Hand sarcomas: a systematic review investigating the role of radiotherapy in non-metastatic patients

Research question and objectives

The central research question was: What is the most effective treatment for non-metastatic hand sarcoma, and what is the role of radiotherapy? The objectives were threefold: (1) summarize treatment options for non-metastatic hand sarcoma; (2) compare outcomes, including overall survival (OS), disease-free survival (DFS), local control (LC), treatment-related toxicity, and quality of life (QoL); (3) evaluate the risk-benefit profile of radiotherapy. The research framework was structured around the PICO model, wherein the Population included patients with hand sarcoma. The Intervention and Comparison components encompassed surgical treatment, radiotherapy, and their combinations with chemotherapy. These interventions were systematically compared to evaluate differences across the aforementioned clinical Outcomes.

Search strategy

A comprehensive search was conducted across PubMed, Cochrane Central Register of Controlled Trials, and Scopus, from database inception to October 31, 2024. The search string used was: (“sarcoma of the hand” OR “hand sarcoma”) AND (“surgery” OR “surgical” OR “amputation” OR “radiotherapy” OR “radiation therapy”). Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) was used to manage the review process, including result organization, screening, and full-text review.

Eligibility criteria

Studies were included if they involved patients with non-metastatic hand sarcoma and reported at least one primary outcome (OS, DFS, LC, toxicity, QoL). Eligible study designs included randomized controlled trials, cohort studies, case-control studies, case series, and case reports, published in English. Exclusion criteria included non-human subjects, unavailable full texts, systematic reviews, editorials, conference abstracts, and studies with insufficient follow-up or focusing on pediatric populations.

Study selection

The selection process included two phases, conducted independently by two reviewers (Valentina Zagardo and Gianluca Scalia) with discrepancies resolved by a third reviewer (Gianluca Ferini). In the first phase, 105 studies were retrieved, and 33 were excluded after title and abstract screening. In the second phase, 72 full-text studies were reviewed, with 47 excluded, leaving 25 eligible studies.

Data extraction

Data were extracted independently by Valentina Zagardo and Gianluca Scalia using a standardized form, capturing key details: study characteristics (year, author, country), patient demographics (age, comorbidities, functional status), clinical presentation (tumor size, location, symptoms), diagnostic information (biopsy, imaging), treatment details (surgery, radiotherapy, chemotherapy), and outcomes (follow-up, survival, recurrence, toxicity, QoL). Disagreements were mediated by Gianluca Ferini.

Risk of bias assessment

The quality of included studies was assessed using the Joanna Briggs Institute checklists for observational studies, case reports, and randomized controlled trials. The risk of bias was independently assessed by the two primary reviewers, with mediation by Gianluca Ferini in case of disagreement.

Data synthesis and analysis

Data were synthesized qualitatively and quantitatively. Case reports were grouped by treatment modality and outcomes (OS, DFS, LC), and analyzed using Kaplan-Meier curves, each accompanied by the number at risk, censored patients, and events at each time point, thereby offering insight into the sample size and the reliability of the results, particularly when small group sizes may influence the statistical significance of differences. Statistical comparisons between groups were made with log-rank tests, with a p-value <0.05 considered significant. Cohort analysis was not performed due to data limitations.

Case reports

We identified 10 case reports, 3 case series (9 patients total), and extracted data from seven cohorts, including a total of 106 patients. The mean age was 44.3 years (range 18–82), with 48 females (45.28 %) and 36 males (33.96 %), while 22 patients’ sex was not reported. Tumors were primary in 60 cases (56.60 %), recurrent in 23 (21.69 %), and not reported in 23 (21.69 %). Tumors were located in the hand (48), hand and fingers or wrist (4), and wrist or fingers (remaining cases).

Histological subtypes were reported in 104 cases, with synovial sarcoma (16.98 %), epithelioid sarcoma (14.15 %), and chondrosarcoma (14.15 %) being the most common (Table 1). High-grade tumors accounted for 48 cases (45.28 %), intermediate-grade for 22 (20.75 %), and low-grade for 16 (15.09 %). Tumor grade was unspecified in the remaining cases (Table 1).

Functional outcome details

The Extremity Performance Scale (EPS) was used in one study, where a decline of at least 2 points was recorded in three of 14 patients [3]. Except for one patient who received radiotherapy alone, the other 2 underwent both surgery and radiotherapy. The EPS was not assessed in 2 patients due to amputation and was not reported in one.

