Patients
Between January 1st, 2008 and December 31st, 2020, 372 patients were diagnosed with solid tumors, while 416 were diagnosed with lymphoma. Among them, forty patients with solid tumors and sixty-six patients with lymphoma were further diagnosed with pleural effusion, ascites, or pericardial effusion. A total of eighty-five patients received IL-2 injection while the remaining twenty-one patients did not. Indeed, twenty-one patients were diagnosed at early stage of this retrospective study. IL-2 injection was not routinely used to treat effusions at the time.
Among eighty-five patients in the effusion group, twenty-one had stage ⅠII diseases (including T cell lymphoblastic lymphoma, B cell lymphoblastic lymphoma, primitive neuroectodermal tumor (PNET), HB, and pediatric pneumoblastoma (PPB) in twelve, three, two, and two cases, respectively), while the remaining sixty-four had stage IV diseases (including T cell lymphoblastic lymphoma, B cell lymphoblastic lymphoma, diffuse large B-cell lymphoma (DLBCL), Burkitt's lymphoma (BL), anaplastic large cell lymphoma (ALCL), rhabdomyosarcoma (RMS), NB, Ewing’s, HB, PNET, and PPB in twenty-four, six, six, three, two, seven, four, four, four, three, and one case, respectively). Meanwhile, fifty-eight patients only had pleural effusion (including bilateral pleural effusions in twenty-nine patients), while fifteen only had ascites. The remaining twelve cases had concurrent pleural effusion, ascites, and pericardial effusion. Moreover, the remaining twelve patients all had bilateral pleural effusions.
Accordingly, for patients without IL-2 injection, 1 had stage II disease (HB in one), and one had stage IVs disease (NB in one). Nine cases had stage III diseases (including T cell lymphoblastic lymphoma in three cases, B cell lymphoblastic lymphoma in one, Ewing's in two, PNET in one, and HB in two), while the remaining ten patients had stage IV diseases (including T cell lymphoblastic lymphoma in two cases, B cell lymphoblastic lymphoma in one, DLBCL in two, ALCL in one, RMS in three, and PNET in one). Meanwhile, eleven patients only had pleural effusion (including bilateral pleural effusions in seven patients), while six only had ascites. The remaining four cases had concurrent pleural effusion, ascites, or pericardial effusion, all of which had bilateral pleural effusions. The detailed characteristics of one hundred and six patients are presented in Table 1.
Table 1. The detailed characteristics of 106 patients
|
Effusion group
|
Control group
|
|
Solid tumor
|
Lymphoma
|
Solid tumor
|
Lymphoma
|
Age(mean±se)
|
5.42±3.97
|
8.02±3.43
|
4.96±3.7
|
9.8±2.39
|
Gender
|
|
|
|
|
Male
|
15
|
32
|
6
|
6
|
Female
|
14
|
24
|
5
|
4
|
stage
|
|
|
|
|
II
|
|
|
1
|
|
III
|
4
|
15
|
5
|
4
|
IV
|
24
|
44
|
4
|
6
|
IVs
|
|
|
1
|
|
Histology
|
|
|
|
|
RMS
|
7
|
|
3
|
|
NB
|
4
|
|
1
|
|
PPB
|
3
|
|
|
|
Ewing's sarcoma
|
4
|
|
2
|
|
HB
|
6
|
|
3
|
|
PNET
|
5
|
|
2
|
|
T cell lymphoblastic lymphoma
|
36
|
|
5
|
B cell lymphoblastic lymphoma
|
9
|
|
2
|
DLBCL
|
|
6
|
|
2
|
BL
|
|
3
|
|
|
ALCL
|
|
2
|
|
1
|
|
|
|
|
|
|
Note: RMS, rhabdomyosarcoma; NB, neuroblastoma; PPB, pediatric pneumoblastoma; HB, hepatoblastoma; PNET, primitive neuroectodermal tumor; DLBCL, diffuse large B-cell lymphoma; BL, Burkitt's lymphoma; ALCL, anaplastic large cell lymphoma
Dyspnea, cough, and discomfort were the most frequently reported symptoms of pleural and pericardial effusions, whereas abdominal distension, abdominal pain, and edema were the most widely recognized symptoms of ascites.
Treatment
While all patients in the effusion group were diagnosed with a malignant tumor and pleural effusion, ascites, or pericardial effusion, IL-2 injection therapy may be administered without a pathological diagnosis as long as malignant lesions associated with pleural, abdominal, or pericardial effusion were identified on imaging. We obtained written informed consent from patients’ parents or legal guardians before starting the therapy. First, patients performed thoracic, abdominal cavity, or pericardial cavity puncture with the indwelling of a drainage tube, and pathology was simultaneously obtained if permitted.
In patients with unilateral pleural effusion, no more than 600 mL fluid was drained on the first day and no more than 1000 mL each day; for patients with bilateral pleural effusions, the total drainage volume was the same. Ascites should not exceed 1000 mL each time, and pericardial effusion should not exceed 100 mL each time. In the presence of multi-cavity effusions, the drained effusion amount should be reduced as appropriate, and static electricity of hydration solution should be applied simultaneously.
