Keywords
tumor - thymus - surgery - complications - pleural disease
Introduction
The prognosis of patients with malignant thymic epithelial tumors (thymoma, thymic
carcinoma) correlates closely with postoperative resection-status, histological subtype,
and Masaoka-Koga stage.[1] In early stages, radical surgical resection of the thymic tumor is still the recommended
treatment.[2] While these patients with early tumor stages (Masaoka-Koga stage I and II) have
excellent prognosis, patients with more invasive thymic malignancies show high rates
of local recurrence and, thus, poor long-term survival.[3]
[4] Masaoka-Koga and TNM stage IVa is defined as pleural or pericardial tumor spread,
which includes both primary tumors and recurrent pleural tumors. Considering only
patients with thymic malignancies (thymoma) in stage IVa, 5-year survival rates of
53 to 87% were reported after surgical resection.[5]
[6] A tumor of this extent is difficult to treat with surgical procedures because residual
tumor cells often remain in the pleural cavity, leading to pleural recurrences.[7]
[8]
[9] Therefore, advanced stages require a multimodal treatment approach involving systemic
chemotherapy, radiotherapy, or radio/chemotherapy. They can be combined with surgery
in neoadjuvant and adjuvant settings or used stand-alone with a palliative intention.[6]
[10] Systemic therapies have been reported to show only temporary impact on tumor progression.[11] Curative therapy for this advanced stage is not yet known.[12]
An innovative technique that is expected to improve local tumor control in patients
with pleural tumor involvement is surgical cytoreduction followed by hyperthermic
intrathoracic chemotherapy (HITOC). The HITOC approach is based on the dose–response
principle, whereby locally high doses of chemotherapeutic agents can be achieved in
the pleural cavity with minimal absorption.[11] Very few studies exist on this innovative procedure in patients with thymic malignanices.[11]
[13]
[14]
[15] Nevertheless, this procedure shows promising results for patient safety in the existing
literature, while the effect on overall survival and recurrence-free survival requires
further investigation.[7]
Patients and Methods
Study Design
This is a retrospective, single-center study of patients with thymic malignancies
according to Masaoka-Koga/TNM stage IVa and primary pleural dissemination or pleural
recurrence. All patients who received a radical tumor resection (surgical cytoreduction)
with subsequent HITOC from September 2008 to December 2017 were included (n = 29). The main tumor manifestation was located at the pleural surface, sometimes
even involving the pericardium, the diaphragm, or the lung. Lymph node sampling was
performed if suspected lymph nodes appeared intraoperatively.
The study was approved by our institutional ethics committee, and informed consent
was obtained from all patients. Demographic data, surgical parameters, data regarding
HITOC, and postoperative data were documented. The follow-up examination was performed
in May 2018 by the regional Tumor Centre Regensburg. Tumor recurrence was defined
by computed tomography scan or positron emission tomography scan. Recurrence-free
interval and overall-survival were examined.
The primary end point of the study was to determine postoperative morbidity with special
respect to renal insufficiency and mortality after surgical cytoreduction including
HITOC. Secondary end points included the recurrence-free and overall-survival.
Surgical Cytoreduction
Depending on the location of the tumor, different surgical procedures were used for
cytoreduction. Almost all patients received two anterolateral thoracotomies (5th and
8th intercostal space) for an optimal view and inspection of the whole thoracic cavity.
The procedures included pleurectomy/decortication (P/D), extended P/D (eP/D including
resection of the pericardium and/or diaphragm and/or lung resection), and extrapleural
pneumonectomy (EPP). Lung-sparing procedures (P/D and eP/D) were preferred for cytoreduction
if possible. Therefore, an extrapleural dissection between lung and tumor was performed
after anterolateral thoracotomy. A pleurectomy was performed from the apex to the
diaphragm until the hilar structures were free. The visceral pleura was then separated
from the lung. If necessary, wedge resections of infiltrated parts of the lung or
resections of the pericardium, diaphragm, or chest wall were performed (eP/D). Resected
areas of the pericardium or diaphragm were reconstructed and covered with bovine patches.