The Musculoskeletal Tumor Society (MSTS) score was used in three studies. A score below 29 (“good” threshold) was recorded in nine of 38 patients. It was not reported or available in 7 patients. 31 of 38 patients received radiotherapy [12], 19], 20].

The Disabilities of the Arm, Shoulder, and Hand (DASH) score was used in one study, with three patients scoring above 40 %, indicating poor performance, all of whom received radiotherapy [20]. In 16 other studies, the outcome was not reported in six patients [11], [21], [22], [23], [24], [25], not evaluated in 2 (due to amputation) [26], 27], and no detrimental effects on quality of life were reported in the remaining cases. The latter ones included 7 (out of 16) patients who received adjuvant radiotherapy [28], [29], [30], [31], [32], [33], [34], [35].

Due to the heterogeneity of scales used, no correlation between treatment and functional outcomes could be established.

Table 2 provides a detailed overview of the treatments administered to patients with hand sarcoma, categorized by histological subtype, with a focus on functional outcomes.

Table 2:

Details of treatments and outcomes reported in case reports of hand sarcomas.

Cancer histology	Treatment	Quality of Life	Local control
Fibrosarcoma
Okunieff et al. [3]	R
Thumser et al. [24]	S
Frassica et al. [34]	S
Mirous et al. [19]	S + R + C
Puhaindran et al. [20]	S + R
Puhaindran et al. [20]	S + R
Alveolar soft part sarcoma
Okunieff et al. [3]	S + R
Troum et al. [21]	R + C + S
Synovial sarcoma
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Bray et al. [12]	S
Thumser et al. [24]	S + C
Thumser et al. [24]	S + C + R
Thumser et al. [24]	S + C
Mirous et al. [19]	S + C
Mirous et al. [19]	S + C
Mirous et al. [19]	S + C
Mirous et al. [19]	S + R
Puhaindran et al. [20]	S + R
Serinelli et al. [11]	S + C + R
Puhaindran et al. [20]	S + R
Nakajima et al. [32]	S
Fleegler et al. [27]	S
Fleegler et al. [27]	S
Epithelioid sarcoma
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Bray et al. [12]	R + S
Bray et al. [12]	S
Thumser et al. [24]	S + R + C
Thumser et al. [24]	S + R + C
Thumser et al. [24]	S
Thumser et al. [24]	S + R
Thumser et al. [24]	S + R + C
Mirous et al. [19]	S + R
Mirous et al. [19]	S + R
Patel et al. [35]	S + R + C
Patel et al. [35]	S + R
Fleegler et al [27]	S
Malignant fibrous histiocytoma
Okunieff et al. [3]	S + R
Puhaindran et al. [20]	S + R
Puhaindran et al. [20]	S + R
Bray et al. [12]	S
Fleegler et al. [27]	S
Neurofibrosarcoma
Okunieff et al. [3]	S + R
Okunieff et al. [3]	S + R
Liposarcoma
Okunieff et al. [3]	S + R
Mirous et al. [19]	S
Puhaindran et al. [20]	S + R
Leiomyosarcoma
Bray et al. [12]	R + S
Mirous et al. [19]
Thumser et al. [24]	S + C
Thumser et al. [24]	S
Mirous et al. [19]	S + R
Clear cell sarcoma
Bray et al. [12]	R + S
Thumser et al. [24]	S
Thumser et al. [24]	S + C
Mirous et al. [19]	S + C
Spindle cell sarcoma
Thumser et al. [24]	S + R + C
Thumser et al. [24]	R + C
Vijayan et al. [28]	S
Solitar fibrous tumor
Thumser et al. [24]	S + C
Peripheral sheat tumor
Mirous et al. [19]	S + R
Pleomorphic sarcoma
Mirous et al. [19]	S + R
Undifferentiated sarcoma
Mirous et al. [19]	S + R
Persitz et al. [23]	S + R
Rhabdomyosarcoma
Johnstone et al. [29]	C + R
Li et al. [30]	S + C + R
Osteosarcoma
Bray et al. [12]	R + C + S
Frassica et al. [34]	S
Frassica et al. [34]	S
Puhaindran et al. [26]	C + R + C
Sadiq et al. [31]	S
Ewing’s tumor
Thumser et al. [24]	S
Johnstone et al. [29]	R + C
Johnstone et al. [29]	R + C
Johnstone et al. [29]	R + C
Johnstone et al. [29]	S + C
Euler et al. [25]	R + C + S
Ozaki et al. [22]	R + C + S
Anakwenze et al. [33]	S + R + C
Anakwenze et al. [33]	S + C
Anakwenze et al. [33]	S + R + C
Chondrosarcoma
Thumser et al. [24]	S
Thumser et al. [24]	S
Thumser et al. [24]	S
Thumser et al. [24]	S
Thumser et al. [24]	S
Thumser et al. [24]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Frassica et al. [34]	S
Hemangioendothelial sarcoma
Frassica et al. [34]	S
Frassica et al. [34]	R
Frassica et al. [34]	R
Frassica et al. [34]	S
Not reported/not available
MSTS score (0–35 points)
TESS score (0–100 points)
DASH score (0–100 %)
EPS scale (1–6 points)
VAS score (0–10)
Patient-reported QoL

