Chemoradiotherapy plus hyperthermia (CRTH) versus chemoradiotherapy (CRT) alone in neoadjuvant treatment of soft tissue sarcoma: tumor response, treatment toxicity and disease control

Abstract

Introduction

Neoadjuvant chemotherapy and radiotherapy for the management of soft tissue sarcomas (STS) are still preferably delivered sequentially, with or without concurrent hyperthermia. Concurrent delivery of chemo-, radio- and thermotherapy may produce synergistic effects and reduce chemotherapy-free intervals. The few available studies suggest that concurrent chemoradiation (CRT) has a greater local effect. Data on the efficacy and toxicity of adding hyperthermia to CRT (CRTH) are sparse.

Materials and methods

A cohort of 101 patients with STS of the extremities and trunk who received CRT (n = 33) or CRTH (n = 68) before resection of macroscopic tumor (CRT: n = 19, CRTH: n = 49) or re-resection following a non-oncological resection, so called ‘whoops procedure’, (CRT: n = 14, CRTH: n = 19) were included in this retrospective study. CRT consisted of two cycles of doxorubicine (50 mg/m² on d2) plus ifosfamide (1500 mg/m² on d1-5, q28) plus radiation doses of up to 60 Gy. Hyperthermia was delivered in two sessions per week.

Results

All patients received the minimum dose of 50 Gy. Median doses of ifosfamide and doxorubicin were comparable between CRT (75%/95%) and CRTH (78%/97%). The median number of hyperthermia sessions was seven. There were no differences in acute toxicities. Major wound complications occurred in 15% (CRT) vs. 25% (CRTH) (p = 0.19). In patients with macroscopic disease, the addition of hyperthermia resulted in a tendency toward improved remission: regression ≥90% occurred in 21/48 (CRTH) vs. 4/18 (CRT) patients (p = 0.197). With a median postoperative follow-up of 72 months, 6-year local control and overall survival rates for CRTH vs. CRT alone were 85 vs. 78% (p = 0.938) and 79 vs. 71% (p = 0.215).

Conclusions

Both CRT and CRTH are well tolerated with an expected rate of wound complications. The results suggest that adding hyperthermia may improve tumor response.

Keywords:

• Sarcoma
• neoadjuvant
• radiotherapy
• chemoradiotherapy
• hyperthermia
• soft-tissue sarcoma
• toxicity
• chemotherapy

Introduction

Soft tissue sarcomas (STS) are rare tumors with variable growth patterns. Consequently, no uniform standard sequence of conservative and surgical treatment has been established to date. What is indisputable is the need for resection, which should be performed at a sarcoma center whenever possible in order to achieve high complete resection rate [1]. There is consensus that radiotherapy (RT) halves the local recurrence rate [2], but disagreement over the timing of radiotherapy. Preoperative radiotherapy offers the advantage of lower long-term toxicity but is associated with a higher incidence of postoperative complications [3]. Preoperative radiotherapy also increases the odds of R0 resection and, thus, the odds of improved overall survival (OS), albeit minimally [4]. Therefore, some centers prefer preoperative radiotherapy, while others have established neoadjuvant chemotherapy concepts. The goal of both approaches is to improve local tumor response and to prevent micro metastasis early in the disease course [5,6]. An EORTC study showed that administering hyperthermia with chemotherapy concurrently increases the odds of local progression-free survival and overall survival [7,8]. When delivered in combination, chemotherapy, hyperthermia and radiotherapy are often performed sequentially. Some experience exists on the combination of radiation and parallel administration of ifosfamide, as recommended within an therapeutic regimen of the ‘German Interdisciplinary Working Party on Sarcomas’(IAWS) [9]. However, concurrent administration can be well justified as an attempt to achieve a local additive effect as well as to prevent the loss of time and treatment intensity before tumor resection. This is also the rationale behind concurrent chemoradiotherapy (CRT). Although the concurrent chemoradiation concept failed to gain acceptance in earlier years, research teams are now reconsidering this approach because CRT can be supported by leukocyte stimulating treatments. More recent observations suggest that integrating CRT into the neoadjuvant treatment regimen can compensate for the negative effect of incomplete resection on the local recurrence rate [10]. The feasibility of combining anthracycline plus ifosfamide chemotherapy with radiotherapy is also being discussed. Only a few studies on this topic exist [10,11]. Hyperthermia has the ability to make cells more sensitive to radiotherapy as well [12]. However, very few sarcoma centers have the possibility to provide hyperthermia and published data are scarce. Data on the effects of adding hyperthermia to concurrent chemoradiotherapy (CRTH) regimens are also scarce. Therefore, this research was designed to address the following questions: How should CRT be administered in comparison to CRTH? What results can be expected following CRT with or without hyperthermia and does hyperthermia have an additive effect? How high is the rate of wound complications with CRTH compared to CRT alone?

