial          42         45.7          1. I       51         32   of DMM, it is not unusual for a biopsy specimen I0 con
Mixed.              29         31.5         0.7         27           g
Sarcomatous         21         22.8         0.3                         tain mostly benign plaque. Pleural plaque is identifiabl.
                                                        19           5
                                                                        microscopically by its hypocelluladty and the orientatiot
Total               92        100           0.7         36          18  of collagen fibers in parallel arrays.

                              PLEURAL DIFFUSE MALIGNANT MESOTHELIOMA ¯                               Ad~m$-et ~/.                                   1549
           TABLE 6. Survival by Gender in 92 Patients With                    TABLE 7. Hislologic Type of Disease by Gender in 92 Patients
                 Diffuse Malignant Mesothelioma.                                       With DiffuSe Malignant Mesothelioma
                                                       Survival                                                              Gender
                                      Median         probability (%) .
                 No.                  survival                                                             Male.                           Female
  ~.nder       patientS      %         (yr)         l-yr       2-yr
                                                                                                      No.                           No.
 :ale            71         77          0.7         31         14         Histologic type           patients           %          patiehts
 :male           21         23          1.5         55         J3
                                                                           Epithelial                 26                37            16                 76
                                                                           Mixed                      27                38             3                 14
                                                                           Sa~-omatous                18                25             2                 10
    DMM versus localizedmesolhelioma: Simple a~eness                       Total                      7f               100            21                100
  fthe entity of localized mesothelioma should avert most
  roblems in differentiating it from DMM. The micro-
 :opic appearances of localized mesothelioma are quite                    mined by which pleural ca~.’.ty manifested the earliest or
  lstinctive and are different frtm those of DMM (Fig.                    most disease. Autopsy studiest have shown that small
  2)..The one case in our ~eries of DMM manifesting as a                  nodules of tumor in contral~teral pleura are common late
  ,litary mass without satellite nodules had histologic fea-             -in the natural history of the disease. This finding would
  :res incompatible with a diagnosis of localized mesothe-                be consistent with either of the two hypotheses.
  area. Localized mesotheliomas rarely; if ever, metasta-                    At a practical level, the knowledge that DMM can
 ze, even with microscopic evidence of malignant dis-                     manifest with a solitary lesion or dominant mass under-
 ~se.t~’tS"3~-38 Localized mesotheliomas are notassociated                scores the need for unequivocal microscopic diagnosis.
  ith asbestos exposure and are usually resectable.                       Thus, odd as it seems, in some instances, microscopic
   Reactive hyperplasia versus DMM of pleura: Semisolid                   examination will help decide whether a pleural neoplasm
 gregates of reactive mesothelial cells imd fibrin may                    is of the diffuse or Iocalizedtype.
 atulate DMM on needle biopsy. Overdiagnosis can be
  ,okled by requiring the demonstration of tumor cells                    Survival
 .’neath the serosal membrane (that is, in tissue).
                                                                             Currently, whether any type of treatment alters the
                                                                           usual rapid progression to death in most patients with
                                                                          DMM of pleura is unclear. In an effort to alter this re-
  Despite their popularity, electron microscopy and im=                   lentless progress, pleuropneumonectomy supplemented
.. unoperoxidase stains for keratin and careinoembryonie                  by chemotherapy and radiation therapy has been offered
 atigen have not yet been shown to discriminate reliably                  to selected patients in the absence of metastatic disease
 :tween DMM and other entities. They must therefore                       or obvious computed tom0graphic evidence of medias-
  regarded as promising.experimental investigative tools.                 final or chest wall invasion: Both physicians and patients
 ~e investigative reports dealing with these procedures                   are reluctant to partici~te in randomized which
~en use carefully selected subsets ofthe available clinical               one of the therapeutic regimens is observation 0nly.
 aterial--namely, the easily diagnosed eases. In our ex-                     Our finding of a difference in survival between patients
:tence, diastase-digested ~riodic aeid-Schiff and mu-                    with purely e.pithelial DMM and those with DMM with
~rmine stains are useful (that is, to exclude the diagnosis              a sarcomatous component confirms prior observa-
 DMM) only when they are positive. Stains for acid                       tions?¯ ta’t4 In our study, this difference was quite distinct.
 ucopolysaccharides such as alcian blue and colloidal                    This difference may be due to our exi:lusion from the
~n are useful in theory; in practice, however, they have                 epithelial category ofany case with even focal sarcomatous
aited usefulness because acid mucopolysaccharides tend                   differentiation¯ Series that have reported better survival
 be washed out of the tissue during routine dehydration                  for patients with sarcom.atous DMM~a t’t~ have apparently
td clearing.39-~’