1. Histological subtypes are listed in the first column. Treatments are reported in the chronological order in which they were administered. Multiple rows corresponding to the same authors represent different patients. In the “quality of life” column, the specific assessment scales used for each patient are indicated, including those related to hand functionality, as detailed in the table legend. The extent of filling or the degree of shading within each bar visually represents the reported outcome for each patient, where available. Green, yellow, and red smiley faces denote mild, moderate, and severe post-treatment pain, respectively, when such information was reported. Red, yellow, and green circles indicate the reported quality of life, categorized as poor, well, and excellent, respectively. In the “Local Control” column, a green bar indicates successful local control, a red bar denotes local failure, and a white bar indicates that data were not reported or were unavailable. S, surgery; R, radiotherapy; C, chemotherapy; MSTS, musculoskeletal tumor society scoring system; TESS, toronto extremity salvage score; DASH, disabilities of the arm, shoulder and hand; VAS, visual analogue scale; EPS, extremity performance scale; QoL, quality of life.

Survival analysis (treatment modalities and oncological outcomes: comparative analysis)

Patient categorization and analytical approach

Patients were categorized based on the treatment modalities performed. Various treatment combinations were analyzed to understand the role of each therapy, with particular emphasis on radiotherapy, in managing hand sarcoma. Overall, against a notably long median follow-up of 54 months (range 4–250 months), the median values for survival outcomes (LC, DFS, and OS) were not reached, highlighting a more favorable prognosis compared to sarcomas located elsewhere. This improved prognosis is likely attributable to the ease of detecting abnormalities in an anatomical site as readily explorable as the hand.

Comparison: surgery vs. no surgery

Surgical resection remains the cornerstone of treatment for hand sarcoma. Patients who underwent surgery had significantly better OS compared to those who did not receive surgical intervention (Figure 2). This highlights the critical role of surgery in achieving optimal oncologic outcomes. The evaluation of LC and DFS in non-surgical patients, with a very small sample size (n=8), is unreliable and unsuitable for meaningful interpretation.

Figure 2:

Kaplan-Meier curves describing the overall survival of patients undergoing surgical treatment or not.

Comparison: surgery + radiotherapy vs. surgery alone

Surgery followed by radiotherapy non-significantly improved LC and OS, but had no effect on DFS (Figure 3).

Figure 3:

Survival curves for patients undergoing surgery + radiotherapy vs. surgery alone. (A) Local control, (B) disease-free survival, (C) overall survival.

Comparison: surgery + radiotherapy vs. surgery + chemotherapy

The addition of radiotherapy to surgery resulted in improved LC, though not statistically significant, compared to surgery combined with chemotherapy. However, DFS and OS were significantly better in the surgery + radiotherapy group (Figure 4). This suggests that radiotherapy may help mitigate systemic disease progression, while LC remains highly dependent on surgical success.

Figure 4:

Survival curves for patients undergoing surgery + radiotherapy vs. surgery + chemotherapy. (A) Local control, (B) disease-free survival, (C) overall survival.

Comparison: surgery + chemotherapy + radiotherapy vs. surgery + chemotherapy

The addition of radiotherapy to surgery and chemotherapy improved LC, DFS, and OS (Figure 5), though not statistically significant.