Materials and methods

Patients

This retrospective study analyzed the treatment outcomes of 101 adults (≥18 years) with high- or intermediate-grade soft tissue sarcomas (STS) or isolated local recurrences of STS of the extremities, trunk or head and neck without distant metastasis and an ECOG performance status of ≤1, who received preoperative chemoradiation (pre-CRT) with doxorubicin plus ifosfamide with or without hyperthermia from 1 January 2000 to 1 November 2021, followed by resection or re-resection, if feasible; re-resection was performed only in cases of R1, R2 or Rx resection following a previous non-oncological resection, so called ‘whoops procedure’ (Figure 1). Before inclusion in the study, informed consent was obtained from each patient following establishment of the indication for neoadjuvant CRT or CRTH by the interdisciplinary tumor board.

Figure 1. CONSORT flow diagram of patient enrollment and treatment.

Neoadjuvant treatment regimen

Radiotherapy

After magnetic resonance imaging (MRI)- and computed tomography (CT)-based radiation planning, patients received conventionally fractionated radiotherapy to the tumor-bearing compartment at a dose of 50 Gy (at the ICRU50 reference point dose up until 2012 with a safety margin of 0.5–1 cm and a boost of up to 60 Gy to the MRI-visible sarcoma, with a safety margin of 4 cm in all directions, adjusted to the compartment borders [13] (Figure 2). Total dosage of 60 Gy was introduced at the same time as the 50 Gy of the approach [3], since our overall experience was good, we retained this approach. In patients treated after a ‘whoops procedure’, a reconstructed boost volume was defined based on information obtained from the patients, their previous treating physicians, and surgical reports, if available, and based on areas of scarring detected by MRI and on the skin. From 2000 to 2010, 16 patients (16%) received a split course of radiotherapy, in some cases, with a pause or acceleration [10,13]. Flab use is no longer traceable in older cases; in more recent cases, radiation is performed using a rotation technique without flab. Due to insufficient documentation, we did not perform an analysis.

Figure 2. Treatment regimen.

Chemotherapy

Patients received two cycles of ifosfamide (1.5 g/m2/d; d 1–5, q28, plus an isodose of mesna) and doxorubicin (50 mg/m2/d, d3, q28). After the first cycle, GCSF was used to treat chemotherapy-induced leukopenia <1,500/µl, as needed. The second cycle of chemotherapy was administered if the patient’s leukocyte or platelet counts were >3000/µl and >100,000/µl, respectively. Otherwise, it was postponed by a week and dose-reduced, as needed.

Hyperthermia

The 68 patients (68%) whose health insurance providers approved the additional treatment received 1–2 sessions/week of either deep regional hyperthermia (n = 50, 50%) or superficial hyperthermia (n = 18, 18%) The request for insurance coverage had to be made individually for each suitable and willing patient and was regularly rejected prior to publication of the EORTC trial results [8]. In this respect, the use of hyperthermia was not linked to the tumor situation but was purely based on insurance coverage and consent of the patient. Deep hyperthermia was performed using the BSD-2000 Deep Regional Hyperthermia System (Dr. Sennewald Medizintechnik, Munich, Germany), and was delivered in part with the MRI thermometry subsystem, if available. If MR-thermometry was possible, temperature was recorded with this technique, if MR-thermometry not possible, HT was done with clinical steering and superficial probes. Superficial hyperthermia was used when the tumor or the tumor bed had a maximum thickness of 3 cm and was delivered with BSD-500 (Dr. Sennewald Medizintechnik, Munich, Germany). Hyperthermia was introduced to the regimen starting in 2008. It was delivered at a target temperature of 40–44 °C for 60 min per session. Hyperthermia was administered immediately before radiotherapy in accordance with the published guidelines for use in clinical studies [14,15]. The time interval between two hyperthermia treatments was at least 72 h. In addition to continuous ECG monitoring, the patients’ blood pressure and oxygen saturation levels were continuously monitored during the treatments [16]. It was aimed to perform administration of ifosfamide parallel to hyperthermia application, Adriamycin was not applicated parallel to hyperthermia.

