Adherence to consensus-based diagnosis and treatment guidelines in adult soft-tissue sarcoma patients: a French prospective population-based study

The present patient-based prospective study evaluates soft-tissue sarcoma care management. Adherence to clinical guidelines was assessed through 23 criteria defined by consensus. Although practices were found to be relatively compliant overall, initial diagnosis and treatment across all stages need improving, particularly outside specialized centers. Standardized incidence rates are also reported.

Abstract

Background

Soft-tissue sarcomas (STSs) are rare tumors with varied histological presentations. Management and treatment are thus complex, but crucial for patient outcomes. We assess adherence to adult STS management guidelines across two French regions (10% of the French population). We also report standardized incidence.

Patients and methods

STS patients diagnosed from 1 November 2006 to 31 December 2007 were identified from pathology reports, medical hospital records, and cancer registries. Guideline adherence was assessed by 23 criteria (validated by Delphi consensus method), and age and sex-standardized incidence rates estimated. Associations between patient, treatment, and institutional factors and adherence with three major composite criteria relating to diagnostic imaging and biopsy as well as multidisciplinary team (MDT) case-review are reported.

Results

Two hundred and seventy-four patients were included (57.7% male, mean age 60.8 years). Practices were relatively compliant overall, with over 70% adherence for 10 criteria. Three criteria with perfect Delphi consensus had low adherence: receiving histological diagnosis before surgery, adequacy of histological diagnosis (adherence around 50% for both), and MDT discussion before surgery (adherence <30%). Treatment outside of specialized centers was associated with lower adherence for all three composite criteria, and specific tumor sites and/or features were associated with lower adherence for diagnostic imaging, methods, and MDT meetings. STS standardized incidence rates were 4.09 (European population) and 3.33 (World) /100 000 inhabitants.

Conclusions

Initial STS diagnosis and treatment across all stages (imaging, biopsy, and MDT meetings) need improving, particularly outside specialized centers. Educational interventions to increase surgeon's sarcoma awareness and knowledge and to raise patients' awareness of the importance of seeking expert care are necessary.

introduction

Adult soft-tissue sarcoma (STS) represents a heterogeneous group of rare malignancies, accounting for around 1%–2% of all adult cancers [1]. Overall survival after STS is around 50%–60% at 5 years, [2] with a 10%–20% risk of local recurrence, and approximately 30%–40% of patients develop metastatic disease. This low incidence and considerable heterogeneity of pathological diagnoses, locations, and prognostic factors, make the clinical management of these highly malignant tumors difficult, especially when the lump is first discovered in primary care. Poor or moderate adherence with clinical practice guidelines (CPGs) from diagnosis to care practices is frequently reported [3–6].

Histological difficulties and multiple coding systems (tumor morphology or anatomic site) have also resulted in variability in the estimation of incidence rates [7]. The estimated standardized incidence rate for STS is generally reported around 1.7 [8]–2.4 [9, 10], but can be up to 3.6 [11] per 100 000 inhabitants. These differences can be explained by the reference population selected for standardization, study setting, or selection of histological types.

Given the documented variation in STS care and the lack of clear incidence data and within the framework of the CONnective Tissues Cancers NETwork project [12], this prospective population-based study used active recruitment and multiple information sources to examine practices in STS care management in terms of adherence with expert-determined guidelines, and to investigate factors associated with lower adherence. A secondary aim was to provide standardized incidence rates.

patients and methods

All adult patients (≥18 years old) diagnosed between 1 November 2006 and 31 December 2007 in the Aquitaine and Midi-Pyrénées administrative districts in South-West France (6 million inhabitants, 10% of the French population) with primary STS of any stage were included. Patients with visceral, bone, uterus or Kaposi's sarcoma, gastrointestinal stromal tumors, or mesotheliomas were not included since their presentation, diagnosis, management, and overall prognosis are quite different than those with STS. Patients being treated for recurrence, and patients diagnosed outside of the administrative districts were not eligible. STS diagnoses were made in public or private pathology laboratories. Data were collected from all relevant sources: pathology reports, medical records from private and public centers, data from the district cancer registries, and weekly multidisciplinary team (MDT) meeting reports from regional cancer networks. French National Commission on Information Technologies and Liberties approved this study.