Figure 5:

Survival curves for patients undergoing surgery + radiotherapy + chemotherapy vs. surgery + chemotherapy. (A) Local control, (B) disease-free survival, (C) overall survival.

Comparison: surgery + radiotherapy + chemotherapy vs. surgery + radiotherapy

In patients receiving multimodal therapy (surgery + radiotherapy + chemotherapy), LC, OS, and DFS (Figure 6) were worse, though not significantly, compared to surgery + radiotherapy alone. This likely reflects that chemotherapy is often used in patients with poor prognostic factors, increasing the risk of recurrence and metastasis. Chemotherapy toxicity may also contribute to worse outcomes. These findings highlight the need for further studies to clarify its role before routine use in post-surgical management.

Figure 6:

Survival curves for patients undergoing surgery + radiotherapy + chemotherapy vs. surgery + radiotherapy. (A) Local control, (B) disease-free survival, (C) overall survival.

Cohort analysis

Five retrospective cohort studies meeting the inclusion criteria involved 254 patients in total [36], [37], [38], [39], [40]. Among them, 135 were male and 119 were female. Tumors were classified as primary in nine cases, recurrent in 15, and unreported in the others due to omission in three studies.

Tumor localization was variably reported. Two studies provided only laterality (right hand in 16 cases and left hand in nine in one study; 92 % of tumors located in the right hand in another). Three studies specified anatomical locations:

1. Study 1 [36]: 33 deep (subfascial) tumors, 24 in the metacarpals, 18 in the wrist, and nine in the fingers.

2. Study 2 [37]: 41 tumors in the metacarpals, 18 in the phalanges, and five in the carpal extension.

3. Study 3 [39]: a continuation of Study 2, involving the same cohort, with five additional metacarpal tumors.

Tumor histology was reported in all studies, with epithelioid sarcoma as the most frequent subtype (Tables 3 and 4). However, TNM classification was not documented in four of the five studies.

Only two studies reported QoL [38], 40]. In the first study [38], median QoL was reported as good, while in the second [40], overall hand function, activities of daily living, and patient satisfaction were significantly lower than those of the contralateral healthy hand.

A detailed breakdown of patient demographics, tumor localization, histological subtypes, treatment patterns, and outcomes is provided in Tables 3 and 4. These findings highlight significant variability in tumor characteristics and management approaches.

The inability to extrapolate granular data on treatment modalities and outcomes prevented an in-depth analysis of potential correlations.

The key findings outlined above are summarized in Figure 7.

Figure 7:

Key findings from our analysis.

Epidemiology

Hand sarcomas, which may arise from either soft tissues or bone, are exceedingly rare, accounting for only 1 % of all sarcomas [41]. Among soft tissue sarcomas, approximately 2 % occur in the hand [1], whereas bone sarcomas are even rarer, with osteosarcoma representing just 0.2 % of cases [31]. In the studies reviewed, the mean age of patients with hand sarcomas was 44.3 years (range 18–82), with a slight female predominance (45.28 % women vs. 33.96 % men) in case reports. In cohort studies, the mean age varied, with values of 46.1 years in one study and its continuation, 36 years in another, and 41 and 48 years in the remaining two. Male predominance was observed across these studies. The rarity of these tumors presents unique diagnostic and treatment challenges, often leading to delays in diagnosis and inappropriate initial management, such as inadvertent excisions [12], 42]. Nicholson et al. reported only 17 cases of soft tissue sarcoma of the hand and wrist over a 15-year period in their regional sarcoma service, further emphasizing the rarity of this condition [43].

Histologic subtypes

Hand sarcomas encompass a wide range of histological subtypes. The most frequent are synovial sarcoma (16.98 %), epithelioid sarcoma (14.15 %), and chondrosarcoma (14.15 %). Other notable subtypes include fibrosarcoma, alveolar soft part sarcoma, malignant fibrous histiocytoma, and leiomyosarcoma. Nicholson et al. found that alveolar rhabdomyosarcoma, synovial sarcoma, and myxofibrosarcoma were the most common in their cohort [43]. Epithelioid sarcoma is particularly notable for its high recurrence rate and tendency for late metastases [44]. This histological variability influences treatment decisions and the likelihood of recurrence. The histologic subtype directly correlates with tumor behavior, metastatic potential, and response to adjuvant therapies, necessitating a tailored approach for each case [10].