Surgery

Six to ten weeks after the end of radiotherapy, surgical treatment, consisting of either compartment resection (in cases with extensive involvement of an entire compartment) or wide resection, was performed with or without plastic coverage as needed. In patients with a previous Whoops procedure, the preexisting scar region was resected, giving special attention to the initially tumor-contaminated incision margins.

Pathological assessment

In patients with macroscopic tumors, a punch biopsy specimen was taken and examined by our reference pathologist and, if the patient had undergone a prior surgery or ‘whoop procedure’ elsewhere, the specimen was sent to the prior surgeon for re-assessment. In cases of resection or re-resection of the initially biopsied tumor, resection quality (R0, R1 – ink on tumor, R2 – no ink on tumor), residual tumor volume and tumor regression were quantitatively assessed based on the area of tumor cells and necrosis and expressed as a percentage of the baseline value. In case of re-resection after a ‘whoops procedure’, only the presence or absence of residual tumor and resection quality were assessed.

Adverse effects assessment

Treatment intensity was assessed based on data from the treatment protocols for chemotherapy, the Record and Verify database for radiation, and the individual hyperthermia protocols. Toxicities were graded according to the Common Toxicity Criteria for Adverse Events (CTCAE Version 5.0) [17] based on data from the analog and electronic medical records and laboratory programs. If CTCAE grading was not possible, the symptom was recorded without a severity grade. Surgical complications and wound complications were classified according to the National Cancer Institute of Canada SR2 trial (NCIC-SRS2) criteria [18].

Follow-up and statistical analysis

Follow-up consisted of computed tomography of the lungs and magnetic resonance imaging (MRI) of the tumor region, and was performed quarterly for two years, semiannually until the fifth year, and annually thereafter. An attempt was made to contact all patients whose status was unknown at the cutoff date, 3 November 2021. In case of patients who were lost to follow-up (n = 7, 7%), data from the last visit was used for the analysis. The oncological endpoints of the study were local recurrence free survival (LRFS), distant metastasis free survival (DMFS), disease free survival (DFS), and overall survival (OS), as calculated from the start of treatment. The statistical analysis was performed using SPSS Statistics, version 29.0.0.0. Frequency comparisons were performed using the chi-square test with or without Fischer’s exact test, as appropriate. Continuous variables were analyzed using the T-test for unpaired samples. Univariate and multivariate analyses of tumor-specific parameters and oncologic endpoints were also performed.

Results

Patient characteristics

Out of 213 STS patients who received radiotherapy from 2000 to 2021, 101 were eligible for inclusion in the analysis (Figure 1). Among the 101 patients, the subset of patients with undifferentiated pleomorphic sarcoma was the largest. Comparison of patient characteristics between the CRT vs. CRTH groups showed no statistical difference (Table 1) This was particularly true for radiation-sensitive (undifferentiated pleomorphic sarcoma, myxofibrosarcoma, synovial sarcoma) versus radiation-resistant histologic subtypes (liposarcoma, leiomyosarcoma, others) as well as for tumor location and diameter. The median follow-up time was 72 months (range 4–172 months).

Table 1. Patient and tumor characteristics.

	N (%)	CRTH	CRT	p-Value
Patients:		n = 68	n = 33
Male	66 (66%)	47 (69%)	19 (58%)	0.18
Female	35 (34%)	21 (31%)	14 (42%)
Age at first diagnosis	55.7 Years (range 19–78 years)	53.9 (19–74 Years)	59.6 (21–78 Years)	0.08
Median follow-up in month	71 Months (range 4–172 months)	70 (4–163)	76 (6–172)	0.57
Tumor:
Histological subtype				0.47
Pleomorphic sarcoma	34 (33%)	21 (31%)	13 (40%)
Myxofibrosarcoma	12 (12%)	9 (13%)	3 (9%)
Synovial sarcoma	10 (10%)	6 (9%)	4 (12%)
Liposarcoma	20 (20%)	18 (26%)	2 (6%)
Leiomyosarcoma	9 (9%)	6 (9%)	3 (9%)
Other subtypes	16 (16%)	8 (12%)	8 (24%)
Location				0.58
Lower Extremity	62 (61%)	43 (63%)	19 (58%)
Upper extremity	15 (15%)	8 (12%)	7 (21%)
Trunk	22 (22%)	15 (22%)	7 (21%)
Head and neck	2 (2%)	2% (3%)	0
Mean diameter	96 mm (Range 6–280 mm)	102 mm (15–280 mm)	85 (6–180 mm)	0.08
Grading				0.30
Gx	5(5%)	4 (6%)	1 (35)
G1	5 (5%)	4 (6%)	1 (3%)
G2	34 (34%)	24 (35%)	10 (30%)
G3	57 (56%)	36 (53%)	21 (64%)
Tumor stage				0.20
Tx	3 (3%)	3 (4%)	0
T1	21 (21%)	11 (16%)	10 (30%)
T2	68 (67%)	45 (66%)	23 (70%)
T3	5 (5%)	5 (7%)
T4	4 (4%)	4 (6%)
UICC 2017				0.20
I	5 (5%)	4 (6%)	1 (3%)
II	18 (18%)	9 (13%)	9 (27%)
IIIA	63 (62%)	42 (62%)	21 (64%)
IIIB	8(8%)	7 (10%)	1 (3%)
Whoops procedure	33 (33%)	19 (58%)	14 (42%)	0.146
Primary biopsy	68 (68%)	49 (72%)	19 (28%)
Risk group according to Sarculator, macroscopic tumors	68 (100%)	49 (72%)	19 (28%)	0.52
10 y OS ≥60%	32 (47%)	25 (51%)	8 (42%)
10 y OS <60%	36 (53%)	24 (49%)	11 (58%)

