Baseline characteristics
There were 164 patients referred for consideration of enrollment into early phase clinical trials that were evaluable for metastatic disease sites and survival over the time period. They were predominantly male (76%) with a median age at MPM diagnosis of 64 years (Table 1). Available histopathology from reports and referral letters was reviewed in all cases. The subtype of mesothelioma was identifiable in 84% of cases (n=138) and was mostly epithelioid (n=115, 83%), with 9% biphasic (n=12) and 8% sarcomatoid type (n=11) (Table 1).
Table 1: Demographics and histopathological diagnosis
Demographics
|
N (%)
|
Gender
|
Male
|
125 (76%)
|
Female
|
39 (24%)
|
Median age at diagnosis
|
64 years (range 37-90)
|
Histopathology
|
Epithelioid
|
111 (68%)
|
Biphasic
|
12 (7%)
|
Sarcomatoid
|
11 (7%)
|
Other
|
8 (5%)
|
Unknown subtype
|
22 (13%)
|
MPM patients had a median of three imaging modalities reviewed for assessment of metastatic disease ranging from one to five. This was predominantly comprised of CT scans, of which patients had a median of two scans reviewed (range 1-9), and X-rays, of which patients had a median of 1 (range 0-6) and this was almost exclusively a chest radiograph.
Patients who were referred for early phase clinical trials had received a median of two prior chemotherapies, ranging from none to six (Table 2). Five patients were referred directly to the early phase trials unit prior to conventional first line therapies. Of the remaining 159 patients, a platinum-based chemotherapy regimen with pemetrexed was initiated in 71% of cases (n=117) in the first line setting. The remaining patients received alternatives to pemetrexed such as vinorelbine (6, 4%), clinical trial enrolment (21, 13%) or MVP (mitomycin-C, vinblastine and cisplatin) (17, 10%). The majority of patients who presented to the early phase trial unit had undergone a second regimen of anticancer treatment (117, 71%) with half referred for clinical trials for their second line of treatment (58, 50%). Of the remaining 59 patients, almost one quarter were rechallenged with a platinum and pemetrexed (27, 23%). The remainder underwent a variety of second line chemotherapies, most frequent of which was vinorelbine (17, 15%), but also ifosfamide, gemcitabine, MVP and TACE (mitomycin-C, gemcitabine and cisplatin).
Table 2: Prior treatments and interventions
Prior treatments / interventions
|
N (%)
|
Surgery / intervention
|
Thoracoscopy +/- intervention
|
53 (32%)
|
No surgery / intervention
|
37 (23%)
|
Cardiothoracic surgery
|
31 (19%)
|
Other intervention
|
11 (7%)
|
Unknown surgery / intervention
|
32 (20%)
|
First line anticancer treatment
Total n=164
|
Platinum + pemetrexed
|
117 (71%)
|
Clinical trial
|
21 (13%)
|
MVP (mitomycin C, vinblastine, cisplatin)
|
15 (9%)
|
Other chemotherapy regimen
|
6 (4%)
|
Incomplete / missing information
|
5 (3%)
|
Second line anticancer treatment
Total n=117 (71%)
|
Clinical trial
|
58 (50%)
|
Rechallenge platinum and pemetrexed
|
27 (23%)
|
Other chemotherapy regimen
|
26 (22%)
|
Incomplete / missing information
|
6 (5%)
|
Median total treatments prior to referral
|
2 (range 0-6)
|
Site of primary MPM and metastases
Of all MPM included for analysis, 64% (n=106) were right sided with the remaining MPM localised to the left-hand side (n=58, 36%).
The spread of metastatic disease was divided into local and distant sites and are outlined in Table 3. Local spread included pleural effusion(s) documented along the course of the MPM disease in 102 patients (62%). Other local sites included mediastinal and paratracheal lymphadenopathy that was radiologically considered and reported as “pathological” in 101 patients (62%). Chest wall involvement was noted in 68 patients (42%). The pericardium was considered infiltrated in 47 patients (29%) with evidence of a pericardial effusion in 19 of these patients (12% total).
Distant sites of disease are also recorded in table 3. A total of 110 patients (67%) had at least one documented site of distant metastasis with many patients having multiple distant sites diagnosed synchronously or metachronously. They include contralateral lung disease noted in 55 patients (35%) and multiple parenchymal lung metastasis in 42 (26%). In 17 cases which equated to 10% of the total patient cohort and 31% of patients with parenchymal lung metastasis, radiographic appearances were of a diffuse military-type metastatic pattern. Peritoneal disease and/or omental involvement was noted in 36 patients (22%), with ascites evident in 24 patients.