Only in selected patients with extensive tumor invasion an EPP was performed. These
tumors were primarily located in the anterior mediastinum and infiltrated the pleural
cavity and lung parenchyma. In these patients, a macroscopic complete tumor resection
was only reachable by EPP.
HITOC Procedure
If possible, HITOC was performed right after the surgical cytoreduction in the same
session. For the safety of the staff, the procedure was always performed with a closed
chest. At least one apical inflow drainage and two or three basal outflow drainages
were placed before the thoracotomies were closed. The drains were connected to the
perfusion system (ThermoChemo HT1000) and 3 to 4 L sodium chloride was introduced
into the circulation as priming volume while the system was heated to a temperature
of 42°C measured at the outflow drains. As soon as circulation was stable, the chemotherapeutic
agents were added. The dosage of chemotherapeutic agents was increased over the course
of the study because the patients rarely presented cisplatin-related complications
and also higher dosages showed better survival in mesothelioma patients in the literature.
In the beginning of the study, cisplatin was used at a dose of 100 mg/m2 body surface area (BSA), which was later increased to 175 mg/m2 BSA and combined with 65 mg doxorubicin. The perfusion was performed over a period
of 60 minutes with a flow rate of ∼1.5 L/min. Afterward, the perfusion solution was
drained and the drains left for the early postoperative period. During the operation
and for approximately the first five days postoperatively we aimed a fluid balancing
and a forced diuresis > 100 mL/h. Starting in June 2013, we added amifostine and sodium
thiosulfate to reduce renal toxicity from cisplatin as reported by the working group
of Dr. Sugarbaker et al.[16]
[17] Amifostine was given intravenously in a dosage of 910 mg/m2 BSA 30 minutes prior to the initiation of the HITOC. Sodium thiosulfate was administered
directly after HITOC (4 g/m2 BSA) and continuously over a period of 6 hours (12 g/m2 BSA) at the intensive care unit.
Statistical Analysis
The data were collected in tabular form using Excel, Version 15.0 (Microsoft Corporation,
Redmond, Washington, United States). Afterward, statistical analysis was performed
using IBM SPSS Statistics, Version 24 (IBM Corporation, Armonk, New York, United States).
The demographic data were summarized as numbers and percentages. Student's t-test and Mann–Whitney U tests were used for group comparisons if appropriate. The
Kaplan–Meier method was used for the analysis of recurrence-free survival and overall-survival.
Recurrence-free and overall-survival were defined as the interval between treatment
and the occurrence of tumor recurrence, death or last follow-up. The log-rank test
was used to examine the influence of tumor type, histological subtype, postoperative
resection status, and the combination of chemotherapeutic agents on survival. A p-value <0.05 was considered statistically significant.
Results
Demographic Data
The mean age of the study population was 49 years, and 58.7% of patients were male
([Table 1]). In ∼52% of patients, the pleural tumor dissemination was located on the right
side. Eleven (37.9%) patients had thymic malignancies with primary pleural tumor spread
(primary stage IVa) when they were scheduled for surgery for the first time, while
18 (62.1%) patients showed pleural tumor recurrence (secondary stage IVa) after previous
surgery of thymic tumors. The median interval between first diagnosis of the thymic
tumor and pleural reoperation (surgical cytoreduction) including HITOC in these patients
was 36 months (interquartile range 37.5 months). The most frequent histological subtype
in the study population was World Health Organization (WHO) grade B2 (n = 10; 34.5%), followed by grade B3 (n = 7; 24.2%). In three cases (10.3%), participants suffered from a thymic carcinoma
(WHO grade C). In 15 cases (51.7%), induction therapy was performed prior to resection
to downsize the tumor and increase the chance for complete tumor resection.