Clinical presentation

Patients typically present with a painless, enlarging mass [21]. Other symptoms may include restricted joint movement or digital deformities, especially in parosteal osteosarcomas, which can lead to swan neck deformities [45]. Misdiagnosis is common due to the overlap in presentation with benign lesions, often leading to multiple surgeries before a correct diagnosis is made. Nicholson et al. reported that 16 of 17 tumors presented as a painless lump, with only two patients reporting discomfort [43]. Early recognition of symptoms and timely use of advanced imaging techniques are crucial to avoiding unnecessary interventions and ensuring early, definitive management.

Diagnostic approach

Diagnosis involves clinical evaluation, imaging, and biopsy. Magnetic resonance imaging (MRI) is critical due to its high contrast resolution, allowing for precise tumor mapping [20], 22]. The ‘string sign’, seen in parosteal osteosarcoma, is a classic radiological hallmark [45]. Biopsy, whether through fine-needle aspiration or core needle sampling, confirms the diagnosis and identifies the histologic subtype. Given the risk of misdiagnosis, early referral to a specialist center is essential [46]. Recent advances in imaging, such as functional MRI and positron emission tomography (PET), enhance diagnostic precision by distinguishing between benign and malignant soft tissue masses [47]. Some authors emphasize the importance of comprehensive imaging and biopsy before surgical excision, as several cases were misdiagnosed as benign lesions prior to formal pathological analysis [43], 48].

Treatment modalities

Management of hand sarcomas is complex due to the need for oncologic control and functional preservation. The treatment typically involves surgical resection, with or without adjuvant therapies, and reconstructive procedures to restore hand function.

Surgical management

In the case reports, 98 out of 106 patients (92.45 %) underwent surgical intervention, with wide local excision being the most common procedure (32 cases). Other procedures included partial or ray amputation, disarticulation, and extensive limb-sparing surgery. Nicholson et al. found that 13 of 17 patients achieved R0 (negative) margins, with only 2 requiring amputation due to local recurrence [43]. Advances in surgical planning, such as intraoperative navigation and 3D imaging, have improved surgical precision and reduced recurrence risk [49].

In the cohort series, surgery was performed in all patients, except for one study where 80 % of patients received surgery. Procedures ranged from wide local excision to ray or partial amputation, with fewer below-elbow or above-elbow amputations.

Neoadjuvant/adjuvant therapies

In the case reports, radiotherapy was used as adjuvant therapy in 44 cases and as neoadjuvant in 3 cases. Chemotherapy was used as adjuvant therapy in 24 patients, neoadjuvant in 6, and both adjuvant and neoadjuvant in three patients. In contrast, in the cohort series, 12 patients received radiotherapy as neoadjuvant, 29 as adjuvant, and in 61 cases, it was not specified. Chemotherapy was administered as neoadjuvant therapy in 21 patients, adjuvant in 8, and was not reported for 16 patients.

Radiation therapy has historically been critical in reducing local recurrence, especially for high-grade tumors or when achieving negative surgical margins is challenging. From our analysis of case reports, we could not demonstrate a significant improvement in LC with the addition of radiotherapy compared to chemotherapy. However, radiotherapy significantly improved DFS and OS, underscoring its role in mitigating disease spread. Furthermore, an analysis of the “number at risk” in Figure 4 reveals a higher proportion of censored data in the small surgery + chemotherapy group for LC, compared to DFS and OS. This higher proportion of censored observations may explain the lack of a statistically significant difference in LC when compared to the surgery + radiotherapy group.

Yet, the lack of LC benefit from radiotherapy may stem from an imbalance in patient characteristics between the radiotherapy plus surgery and chemotherapy plus surgery groups. Specifically, 36.36 % of patients in the chemotherapy group underwent amputation, compared to only 12.76 % in the radiotherapy group. This higher amputation rate in the chemotherapy group may contribute to superior LC in these patients, possibly explaining the observed difference in LC.

Regarding adjuvant radiotherapy, Alektiar et al. [50] demonstrated that while radiotherapy improved LC, it did not affect OS or DFS. This aligns partially with our results, where radiotherapy added to surgery improved LC and OS – though not statistically significant, possibly due to prognostic imbalances – but did not affect DFS. Alektiar et al. suggested that unrecognized metastatic disease at presentation could explain this discrepancy. In our cohort, no patients had metastatic disease at diagnosis, though undetected metastases may have influenced the outcomes.