~L~I as a Mass                                                              TAI~LE 8. Survival b.v Gender of 42 Patients With Epi aelial
                                                                                        Diffuse Malignant Mesothelioma
 The finding in a few cases of a dominant mass sur-
unded by satellite nodules suggests one of two possi-                                                          Median                 probability (%)
.ifies: (I) DMM may begin growth as a solitary nodule                                      No.                 survival
d, via lymphatics, rapidly seed adjacent pleura and later                Gender          patients                (yr)             l-yr            2-yr
ntralateral pleura or (2) DMM may begin as bilateral                     Male               26                   1.0              46              3I
ultiple primary nodules, and clinical laterality is deter-               Female             16                   1.5              60              33

  1550                                                 CANCER October I 1986

                                                                                                      FI~. 12. Localized fibrous mesothdior
                                                                                                   Unlike desmoplustie DMM, in which
                                                                                                   is one cell per slit and the slits are genera
                                                                                                   discrete, this tumbr has a distinct appe
                                                                                                   ante characterized by a maze of undulati
                                                                                                   communicating slits popfftated by multii
                                                                                                   uniform nuclei in single file (H & E ×!6’

 included the malignant form of the localized fibrous me-              mesothelioma presenting as a solitary lung mass. Chest 1983; 84:9
 sothelioma, which is an entirely different tumor with no              101.
                                                                          3. Kannersteiu M, Churg J, McCaughey WTE. Asbestos and rues
 relationship to asbestos exposure.3"t7’~8                             thelioma: A review. PatholAnnu 1978; 13(pt 1}:81-129.
     The finding by one British group of no difference in                 4. Suzuki Y. Pathology
                                                                       Onco11981; 8:268-282. of’human malignant mesothelioma. Sen:
 survival among patients with different histologic sub-                   5. Brenner J, Sordillo PP, Magill GB, Golbey RB. Malignant rues
 types42. is at variance with our findings. Differences in the         thelioma of the pleura: Review of 123 patients. Cancer 1982; 49".243"
 histologic criteria for subtyping may well be responsible.            2435.
                                                                          6. Chahinlan AP,
    We confirmed the finding of a better survival in women Mandel EM. DiffusePajak TF, Holland JF, Norton L, AmbinderR[,
                                                                                             malignant mesothelioma: Prospective evaluatk