Table 1
Demographic data
|
Study sample (n = 29; 100%)
|
|
Male gender
|
17 (58.7)
|
|
Mean age ± SD (years)
|
49 ± 12.2
|
|
Mean BMI ± SD (kg/m2)
|
26.2 ± 4.9
|
|
Mean BSA ± SD (m2)
|
1.93 ± 0.23
|
|
Tumor localization on the right side
|
15 (51.7)
|
|
Tumor
|
|
• Primary tumor
|
11 (37.9)
|
|
• Recurrence
|
18 (62.1)
|
|
Induction therapy
|
15 (51.7)
|
|
Who
|
|
• B1
|
4 (13.8)
|
|
• B2
|
10 (34.5)
|
|
• B2/3
|
5 (17.2)
|
|
• B3
|
7 (24.2)
|
|
• C
|
3 (10.3)
|
|
Patients with recurrent thymoma: Median interval first diagnosis - reoperation with
HITOC; IQR (months)
|
36; 37.5
|
Abbreviations: BMI, body mass index; BSA, body surface area; HITOC, hyperthermic intrathoracic
chemotherapy; IQR, interquartile range; SD, standard deviation; WHO, World Health
Organization.
Operative Data
[Table 2] shows the operative parameters. Most patients received eP/D (n = 15; 51.8%). For almost all patients (n = 27; 93.1%), HITOC was performed after the resection in the same session. In two
cases (6.9%) with extensive surgical tumor resection and hemodynamic instability,
all thoracic drains were placed as described above and HITOC was performed 1 and 19
days later. The initially used dose of 100 mg/m2 BSA cisplatin was applied to eight patients (27.6%). In seven patients (24.2%), the
cisplatin dose ranged between 150 and 175 mg/m2 BSA. In the later study period and with increased experiences, most patients received
a combination of cisplatin and doxorubicin, with cisplatin at the maximum dosage of
175 mg/m2 BSA (n = 14; 48.2%). Average operation time including HITOC was 275 minutes (min: 118 minutes,
max: 440 minutes).
Table 2
Operative data including HITOC
|
Study sample (n = 29; 100%)
|
|
Surgical cytoreduction
|
|
• P/D
|
11 (37.9)
|
|
• eP/D
|
15 (51.8)
|
|
• EPP
|
3 (10.3)
|
|
HITOC
|
|
• Same session
|
27 (93.1)
|
|
• Second session
|
2 (6.9)
|
|
Chemotherapeutics
|
|
• Cisplatin 100 mg/m2 BSA
|
8 (27.6)
|
|
• Cisplatin 150–175 mg/m2 BSA
|
7 (24.2)
|
|
• Cisplatin 175 mg/m2 BSA + doxorubicin 65 mg
|
14 (48.2)
|
|
Cytoprotection
|
16 (55.2)
|
|
Mean duration of operation ± SD (minutes)
|
275 ± 79
|
Abbreviations: BSA, body surface area; eP/D, extended pleurectomy/decortication; EPP,
extrapleural pneumonectomy; HITOC, hyperthermic intrathoracic chemotherapy; P/D, pleurectomy/decortication;
SD, standard deviation.
Postoperative Data
In 25 (86.2%) patients, a macroscopically complete tumor resection (R0/R1) was achieved.
Four patients had macroscopically residual tumors (R2: 13.8%). In two patients, a
transdiaphragmal, intra-/retroperitoneal spread of the tumor was observed. These resections
with macroscopically low tumor residues were intraoperatively classified as R2. Other
two patients showed infiltration of the superior or inferior vena cava so that tumor
debulking was performed. Since a complete resection could not be determined with certainty,
they were also classified as R2. Intraoperative complications occurred only in one
(3.4%) patient (temporary right heart decompensation before HITOC). Surgical revision
was necessary in a total of six (20.7%) patients. Half of these patients (n = 3) suffered from a prolonged parenchymal fistula after extensive decortication,
which made a rethoracotomy with suture of the lung parenchyma necessary. The other
three cases included patch rupture of the diaphragm (re-attachment), chylothorax (closure
of lymphatic fistula), and hemothorax (evacuation of hematoma). One patient developed
pulmonary embolism, which could be treated with anticoagulation. Another patient developed
sepsis with subsequent multiple organ failure and died in the hospital. The median
stay in the intensive care unit was 1 day, and the total stay in the hospital was
18 days. In 13 patients (44.8%), adjuvant chemotherapy was performed, and eight (27.6%)
patients received adjuvant radiotherapy, depending on the resection status and the
histological subtype ([Table 3]).