Feasibility of treatment

Radiotherapy

Sixty-seven of the 101 patients (67%) received conventionally fractionated radiotherapy. Prior to 2010, patients were treated according to the old protocol as follows: 34 patients (34%) received accelerated radiotherapy during weeks 1 and 5, with intermittent seven-day breaks after a dose of 30 Gy in eight cases (8%) [11,13]. The median radiation dose was 60 Gy. Eighty-one patients (80%) received the planned dose of 60 Gy, and 20 patients (20%) achieved a dose of 50–60 Gy. Only three patients (3%) needed a 1 to 3-day break from radiation due to toxicity. The median dose was 59.7 Gy (range 50–67 Gy) in patients with hyperthermia and 60.2 Gy (range 50–70 Gy) in those without hyperthermia (p = 0.19).

Chemotherapy

Ninety-seven patients (97%) received two cycles of chemotherapy in weeks 1 and 5, and three patients (3%) received >2 cycles of chemotherapy. In one case, chemotherapy was discontinued after one cycle due to hematologic toxicities. Forty patients (40%) received the full dose of ifosfamide (10 doses of 1.5 g/m²). Eleven patients (11%) received 90–99% of the intended dose. In 44 cases (44%), a planned dose reduction to 50–89% of the calculated total dose was needed due to comorbidities. Seventy-eight patients (78%) received the full dose of doxorubicine, nine (9%) received 90–99% of the planned dose, and another nine (9%) received only 75–90% of the total dose due to comorbidities. Patients in the CRTH group received a median of 78% (45–100%) of the per-protocol dose of ifosfamide and 95% (10–100%) of the per-protocol dose of doxorubicin. Similarly, patients in the CRT group received a median of 75% of the per-protocol dose of ifosfamide (30–100%) and a median of 97% (60–100%) of the per-protocol dose of doxorubicin (p = 0.20 for ifosfamide, p = 0.20 for doxorubicin).

Hyperthermia

Sixty-eight (68%) of the analyzed patients received additional hyperthermia. Of these, 50 (50%) received regional deep hyperthermia and the other 18 (18%) received superficial hyperthermia. The median number of hyperthermia sessions was 7 (range: 2–15). Reasons for reducing the planned number of hyperthermia sessions were lack of patient compliance and lack of availability of the system. Of the patients with macroscopic tumors, 64 (94%) received hyperthermia and 4 (6%) did not. This reflects the trend toward neoadjuvant therapy in centers and the increasing use of hyperthermia after publication of the EORTC study in 2018 [8].

Surgical resection and flap reconstruction procedures

En bloc resection or re-resection of the tumor bed, the standard treatment, was performed in all but three patients. In the latter cases, the initial decision to perform re-resection was revised immediately before surgery due to the expected difficulties in achieving plastic reconstruction. Twenty-six patients (26%) needed flap reconstruction for defect coverage. In all cases, primary closure was performed (if necessary, with interim VAC coverage until the histology results were available).

Toxicity profile of neoadjuvant CRTH vs. CRT

Hematologic toxicity

Leukocytopenia was the most common hematologic toxicity (Table 2). Except for anemia, which tended to occur more frequently in the CRTH group (p < 0.0001), we detected no statistical differences in toxicity, especially for higher grade toxicities. No toxicity-related deaths occurred.