While adding radiotherapy to surgery can improve outcomes, it is associated with significant complications, including fibrosis, joint stiffness, and functional impairment of the extremities, especially when used as adjuvant therapy, as noted by Schoenfeld et al. [51]. Neoadjuvant radiotherapy, intended for tumor downstaging, may lead to surgical wound healing failure [52], and in some cases, necessitate a second surgery due to complications it induces [38]. Furthermore, radiation exposure can lead to radiation-induced sarcomas, which are rare but typically high-grade and have a poor prognosis [53].

Radiotherapy is also effective in cases of R2 resection, offering disease stability post-treatment [54]. However, increasing radiation doses to achieve more satisfactory results in R2 resections may result in more side effects compared to typical doses used in adjuvant or neoadjuvant settings [55], 56].

Radiotherapy appears to provide significant benefits in managing hand sarcomas, particularly in cases with a high risk of recurrence, such as positive margins, aggressive tumor features, or situations where surgery is not feasible due to unresectability or patient refusal. In such cases, spatially fractionated radiotherapy may offer control of the bulky unresectable disease [57], 58], as seen in other cancer histologies [59], 60]. Generally, combining radiotherapy with surgery is a key strategy for improving LC, DFS, and OS.

In our analysis, the addition of chemotherapy to surgery resulted in significantly worse outcomes in terms of DFS and OS compared to the combination of radiotherapy and surgery, with no differences observed in LC. Similarly, chemotherapy added no benefit to the combination of surgery and radiotherapy, neither in LC, DFS, nor OS. This finding is consistent with the conflicting evidence in the literature regarding the efficacy of chemotherapy in soft tissue sarcomas. For instance, Brunello et al. [61] reported a 10-year mono-institutional experience in which a mixed-tumor site cohort of patients receiving adjuvant chemotherapy exhibited a statistically significant improvement in DFS, albeit at the cost of a high incidence (55.6 %) of severe (≥G3) toxicities, which in some cases required treatment modifications or even discontinuation. Moreover, they noted that, under the same therapeutic intervention (adjuvant chemotherapy), the median DFS was markedly higher among patients with limb or girdle disease compared to those with other primary sites (82.4 months vs. 18.3 months). This finding may suggest a lower propensity of limb sarcomas, including those of the hand, to metastasize compared to trunk sarcomas, which could explain the absence of benefit from chemotherapy in hand sarcomas observed in our study. In contrast to Brunello et al. [61], but consistent with our findings, Le Cesne et al. [62] reported no improvement in OS with adjuvant chemotherapy across any studied subgroups, instead emphasizing the critical role of successful surgical intervention.

Certain tumor histologies are more responsive to chemotherapy. For example, clear cell sarcoma and alveolar soft part sarcoma are highly chemoresistant, while liposarcoma and synovial sarcoma exhibit greater chemosensitivity [63]. Despite their chemosensitivity, synovial sarcomas are associated with a high risk of distant metastasis [11]. These findings highlight the need for tailored treatment plans based on margin status, tumor aggressiveness (e.g., grading), and the chemosensitivity and metastatic potential of each sarcoma subtype. Such personalized approaches could optimize therapeutic outcomes.

Ultimately, case reports revealed that patients who underwent surgery had significantly better OS than those who did not receive surgical intervention.

Due to the heterogeneity of the cohort studies, statistical analysis was not possible, but the five studies included in our review suggest that effective surgery significantly reduces both local and distant recurrence rates. As Lans et al. [37] demonstrated, metastasis development is associated with positive margins, tumor aggressiveness, and infiltration depth. Thus, suboptimal surgery negatively affects oncological outcomes [23]. Factors such as tumor size, advanced stage, and histological characteristics also negatively impact outcomes. Certain tumors carry a higher risk of local spread, emphasizing the need for individualized approaches [38].

The cohort review also stresses the importance of specialized centers in treatment. These centers improve the likelihood of achieving optimal surgical margins and reduce the need for secondary surgeries, which can negatively affect functional outcomes [37]. These findings reinforce that radical surgery remains the cornerstone treatment for adult hand sarcomas, playing a pivotal role in shaping treatment strategies and improving oncological outcomes.