Table 3
Postoperative data
|
Study sample (n = 29; 100%)
|
|
Resection
|
|
• R0/1
|
25 (86.2)
|
|
• R2
|
4 (13.8)
|
|
Median ICU stay; IQR (d)
|
1; 2
|
|
Median hospital-stay; IQR (d)
|
18; 18
|
|
Intraoperative complications
|
1 (3.4)
|
|
Postoperative complications
|
9 (31)
|
|
• Surgical revision required
|
6 (20.7)
|
|
• Postoperative dialysis
|
2 (6.9)
|
|
• Postoperative pulmonary embolism
|
1 (3.4)
|
|
In-hospital mortality
|
1 (3.4)
|
|
90-d mortality
|
1 (3.4)
|
|
Adjuvant chemotherapy
|
13 (44.8)
|
|
Adjuvant radiotherapy
|
8 (27.6)
|
Abbreviations: ICU, intensive care unit; IQR, interquartile range; SD, standard deviation.
Postoperative Renal Function
Only a moderate increase in serum creatinine levels was observed in the first postoperative
days. Looking at all patients, the collected data showed a maximum mean value of creatinine
of 1.13 mg/dL 2 days after surgery. As illustrated in [Fig. 1], patients who received amifostine and sodium thiosulfate for cytoprotection showed
an overall milder increase in serum creatinine. In patients without cytoprotection,
the maximum was reached after 1 week and was higher overall 1.24 mg/dL. At this time,
patients with cytoprotection showed a considerably lower creatinine level of 0.7 mg/dL.
The differences in median serum creatinine levels turned out to be statistically significant
(p = 0.036), although this finding was not clinically relevant. Despite the increasing
cisplatin concentration during the study and the subsequent combination with doxorubicin,
there was no increased rate of postoperative renal insufficiency. Postoperative temporary
dialysis was necessary in two patients (6.9%) in the early study period without cytoprotection
([Table 3]).
Fig. 1 Serum creatinine concentration after surgical cytoreduction and hyperthermic intrathoracic
chemotherapy. SD, standard deviation.
Survival Analysis
Overall survival and recurrence-free survival are shown in [Fig. 2]. One patient with thymic carcinoma died after EPP during hospital stay (30-day mortality
3.4%) and was, therefore, excluded from the long-term survival analysis. A 5-year
survival rate of 80.1% was observed in all patients (n = 28). The mean follow-up was 35.6 months (min: 0 months, max: 106 months). At the
end of the follow-up, 62.5% of the patients were alive, and the recurrence-free survival
after 5 years was 53.8% for patients with R0/R1 resection (n = 24). The median recurrence-free interval was 69 months (95% confidence interval:
40.4–97.6; min: 0 months, max: 69 months). Concerning recurrence-free survival, no
significant prolongation was observed by combining cisplatin with doxorubicin (p = 0.59). The log-rank test showed no significant difference in survival between patients
with primary pleural tumors and patients with pleural recurrence (p = 0.88). Concerning histological subtypes, the lowest survival was found in patients
with thymic carcinoma (p = 0.01; [Fig. 3]). No difference in survival could be observed by resection status (p = 0.394) or by choice of chemotherapeutic agents for HITOC (p = 0.39). The corresponding graphs and associated rates are presented in [Fig. 4]. Similarly, no difference was observed by the concentration of cisplatin (p = 0.99) or the surgical procedure used (p = 0.27).
Fig. 2 Overall survival and recurrence-free survival.
Fig. 3 Survival depending on type of tumor and histological subtype. WHO, World Health Organization.
Fig. 4 Survival depending on resection status and chemotherapy for hyperthermic intrathoracic
chemotherapy (HITOC).