Table 2. Toxicity profile.

	All patients (n = 101)	CRTH (n = 68)	CRT (n = 33)	p-Value
Leukocytopenia (grades 1–4)	92 (92%)	62 (91%)	29 (88%)	0.72
Including:				1.0
Grade 3	37 (37%)	23 (37%)	14 (49%)
Grade 4	38 (38%)	27 (39%	11 (38%)
Anemia (grades 1–4)	75 (75%)	68 (100%)	19 (58%)	<0.0001
Including:				0.34
Grade 3	7 (7%)	7	0
Grade 4	0 (0%)	0	0
Thrombocytopenia (grades 1–4)	45 (45%)	33 (48%)	12 (36%)	0.29
Grade 3	5 (5%)	3	2	1.0
Grade 4	5 (5%)	4	1
Elevated creatinine	13 (13%)	6 (9%)	7 (21%)	0.11
Grade 1	10 (10%)	5	5
Grade 2	3 (3%)	1	2
Chemotherapy and radiotherapy
Radiation dermatitis (CTCAE ≥3)	12 (12%)	9 (13%)	3 (9%)	0.74
Nausea and emesis	11 (11%)	6 (9%)	5 (15%)	0.33
Edema	5 (5%)	4 (6%)	1 (3%)	0.37
Constipation	3 (3%)	3 (4%)	0 (0%)	0.55
Superficial thrombosis	2 (2%)	2 3%)	0 (0%)
Neutropenic fever	2 (2%)	1 (1%)	1 (3%)
Fatigue	1 (1%)	1 (1%)	0 (0%)
Hematemesis	1 (1%)	1 (1%)	0 (0%)
Tumor bleeding	1 (1%)	1 (1%)	0 (0%)
Vision impairment	1 (1%)	1 (1%)	0 (0%)
Dysphagia	1 (1%)	1(1%)	0 0(%)
Polyneuropathy	1 (1%)	1 (1%)	0 (0%)
Tumor perforation	1 (1%)	1 (1%)	0 (0%)
Hyperkalemia	1 (1%)	1 (1%)	0 (0%)
Postoperative complications
Wound complications	22 (22%)	17	5	0.46
Infection	7 (7%)	6	1	0.42
Seroma	8 (8%)	4	4	0.43
Persistent pain	5 (5%)	3	2	0.66

Non-hematologic toxicity

The most common non-hematologic complication was radiodermatitis, which occurred in 57 cases (57%). Radiodermatitis was generally mild: CTCAE grade 1–2 in 45 cases (45%) but was more severe (CTCAE grade 3) in 12 cases (12%): nine in the CRTH group vs. three in the CRT group. CTCAE grade 4 radiodermatitis did not occur. There was no difference in the overall frequency of all grades of radiodermatitis combined (40/68; 58% vs. 19/33, 59%) or in the frequency of higher-grade dermatitis between the two groups (p = 1.0 and p = 0.74, respectively)

Postoperative complications

Wound complications were observed in 22 patients and assessed according to the criteria delineated in the NCIC-SR2 trial. Minor wound complications were not detected in any patients, but underreporting of this complication can be assumed. Ten patients developed moderate wound complications: fistula (n = 1), abscess (n = 3), infected seroma or seroma requiring drainage (n = 2), osteomyelitis (n = 2), and protracted wound complications without need for surgical intervention (n = 2). Twelve patients developed major wound complications: wounds with or without abscess requiring surgical debridement (n = 11) and compartment syndrome (n = 1). The CRTH group tended to have a higher rate of overall wound complication compared to the CRT group: CRTH 17/68 patients (25%) vs. CRT 5/33 patients (15%), respectively (p = 0.19). Likewise, severe wound complications occurred more frequently in the CRTH group than in the CRT group: 10/68 (15%) vs. 2/33 (6%), respectively, yet not significant (p = 0.32). The distribution of wound complications relative to relative to tumor location was as follows: upper extremity 3/15 (20%), lower extremity 18/62 (29%), head-neck-torso wall 1/24 (4%). The p-value for trunk vs. lower extremity was p = 0.018.

Resection quality

The results for resection quality were as follow: R0 (n = 88/101; 87%), R1 (n = 8/101,8%), R2 (n = 1; 1%), and Rx (n = 4, 4%). Among the subset of patients with macroscopic tumor prior to the start of CRT, R0 was achieved in 62/68 cases (91%) and R1/R2/Rx in 6/68 cases (9%); this was non-significantly higher than in patients who underwent re-resection after primary tumor resection before chemoradiotherapy (R0: 26/33; R1/R2/RX: 7/33, p = 0.122).