 with DMM than in men with DMM reported by Antman of 69 patienlr,. Ann Intern Med 1982; 96:746-755.
 and associates.21 On the basis of our finding that most                  7. Hillerdal G. Malignant mesothelioma 1982: Review of 4710 pu
                                                                                       BrJDis Chest 1983; 77:321-343.
 women had epithelial DMM, we attribute the survival lished cases. Spjut HI, Seybold WD. Diffuse malignant mesotheliom.
                                                                         g. Klima M,
 difference between men and women to differences in the               Am J Clin Patrol 1976; 65:583--600.
 proportion ofhistologic subtypes of DMM in each gender                 ¯ 9. Leraer H J, Scho~nfeld DA, Fallc~n G, Borden E. Malignant
 category.43,                                                         sothelioma: The Eastern Cooperative Oneology Group (ECOG)
                                                                      rience. Cancer 1983; 52:i981-1985.
    Our finding of poor survival in patients with desmo-                  10. Oels HC, Harrison EG Jr, Carr DT, Bematz PE. Diffuse malignar
 plastic DMM is in accord with the observations of Cantin             mesothelioma of the pleura: A review of 37 cases. Chest 1971; 60:564
 et al.44 and Kannerstein and Churg.45                                570.
                                                                          I I. Ratzer ER, Pool JL, Melamed MR. Pleural mesotheliomas: Clir.
    The results of this study confirm previous reports of             ical experience~ with thirty-seven patients. Am J Roentgenol 1967;
an overall short survival for patients with DMM and may               863-8~0.
serve as an historical control and aid in the evaluation of               12. \Vanebo H J, Martini N, Melamed MR, Hilaris B, Beatlie EJ J
                                                                      Pleural mesolhelioma. Cancer 1976; 38:2481-2488.
 the survival of patients receiving uniform protocol treat-               13. Butchart EG, Ashcroft T, Barnsley WC, Holden MP. Pleurc
ments.                                                                pneumonectomy in the management of diffuse malignant mesotheliom
                                                                      of the pleura: Experience with 29 patients. Thorax 1976; 31:15-24.
                                                                          14. Gfil~ths MH, Riddell R.J, Xipell JM. Malignant mesotheliom;
                                                                      A re,Se~ of 35 cas~s with diagnosis and prognosis. Pathology 1980;
                           REFERENCES                                 591-603.
                                                                         15. McCarmack PM, Nagasaki F, Hilaris BS, Martini N. Surgi~
   I. Adams VI, Unni KK. Diffuse malignant mesothelioma of pleura:    treatmenl of pleurat mesothelioma. J Thorac Cardiovasc Surg 1982; 8~
Diagnostic cfilefia based on an autopsy study. Am J Clin Patrol 1984; 834-840.
82:15-23.                                                                16. Klempcrer P, Rabin CB. Primary neoplasms of the pleura:
   2. Goffried MH, Quan SF, Sobonya RE. Diffuse epithelial pleuml     repor~ or" five cases. Arch Pathol 1931; i 1:385-412.

¯ u. 7                       PLEURAL DIFFUSE MALIGNANT MESOTI~EI~IOMA Adams ~ ~!.                                                     .. 15~ I