Reconstructive surgery

After tumor resection, reconstructive techniques, such as tendon transfers, nerve grafts, and soft tissue coverage with local or free flaps, are essential for restoring hand function [64], 65].

Outcomes

Oncologic and functional outcomes depend on achieving negative surgical margins. R0 margins were obtained in 78.94 % of 76 cases with known margin status. Nicholson et al. reported a 2-year OS rate of 92 % [43]. Functional outcomes, assessed by the MSTS, DASH, EPS, and Michigan Hand Questionnaire, highlight the trade-off between oncologic control and hand function. Limb-sparing surgeries are preferred over amputations for preserving functionality [4], 12], 40].

Prognostic factors

Prognosis is influenced by tumor size, grade, and surgical margins. Tumors >5 cm are linked to higher recurrence and poorer survival [66], 67]. Higher-grade tumors (G2 or G3) correlate with worse outcomes [68]. Negative margins remain key to LC and survival [69]. Age and comorbidities also impact survival, as older patients have reduced physiological reserve [70].

Complications

Common complications include wound infections, stiffness, phantom limb pain, and tendon adhesions. Tumor size and surgical resection depth are major risk factors for postoperative issues [38], 71]. Recovery is often prolonged, and long-term follow-up is necessary for recurrence monitoring. Radiotherapy is associated with complications like periarticular fibrosis, joint stiffness, skin toxicity, and radiation-induced sarcoma, highlighting the importance of optimized protocols and rehabilitation to mitigate long-term disability [53], 72], 73].

In our opinion, neither the use of advanced radiotherapy techniques, such as VMAT and IMRT, which provide improved dose conformity to the target, nor the proper implementation of immobilization systems appears to mitigate these side effects. For instance, as described in a recent report on the treatment of a patient with hand sarcoma [15], the arm was immobilized using a radiotransparent polyethylene support, over which a stereotactic mask was applied, similar to a setup used in intracranial stereotactic treatments, to enhance reproducibility as in Ref. [74]. Despite these precautions, the patient still developed necrosis of the skin graft following radiotherapy. This may suggest that such side effects could be unavoidable, regardless of the technique employed or the adequacy of the immobilization system.

Limitations

This systematic review consolidates evidence from case reports, case series, and cohort studies. However, the rarity of hand sarcomas introduces limitations requiring examination.

First, the infrequency of sarcomas in such a peculiar anatomical location increases the risk of performance and referral filter biases. The surgical management of hand sarcomas requires specialized expertise and spans procedures from limb-sparing resections to amputations. This variability, combined with disparities in expertise between primary and tertiary care centers, may lead to differences in treatment and outcomes.

Radiotherapy is a cornerstone for sarcomas. However, specific guidelines for its use in hand sarcomas are lacking. Clinical decisions are often based on protocols for other sarcomas. This extrapolation may reinforce assumptions about efficacy, introducing confirmation bias, where evidence is interpreted to align with pre-existing beliefs, and spin bias, which emphasizes favorable outcomes while overlooking limitations.

The use of local control as a primary outcome adds complexity. In cases of extensive amputations, the removal of the anatomical site automatically excludes local recurrence, creating a confounding effect that complicates comparisons between ablative and limb-preserving approaches.

Moreover, histological heterogeneity in the studies, encompassing sarcomas with varying aggressiveness, limits the applicability of findings to the broader hand sarcoma population, introducing selection bias. The preferential publication of positive outcomes further exacerbates this issue, leading to publication bias, which could be further amplified by the exclusion of grey literature, as in the present case.

Selective reporting was present in some of the studies included in this review, as certain outcomes, such as quality of life or adverse effects, were not consistently reported across studies. This could lead to a biased representation of the role of radiotherapy in hand sarcomas.

Given the limited number of studies and their qualitative nature, statistical tools for assessing publication bias (e.g., funnel plots, Egger’s and Begg’s tests) and heterogeneity (I² statistic and Cochran’s Q) were not applicable. The small and heterogeneous data set made these methods unreliable and unsuitable for this review.

Regarding funding bias, given the nature of the studies included in this review, it is unlikely that it has had more than a minimal impact on the results.

Finally, the broad temporal span of the studies reflects advancements in sarcoma management over time. Earlier studies favored amputation, while recent approaches emphasize microsurgery, more precise radiotherapy, and novel systemic therapies. This chronological bias limits the applicability of earlier findings to current clinical practice.