Extent of disease remission

For the overall population, remission data were available in 92 cases and missing in 9 cases. In the subset of 33 patients who had undergone primary resection, 7 (21%) were excluded from the analysis because they did not undergo re-resection despite initial consent or due to the lack of regression data. Of the remaining patients, 12/33 (36%) had no residual tumor or complete tumor regression, 5/33 (15%) had ≥90% tumor regression, 6/33 (18%) had 60–80% tumor regression, and 3/33 (9%) had ≤ 50% tumor regression.

In the subset of 68 patients with macroscopic primary tumors, the remission status could not be determined in 2 cases (3%) because the surgery was performed elsewhere. Analysis of the remaining 66 patients revealed that percentage regression was ≥90% in 25/68 (37%) cases, between >50% and <90% in 21/68 cases (31%), and <50% in 20 cases (29%).

Remission after chemoradiotherapy with and without hyperthermia

In patients with macroscopic soft tissue sarcomas, adding hyperthermia resulted in a tendency toward greater tumor regression: ≥90%: 21/48 (43%), 50%-≤90%: 15/48 (31%) and <50% regression: 12/48 (25%) with CRTH versus ≥90% 4/18 (22%), 50%–≤90%: 6/18 (33%) and <50% regression: 8/18 (44%) with CRT alone (p = 0.197). The number of patients with undifferentiated pleomorphic sarcoma (UPS) tended to be lower in the CRTH group than in the CRT group: 16/48 (33%) vs. 9/18 (50%), respectively. Patients with pleomorphic sarcomas had a mean regression grade of 79% after CRTH (n = 15) compared to 67% without hyperthermia (n = 10), p = 0.20.

Local recurrence

Overall, 15/101 (15%) patients developed local recurrences within a mean of 61 months (range 4-174 months). There were no regional recurrences. Local recurrence-free survival (LRFS) was 84.5%±4.2% at 5 years and 75.3%±6.4% at 10 years; the median follow-up time was 72 months. In the overall population, none of the analyzed parameters correlated with the local recurrence rate: primary surgery vs. primary CRT (p = 0.931), gender (p = 0.993), age <60 vs. age ≥60 (p = 0.240), radiation-sensitive vs. non-radiation-sensitive tumors with a classification (p = 0.430), tumor location (p = 0.640), size ±10cm (p = 0.20), grade G1/Gx vs. G2/3 (p = 0.820), T1/2/x vs. T3/T4 (p = 0.80), and UICC stage I/II vs. IIIA/B (p = 0.78). Likewise, there was no difference in LRFS between CRTH and CRT (85.1% ± 5.0% vs. 83.7% ±7.5%, p = 0.938). The subset of 68 patients with macroscopic tumors showed a trend toward better local control after CRTH in comparison to CRT after 6 and 10 years: 86.6%±5.7% vs. 66.3%±14.2% (p = 0.420), respectively (Figure 3). High-risk patients, identified via Sarculator (n = 36), [19,20], see Table 1, showed a 5-year LRFS of 95%±4.9% (CRTH, n = 25) vs. 83.3%±15.2% (CRT, n = 11), (p0.170).

Figure 3. Local recurrence free survival (LRFS), macroscopic tumors only, p0.420.

Figure 4. Overall survival (OS), macroscopic tumors only, p0.364.

Distant metastasis

Distant metastasis was diagnosed in 26/101 patients at a median of 60 months (range 2-172 months). DMFS was 73.1 ± 4.7% at 5 years and 66.1 ± 6.6% at 10 years. Tumor size correlated with freedom from distant metastases: 5-year DMFS was 85.4%±5.1% for small tumors and 58.0%±8.7% for tumors ≥ 10 cm (p = 0.005). Tumor size was an independent prognostic factor DMFS, with a hazard ratio (HR) of 3.2 (p = 0.007). Univariate analysis revealed differences for tumor grade (p = 0.025), T stage (p = 0.029), and UICC stage (p = 0.025). Hyperthermia had no statistical effect. The 5-year DMFS was 72.2%±5.8% for CRTH and 75.4 ± 8.3% for CRT alone (p = 0.698). In the 68 patients with macroscopic tumors, there was no difference in DMFS between the CRTH and CRT (5 y: 68.5 ± 7.0% vs. 62.7 ± 12.6%, p = 0.998). High-risk patients, identified via Sarculator (n = 36), [19,20] (see Table 1), showed a 5-year DMFS of 67.9%±10% (CRTH, n = 25) und 50.5%±15.8% (CRT, n = 11) (p0.540).