Discussion
The treatment of patients with thymic malignancies and pleural dissemination (stage
IVa) has been the subject of controversial discussions.[3]
[4] Combinations of surgical cytoreduction and radiochemotherapy currently represent
the gold standard in treatment, but the beneficial effects of systemic therapies are
limited. However, the advantages of additional local HITOC are still not clear.[11] Most clinical experience with HITOC has been gained in the treatment of malignant
pleural mesothelioma with a prolongation of overall-survival (35.3 vs. 22.8 months).[7]
[18]
In our study, HITOC was performed either with cisplatin alone or in combination with
doxorubicin. Cisplatin is well-known as first-line therapy for thymomas in systemic
approaches.[6] In the literature, cisplatin is always a component for HITOC. A maximum tolerable
dose of cisplatin of 225 mg/m2 BSA was determined in a previous mesothelioma study and we also increased the dosage
of cisplatin during the study as higher doses demonstrated a better survival in patients
with malignant pleural mesothelioma.[16]
[19] We expected that also in thymoma patients an increased dosage would lead to better
long-term results. The use of a second agent, however, differs and is still debated.
While some studies use cisplatin alone, others combine it with a second substance,
often an anthracycline.[13]
[14]
[15]
[20] In our study, the addition of doxorubicin did not yet significantly prolong overall
and recurrence-free survival. For this reason, excluding doxorubicin could be considered.
It has already been proven that only a fraction of the cisplatin concentration of
the intrathoracic perfusion solution passes into the serum.[21]
[22] Nevertheless, systemic complications caused by chemotherapeutic agents (cisplatin
concentration) have to be considered.[19] Cisplatin has dose-dependent renal toxicity, and the most important chemotherapy-related
complication to be mentioned here is the postoperative increase in creatinine. Moderate-to-severe
nephrotoxicity was noted in 25 to 33% of patients receiving a single intravenous dose
of cisplatin of 50 to 75 mg/m2.[23] In several studies, HITOC caused renal dysfunction in 8.7 to 57% of patients.[19]
[24]
To attenuate cisplatin-induced nephrotoxicity, fluid balancing including hydration
(with diuresis as needed) is generally introduced in the cisplatin-based chemotherapy
regimen as a nephroprotective strategy.[25]
In our opinion, additional cytoprotection is one of the most important tasks during
HITOC. Amifostine pretreatment should reduce tubulotoxicity and sodium thiosulfate
is a neutralizing agent for cisplatin that also protects against renal damage.[26] The application of amifostine and sodium thiosulfate as cytoprotective agents significantly
reduced creatinine levels in the postoperative period during the study. The two patients
in our study with postoperative transient dialysis were treated at the beginning of
our HITOC program without cytoprotection. After the introduction of cytoprotection,
no dialysis was necessary despite the increase in cisplatin concentration and the
additional combination with doxorubicin. Few other studies have already reported the
positive effect of cytoprotection on renal function.[16]
[17]
In the present study, the recurrence-free 5-year survival of patients with stage IVa
thymic tumors after multimodal treatment including HITOC was 53.8%, with an average
recurrence-free interval of 53.7 months. Our study exceeds the very good recurrence-free
interval of 42 months from a study of 19 patients treated with additional HITOC.[13] In the literature, multimodal therapy without HITOC resulted in distinctly lower
5-year recurrence-free rates of 13.3 to 33.6%.[27]
[28] Intrapleural chemotherapy, thus, offers the opportunity to kill residual tumor cells
that persist even after complete macroscopic removal and lead to local recurrence.[7]
[8]
[9] Additional hyperthermia increases the cytotoxic effect by improving penetration
depth and reducing chemotherapeutic resistance.[21]
Even though recurrence-free survival and overall survival are only partially linked
in thymomas (recurrence does not necessarily lead to death and deaths are often not