  17. Dalton W’l’, Zollike~ AS, McCaughey WEE, Jacques J, Karma-                32. Wright WE, Sherwin RP, Dickson EA, Burnstein L, Fromm
dn M. Localized primary tumo~ of the pleura: An analy~s of 40 cases.         Hende~on BE. Malignant mesolhclioma: Incidence, asbestos exposure,
~ncer 1979; 44:1465-1475.                                                   and reclassification of histopathology. Br J lnd Meal 1984; 41:39-45.
  18. OkJke N, B~rnatz PE, Woolner LB. Localized o~thclloma of                  33. Adams VI, Jett J.R. Difl’use mafignant mesothdioma: Desmoplastic
; pleura: Benign and malignant variants. J Thorac ~-~’dioras~ Surg          variant~ASCP Check Sample (in preas).
~78; 75:363-370.                                                                34. Herbert A, Gallagher PJ. Pleural biopsy in the diagnosis of ma-
  19. Kaplan EL, Meier P. Nonparametric estimation ~ incomplele             lignantmcsothelioma. Thorax 1982i 37:816-821.
~¢rvations. JAm Star Assoc 1958; 53:457-481. "                                  35. ’Briselli M, Mark El, Dicke~sin GR. Solitary fibrous tumoss of
 20. Peru R, Peru J. Asymptotically efficient rank inw~iant te~ pro-        the pleura: Eight new cases and review of 360 cases in the literature.
-duty’s. J R Stat Soc Series A 1-9724 135: ! 85-207.                        Cancer 198 I; 47:2678-268.9.
 21. Antman KH, Blum RH, Greenbe~ger JS, Flowenkw G, Skar~n                     36. Clagett OT, McDon’ald JR, Schmidt HW. Localized fibrous me-
I’, Canellos GP. Multimodality therapy for malignant n~-sothelioma          sothelioma of the pleura. J Thorac Surg 1952; 24:213-220.
~ on a study of natural history. Am J Med 1980; 68-.$56,362.                    37. Foster EA, Ackerman L’0’:.Localiz~l mesotheGomas of the pleura:
 22. Antman KH. Clinical presentation and natural l~s~o~, ofbenign          The pathologic evaluation of 18 cases. ~m J Clin Patho11960; 34:349-
:d malignant mesothelioma. Semin Oncol 19gl; 8:315--320.                    364.
 23. Edge JR. Mesothelioma. BrdHosp Med 1983; 29".521-536.                   ¯ 38. Schariflcer D, Kaneko M. Localized fibrous "mesothelloma" of
 24. Lagha SS, Muggia FM. Pleural mcsotbelioma: L"l~a~al                    pleura (submesothelial fibroma): A clinicopathologic study of 18 cases.
~d therapeutic implications. Ann Intern Med 1977; 8"/:613-621.              Cancer 1979; 43:627--635.
25. Newhouse ML, Thompson H. Mesolhelioma oflai~ and peri-                     39. Hourihane DO’B. The pathology ofmcsotheliomata and an anal-
neum following exposure to asbestos in the London ~ Br                     ~s oftbeir association wi~h asbe~, us exposure. Thorax 1964; 19:268~
"ed 1965; 22:261-269.                                                      278.
26. Mi~’is S, Dutcher JP, Whitley NO, Aisner J. C’I" of malignant              40. Kannerstein M, Churg J, Magnet D. HistochemisU:y in the
:ural me.sothelioma. A JR 1983; 140:665-670.                               agnosis of malignant mesothelioma. Ann Clia Lab Sci 1973; 3:207-21 !.
27. Fraser RG, Par~ JAP. Diagnosis of Diseases ofd~ Chest; ed. 2,              41. Whitwell F, Rawclifl’e RM. Diffuse malignant pleural mesothe-
1. 3. Philadelphia: WB Saunders, 1979; 1785-1792.                          fioma and asbestos exposure. Thorax 1971; 26:6-22.
28. Kannerstein M, Churg J. Pathology of carcinoma of ~ lung                 - 42. Law MR, Gregor A, Hudson ME, Bloom HJG, Turner-Warwick
~ociated with asbestos exposure. Cancer 1972; 30:|4-21.                    M. Malignant mesothefioma of the pleura: A study of 52 treated and 64
29. Whitwell F, Newhouse ML, Bennett DR. A stilly oflhe histo-             untreated patienB. Thorax 1984; 39:255-259.
"ical cell types of lung cancer in worke~ suffering from asl~stos in the       43. MArtens.son G, Hagmar B, Zcttergren L. Diagnosis and prognosis
titr.d Kingdom. BrJlndMed 1974; 31:298-303.                                in malignant pleural mesothelioma: A pros-p~’~ive study. Eur J Respir
J0. SelikoffU, Hammond EC, 8eidman H. Latency ofasLx~stos disease          Dis 1984; 65:169-178.
long insulation workers in the United States and C~na~L Cancer 1980,           44. Cantin R, AI-Jabi M, McCaughey WTE. Desmoplastic diffuse
.2736-2740.                                                                mesothelioma. Am J Surg Pathol 1982; 6:215-222.
H. Wlfitaker D, Shilkin KB. Diagnosis ofpleural malignaut meso-               45. Kannersteia M, Churg L Desmoplastic diffuse malignant meso-
’lioma in life: A practical approach. JPathol 1984; 143=147-175.           thelioma. Prog Surg Pathol 1980; 2:19-29.