Disease-free survival

Thirty-seven failure events occurred at a median of 54.6 months (2-172 months). DFS was 63.4% ± 5.2% at 5 years and 49.9% ± 7.1% at 10 years. None of the analyzed factors correlated with disease-free survival. Five-year DFS was 64.9%±6.2% for CRTH vs. 61.4%±9.3% for CRT alone (p = 0.988). In the subset of patients with macroscopic tumors, DFS was 62.9%±7.4% for CRTH compared to 46.1%±12.1% for CRT (p = 0.524). High-risk patients, identified via Sarculator (n = 36) [19,20], see Table 1, showed a 5-year DFS of 66.3%±10.4% (CRTH, n = 25) and 40.4%±15.5% (CRT, n = 11) (p0.217), see supplementary material.

Overall survival

Seventy-two out of 101 patients were alive on the cutoff date for the analysis or the 72-month follow-up date. The overall survival (OS) rate was 76.6%±4.6% at 5 years and 67.4%±5.6% at 10 years. There were no inter-group differences in overall survival for any of the parameters. Patients with primary resection (‘whoops’) had OS rates similar to those of patients with primary CRT had (p = 0.792). Five-year overall survival was 81.6%±5.1% vs. 71.9%±8.5% (p = 0.215) in patients receiving CRT with and without hyperthermia, respectively. In the subset of patients with macroscopic tumors, CRTH showed a tendency toward 5-year overall survival compared to CRT: 81.5%±6.0% vs. 63.3 ± 12.1% (p = 0.364), respectively (Figure 4). High-risk patients, identified via Sarculator (n = 36) [19,20] (Table 1), showed a 5-year OS of 87.2%±7% (CRTH, n = 25) and 60%±15.5% (CRT, n = 11) (p0.230), see supplementary material.

Discussion

Three insightful conclusions about neoadjuvant CRTH can be drawn from this retrospective analysis: First, neoadjuvant concurrent chemoradiotherapy plus hyperthermia is feasible, despite the addition of doxorubicin to ifosfamide during radiotherapy. Second, as assessed using established criteria, wound complication rates associated with neoadjuvant CRTH were as expected and manageable, albeit slightly higher and clearly more focused in the distal limb area compared with neoadjuvant CRT alone. Third, the evidence suggests that adding hyperthermia to the neoadjuvant CRT regimen may have clinical benefit in terms of remission induction, but not in distant metastasis in accordance with Prosnitz et al. [21].

In patients with macroscopic disease, additional hyperthermia led to a trend toward greater remission in the CRTH vs. CRT group comparison. Especially in high-risk patients according to the Sarculator [19], adding hypertermia suggested a benefit in loca control, DFS and overall survival.

This research addresses the need to devise new strategies to optimize the timing of neoadjuvant treatments for soft tissue sarcomas, which has emerged in recent years as previous studies have shown essentially no improvement of efficacy with neoadjuvant chemotherapy versus standard chemotherapy with doxorubicin plus ifosfamide [6]. At best, promising results were obtained by combining dacarbazine with doxorubicin instead of ifosfamide for the treatment of leiomyosarcoma [22] or Trabectedin, respectively ifosfamide plus Pazopanib with radiotherapy [23–25]. Another study demonstrated the benefit of adding deep regional hyperthermia to perioperative chemotherapy [8]. The benefit of neoadjuvant radiotherapy is well documented. The rationale behind administering all three treatment modalities concurrently is not only based on radiobiological considerations but can also serve to prevent the interruption of treatment during chemotherapy-free intervals before resection. Based on data collected over many years of experience treating STS patients in routine clinical practice, we successfully demonstrated that combining radiotherapy, chemotherapy and hyperthermia in the neoadjuvant setting is possible and feasible. Our data show that basically no reduction of the radiation dose is needed for neoadjuvant CRTH. By and large, it was also possible to administer chemotherapy at the scheduled times and doses and, as expected, hematologic toxicity was the most common complication of chemotherapy. As expected, these results are completely in agreement with the toxicity profile observed in our last study and provide further evidence of the feasibility of concurrent chemoradiotherapy for the treatment of advanced soft tissue sarcoma [11]. Unlike the first RTOG study [26], no treatment-related deaths occurred in the present study. This also confirms the Italian-Spanish researcher group’s conclusion regarding the feasibility of preoperative chemoradiotherapy [10]. The addition of hyperthermia had no major effect on the side effect profile, although anemia was slightly increased. No increase in the rates of leukopenia, thrombocytopenia, and skin toxicity was observed. These results concerning toxicity data are in line to those of the randomized EORTC study, which, however, could show a survival benefit [8]. Thus, dermatitis remains the leading non-hematologic side effect of both treatment modalities. With the established supportive therapy options, the treatment regimen was feasible.