due to recurrence), good recurrence-free survival is also reflected in the 5-year
overall-survival rate of 80.1%.[29] Other studies of patients with stage IVa thymomas and HITOC found encouraging 5-year
survival rates of 67 to 92% ([Table 4]).[13]
[14]
[15]
[20] However, stage IVa patients who did not receive HITOC have published 5-year survival
rates ranging between 53 and 87%.[5]
[6]
Table 4
Study overview of stage IVa thymic tumors and application of HITOC
|
Author, y
|
Patients
|
Chemotherapeutics (dosage)
|
Mortality
|
Recurrence free survival
|
Overall survival
|
|
Refaely et al, 2001[14]
|
n =15
|
Cisplatin (100–200 mg/m2 BSA)
|
0%
|
–
|
70% (5 y)
|
|
de Bree et al, 2002[11]
|
n = 14
|
Cisplatin (80 mg/m2 BSA) + doxorubicin (15–30 mg/m2 BSA)
|
0%
|
100% (18 mo)
|
100% (18 mo)
|
|
Ried et al, 2012[7]
|
n = 8
|
Cisplatin (100–150 mg/m2 BSA)
|
0%
|
88% (22 mo)
|
88% (22 mo)
|
|
Yellin et al, 2013[15]
|
n = 35
|
Cisplatin (100 mg/m2 BSA) + doxorubicin (50–60 mg)
|
2.5%
|
48–61% (5 y)
|
67–81% (5 y), thymic carcinoma 0%
|
|
Ambrogi et al, 2016[20]
|
n = 13 (pleural recurrences)
|
Cisplatin (80 mg/m2 BSA) + doxorubicin (25 mg/m2 BSA)
|
0%
|
77% (64 mo)
|
92% (5 y)
|
|
Maury et al, 2017[13]
|
n = 19
|
Cisplatin (50 mg/m2 BSA) + mitomycin (25 mg/m2 BSA)
|
0%
|
50% (42 mo)
|
86% (5 y)
|
Abbreviations: BSA, body surface area; HITOC, hyperthermic intrathoracic chemotherapy.
The good overall-survival of surgically treated patients with thymoma stage IVa is
an expression of the low metastatic potential of thymomas. One possible reason for
this positive result may also lie in the high proportion of macroscopically completely
resected tumors (86.2% in our study), since resection status is known to be an important
prognostic factor.[2] Only in four cases with involvement of the great vessels or intra-/retroperitoneal
spreading a postoperative R2 situation was given.
Another important prognostic factor is the histological subtype. Patients with thymic
carcinoma (WHO type C) showed significantly worse prognosis in the present study (0
vs. 83.1–91.7% 5-year survival rates). Especially studies with high survival rates
have excluded thymic carcinomas.[5]
[6] Thus, these patients with stage IVa thymic carcinomas seem not to benefit from surgery,
and therefore, HITOC also had no significant effect.
There was no difference in overall-survival between primary stage IVa thymic tumors
and patients with pleural recurrence. Therefore, in patients with pleural recurrence
of thymoma, surgery and HITOC can be indicated. Until now, no significant difference
in survival was observed by resection status (85.5 vs. 50%), choice of chemotherapeutic
agents for HITOC, or the concentration of cisplatin. However, a longer follow-up might
be required for this tumor entity.
The extent of surgery (P/D vs. EPP) in our study did not affect survival. Also in
the literature, lung-protecting procedures such as P/D and eP/D showed similar survival
but lower complication rates compared with EPP.[30] Due to this fact, P/D and eP/D were preferred depending on the patient's condition,
and complete resection of the tumor could be achieved in 88.5% of these patients.
This study has limitations due to its retrospective nature and single-center design
in a selected and heterogeneous group of patients. Due to this special tumor entity
and advanced stage, the number of patients is limited. Furthermore, there was no control
group of patients who did not receive HITOC. A longer follow-up might substantiate
our findings.
Conclusion
Surgical cytoreduction with HITOC is safe and feasible in selected patients with thymic
malignancies and pleural dissemination. It offers encouraging survival rates, both
in primary and secondary (pleural recurrence) stage IVa thymomas. In contrast, thymic
carcinomas show a considerably lower survival rate and should be treated as a distinct
tumor entity. To improve local tumor control resulting in prolonged recurrence-free
survival, HITOC with cisplatin can be used in addition to surgical cytoreduction.
Severe HITOC-associated complications are rare and renal complications can be reduced
by fluid balancing and cytoprotection.