After discharge, 0.9% sodium chloride injection (0.9%NaCl, maximum 100 mL) combined with IL-2 (5.0-10.0×106 IU/m2, maximum dose 10.0×106 IU) and DEX (5 mg) was injected via the drainage tube. The injection was administered every other day, and the total number of injections was not strictly limited, which was stopped when ultrasound confirmed that the effusion was no more than 2 cm or disappeared. For patients with bilateral pleural effusions or multi-cavity effusions, the maximum total doses of IL-2 and DEX were maintained at 10.0×106 IU and 5 mg, respectively, which should be divided according to 0.9%NaCl volume. It should be noted that 0.9%NaCl volume should not exceed 50 mL during unilateral thoracic injection and should not exceed 20 mL during pericardial cavity injection. Besides, the injection time should be more than 1 hour. It was advisable to use an injection pump to pump the fluid at a uniform rate. No strict requirement was found for the intraperitoneal injection rate or the fluid amount. After injection, the drainage tube was closed until the following day, and the child was instructed to change the position as much as possible to ensure the wider distribution of IL-2.
Chemotherapy might be initiated during IL-2 therapy. All patients in the control group were treated according to pathology diagnosis without IL-2 therapy.
Response
A total of four hundred and eighty one injections were administered for eight-five patients. The average number of injections into the pericardial cavity was two, while that into pleural and intraperitoneal cavities were three to four. Only one patient with T cell lymphoblastic lymphoma received the maximum seven pleural injections.
Among the eighty-five patients, half had bloody drainage fluid, and tumor exfoliated cells were detected in the drainage fluid from thirty-one patients. The injections generally had limited toxicity, and only eleven patients developed a moderate fever on the first day of injection. No patient developed respiratory distress related to IL-2 injection therapy. Simultaneously, no allergic reaction or catheter-related infection occurred.
Outcome
In this study, patients with solid tumors were mainly treated according to the protocols from COG or International Society of Pediatric Oncology group (SIOP)[7-13], whereas those with lymphoma were mainly treated in line with BFM protocols[14-17]. All patients in the effusion group achieved CR of effusions from IL-2 injection therapy, even though one patient received seven injections altogether. No recurrence of pleural effusion, ascites, or pericardial effusion was noticed.
Among patients in the effusion group with lymphoma, four died of disease progression, and two had relapsed disease (including three with T cell lymphoblastic lymphoma, one with DLBCL, one with BL, and one with ALCL). One of the two patients with relapsed disease died, while the other with ALCL achieved SD after crizotinib treatment[18]. The five-year EFS and five-year OS for patients with lymphoma were 89.3%±31.2% (95%CI, 80.9% to 97.6%) and 91.1%±28.8% (95%CI, 83.4% to 98.8%), respectively. Among patients in the control group with lymphoma, two died of disease progression, and two had relapsed disease (including one with T cell lymphoblastic lymphoma, two with DLBCL, and one with ALCL). Both patients with relapsed disease died. The five-year EFS and five-year OS for patients with lymphoma were 60%±51.6% (95%CI, 23.1% to 96.9%). The Kaplan Meier analysis demonstrated significant differences between the two groups with both of which p<0.01. When we calculated the hazard ratio (HR), we found that it was 0.191 for EFS and 0.161 for OS (Figure 1).
For patients in the effusion group with solid tumors, four died due to disease progression, and eight got relapsed diseases (including three with RMS, one with NB, three with PPB, one with Ewing's sarcoma, one with HB, and one with PNET). Seven of the eight patients with relapsed disease died, while the other with HB achieved secondary CR after irinotecan treatment[19]. The five-year EFS and five-year OS were 62.1%±49.4% (95%CI, 43.3% to 80.9%) and 65.5%±48.4% (95%CI, 47.1% to 83.9%), respectively. For patients in the control group, among patients with solid tumors, two died due to disease progression, and four got relapsed diseases (including three with RMS, one with NB, one with Ewing's sarcoma, and one with PNET). Three patients with relapsed disease died, while the other with NB achieved PR after irinotecan treatment and alive with tumor. The five-year EFS and five-year OS were 45.5%±52.2% (95%CI, 10.4% to 80.5%) and 54.5%±52.2% (95%CI, 19.5% to 89.6%), respectively. The Kaplan Meier analysis showed no statistical difference between the two groups with both of which p>0.05 (Figure 2).
However, when lymphoma and solid tumor patients were combined, the Kaplan Meier analysis revealed a significant difference between the two groups, with p<0.01 for EFS and OS. HR=0.344 between OS and 0.352 between EFS (Figure 3).
The mean effusion control time (<2m pleural effusion, or disappearance of ascites or pericardial injection) for the effusion group was 5.76±1.95 days (95%CI, 5.34 to 6.19 days), while for the control group was 18.3±5.25 days (95%CI, 15.94 to 20.72 days), which had statistical difference (p<0.01).