Bone metastases were diagnosed in 31 patients (19%) (Table 3). Of the bone lesions, the majority (18, 60%) were documented as “lytic” on radiological imaging, with 7 (23%) defined as “sclerotic” and 5 (17%) unknown or unremarked upon. Visceral metastases included distant spread to the major intra-abdominal organs including the liver, kidney, adrenal and spleen. Twenty-seven sites of disease were documented in 23 patients (14% of total). Of these visceral lesions, most were located in the liver (19, 70%), three in the kidney, three in the adrenal gland(s) and two patients (8%) had radiologically diagnosed splenic metastases. Only five patients (3%) had brain metastases confirmed radiologically. All these patients had neurological symptoms that prompted imaging. Subcutaneous metastatic nodules distant to the site of MPM and distant to any potential chest drain or biopsy site was noted in 20% (n=32). Seven patients (4%) had distant intramuscular metastasis including gluteal muscle, psoas muscle and distant intercostal muscle deposits. Figure 1 depicts radiological images of distant metastatic sites. On comparison of distribution of sites of metastases by histological subtype, peritoneal metastases were more likely to occur in epithelioid compared to biphasic/ sarcomatoid MPM (frequency of 27.0% in epithelioid and 4.3% in biphasic/ sarcomatoid; p=0.015). Though there were higher rates of bone metastases in biphasic/ sarcomatoid compared to epithelioid subtypes, this did not reach statistical significance (frequency of 18.2% in epithelioid and 34.7% in biphasic/ sarcomatoid; p=0.0936) (Table 4).
Table 3: Site of metastatic disease
|
Location
|
N (%)
|
Local disease spread
|
Pleural effusion
|
102 (62%)
|
Thoracic nodal disease
|
101 (62%)
|
Chest wall involvement
|
68 (42%)
|
Pericardial infiltration
|
47 (29%)
|
Pericardial effusion
|
19 (21%)
|
Distant metastatic sites
(n=110, 67% of total)
|
Contralateral lung disease
|
55 (35%)
|
Parenchymal lung metastasis
|
42 (26%)
|
Peritoneal / omental metastasis
|
36 (22%)
|
Ascites
|
24 (15%)
|
Bone metastasis
|
31 (19%)
|
Visceral metastasis
|
23 (14%)
|
|
Liver
|
19 (70%)
|
|
Renal
|
3 (11%)
|
|
Adrenal
|
3 (11%)
|
|
Spleen
|
2 (8%)
|
Brain metastasis
|
5 (3%)
|
Subcutaneous metastatic nodules
|
32 (20%)
|
Intramuscular metastasis
|
7 (4%)
|
Table 4: Metastatic spread according to histopathological subtype.
Subcut: subcutaneous, Mets: metastases
Pathology
|
N
|
Sites of Metastases
|
|
|
Effusion
|
Contralateral lung mets
|
Bone mets
|
Visceral mets
|
Brain mets
|
Peritoneal
|
Subcut/
muscular
|
Epithelioid
|
115
|
74 (64.3%)
|
38 (33.0%)
|
21 (18.2%)
|
16 (13.9%)
|
3 (2.6%)
|
31 (27.0%)
|
23 (20.0%)
|
Biphasic/ Sarcomatoid
|
23
|
12 (52.1%)
|
9 (34.7%)
|
8 (34.7%)
|
4 (17.4%)
|
2 (8.6%)
|
1 (4.3%)
|
5 (21.7%)
|
Comparison of proportions
(Fishers exact test)
|
P=0.3464
|
P=0.6277
|
P=0.0936
|
P=0.7455
|
P=0.1938
|
P=0.0159
|
P=0.7839
|
Patient Survival
In this dataset, total overall median survival of patients in our dataset was 23.8 months (range 1 month to 106 months) from initial diagnosis of MPM to death (Figure 2a). The precise date of death was known in 122 patients (74% of total cohort), the remainder were discharged from the Drug Development Unit to their primary oncology unit and outcome was unknown. Epithelioid histological subtypes had a statistically significant better overall survival than biphasic and sarcomatoid subtypes (p=0.004) (Figure 2b).
The presence or absence of metatatic disease or locally invasive disease did not correlate with overall survival (Figure 2 c and d). In those with metastatic disease, those with peritoneal or omental only metastases had no difference in overall survival compared to those without (23.9 vs 22.0 months, HR 0.64, p=0.54). There was a difference of 4 months survival with the presence or absence of bone only disease alone though this did not reach statistical significance (23.4 months vs 27.3 months, HR 0.72, p=0.75) (Figure 2e). There was a difference of just under 7 months survival with the presence or absence of visceral only metastasis though not reaching statistical significance (18.9 months vs 25.7 months, HR 2.36, p=0.07) (Figure 2g). The presence of brain metastases had a poor median overall survival of 14.1 months compared to a median survival of 23.9 months though the comparison was limited by low numbers of patients with brain only metastases (HR 3.19, p=0.12) (Figure 2h).
Early phase clinical trial details
Of the 164 patients who were referred for consideration of early phase clinical trials, 92 (56%) were identified as suitable for allocation onto a trial with 60 (37%) deemed ineligible for enrollment and the remaining 12 (7%) were unknown whether they were deemed eligible for enrolment upon a clinical trial. Of the 92 patients who went forward for clinical trial screening, 69 went commenced on an early phase trial (75%) and 23 did not ultimately embark upon a study due to trial ineligibility (10, 43%), patient deterioration (8, 35%), or patient decision (5, 22%). Of those that received an investigational compound, the predominant radiological tumour response was progressive disease in the majority (39, 57%), with 25 (36%) having stable disease and one confirmed complete response and one confirmed partial response (1% respectively). Three patients (4%) were non-evaluable due to discontinuing the trial prematurely and not being radiological assessable.