There is a potential to minimize the rates of both radiation dermatitis and wound complications by reducing the radiation volume. Studies by O'Sullivan [18] and Wang et al. [27] show that a moderate decrease in the incidence and/or severity of radiotherapy-related wound complications can be achieved by avoiding normal tissues potentially required for wound closure. Moreover, the addition of hyperthermia to the chemoradiation protocol did lead to slightly increased worsening of wound complications in our patients. Complications stay in the expected range, this confirms the observations of Potkrajic et al. [28] However, a weak point of our study is the lack of long-term data on quality of life and function because only information regarding toxicities was collected in the follow-up visits according to RTOG toxicity criteria and CTCAE 5.0 criteria [17].

Also, and in our opinion most importantly, our findings suggest that adding hyperthermia to the CRT regimen improves remission. Since macroscopic tumor regression was basically assessed by a single pathologist, it can be assumed that uniform assessment standards were applied in each case. This is all the more significant in view of the lack of uniform assessment standards. Moreover, the evaluating pathologist had no knowledge of the intensity of treatment. To address the problem of potential differences in the distribution of immunologically addressable sarcomas between the different treatment groups, we performed an isolated analysis of pleomorphic sarcomas. In the subgroup of patients with undifferentiated pleomorphic sarcoma, CRTH led to deeper remission than CRT alone. This could very well translate into better local control, making CRTH an attractive option for patients with sarcomas of questionable complete resectability. Appropriate tumor selection criteria for this should be developed Especially in high-risk patients according to the Sarculator [19,20], the addition of hypertermia in locoregional control, DFS, and overall survival indicated an advantage. Possibly, high-risk patients do benefit from CRT and especially CRTH. In comparison to Gronchi et al. [29], who could achieve a 5-year DFS of 45% in this patient group, Patients showed a DFS of 57.9% after CRT and 63.8% after CRTH. This was partially shown in the overall survival of patients, here the 5-year overall survival was 67% for Neoadjuvant Chemotherapy, 66.7% (CRT) vs. 83.7%(CRTH). At this moment, we cannot demonstrate a significant difference in oncological endpoints due to the heterogeneity of the patient group analyzed and their small number. However, there is a trend, especially for local control of macroscopic tumors in the subgroup analysis. A final observation is that distant metastasis rates for neoadjuvant CRTH were high yet comparable to those reported in other neoadjuvant treatment studies. This is not surprising, since the applied chemotherapy dose was low and it is known that the addition of hyperthermia alone to radiotherapy does not provide systemic effect [21].

This problem remains to be addressed. We found that the risk of distant metastasis is higher with large than with small soft tissue sarcomas. This is not surprising considering that this was observed in all populations used as the basis for risk stratification models for chemotherapy [19,20]. As expected, the intensification of treatment resulted in a local effect. The challenge for the coming years is to expand and steer this treatment concept in the direction of total neoadjuvant therapy by incorporating higher doses of chemotherapy, for example an integration of at least 3 up to 6 cycles of neoadjuvant chemotherapy. We conclude that, in the interim, chemoradiotherapy plus hyperthermia is a viable option for the treatment of borderline resectable soft tissue sarcomas in the neoadjuvant setting. As a next step, integration of this combination therapy as a total neoadjuvant approach should be further assessed in clinical trials.

Ethical approval

This study was carried out in accordance with all relevant guidelines and regulations. Informed consent to treatment was obtained from all patients. Patient consent for retrospective study participation was waived based on local legislation (BayKrG Art. 27 (4)) and institutional policy; formal consent was not required for retrospective study participation.

Acknowledgements

Alexander Willner performed research for the present work in fulfillment of the requirements for obtaining a doctoral degree in medicine (Dr. med.). We acknowledge financial support by Deutsche Forschungsgemeinschaft and Friedrich-Alexander-Universität Erlangen-Nürnberg within the funding programm “Open Access Publication Funding”.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets used and/or analyzed in the current study are available from the corresponding author on reasonable request. Please contact the corresponding author ([email protected]).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.