data collection

Patient management was recorded in terms of: diagnosis, pretherapeutic work-up, assessment of disease extension, surgery, pathology, quality of surgery according to the Union for International Cancer Control (UICC)-R classification, discussion in MDT meetings, and radiotherapy or adjuvant chemotherapy. Tumor depth was defined as ‘superficial’ if it was a supra-aponeurotic tumor of the skin and/or the hypoderm (complete information on data collection is available in supplementary material, available at Annals of Oncology online).

development of quality-of-care criteria: Delphi consensus method

We used the DELPHI consensus method [13, 14] to develop criteria for the quality of care through consultation with selected French national sarcoma group experts (Groupe Sarcome Français—GSF). We selected key elements from available CPG and relevant literature, and drafted a written questionnaire. A multidisciplinary panel of 19 national experts discussed and rated the potential quality-of-care criteria and selected a final set that we report according to the level of consensus (perfect, high, or good) (details in supplementary material, available at Annals of Oncology online).

statistical considerations

Adherence rates were calculated for each criterion as the number of patients receiving the treatment or care as set out by the criterion (‘patients observed’) divided by the number of patients to whom the criterion applied (for example, surgery quality criteria only applied to patients receiving surgery).

Three composite criteria were constructed to describe practices. The first two criteria relate to diagnosis practices: (i) imaging: compliant for patients with deep tumors if they received magnetic resonance imaging (MRI) or scan of the tumor zone before surgery (Delphi 1), and for patients with superficial tumors if they received an MRI, scan, or ultrasound before surgery (for all operated patients) (Delphi 6); (ii) diagnostic method (biopsy): compliant for deep tumors over 5 cm in size or tumors <5 cm increasing in size, if the diagnosis was made from a percutaneous or surgical biopsy. All methods for superficial or deep tumors <5 cm not increasing in size were considered as a compliant unless there was no information on depth or it was a deep tumor with no information on size or possible increase in size (Delphi 2 and 3). (iii) MDT discussion: compliant if patient file discussed in MDT meetings (Delphi 8, 11, 15, 17, 18, and 23), with the added condition that MDT discussion must have occurred before or within 31 days of diagnosis for all tumor types.

Based on the literature, we selected the following factors to test for associations with adherence on these three composite criteria: age at diagnosis [6], tumor site (limbs, head, and neck or trunk wall versus internal trunk) [7], histological grade (1 versus 2/3 versus nongradable) [6], and the type of institution offering the care (specialized versus nonspecialized treatment centers) [5, 6]. Logistic regression was used to model the probability of adherence. We calculated age and sex-standardized incidence rates based on standard populations: French [15], European [16, 17], World [18], US theoretical population of 1 million, and the year 2000 projected US population [19]. SAS 9.2 was used for all analyses. Additional details on the statistical analysis are provided in supplementary material, available at Annals of Oncology online.

results

Overall, 274 adult STS diagnoses were recorded in this study. Patients' characteristics are given in Table 1.

Table 1.

Patient and tumor characteristics for soft-tissue sarcoma patients diagnosed in two French regions between 1 November 2006 and 31 December 2007 (274 patients)

Median age (years; mean, range)	63.8 (60.8, 18–96)
<30 years, n (%)	23 (8.4)
30 < 45 years, n (%)	41 (15.0)
45 < 60 years, n (%)	62 (22.6)
60 < 75 years, n (%)	68 (24.8)
≥75 years, n (%)	80 (29.2)
Sex—male, n (%)	158 (57.7)
Performance Status 0–1, n (%)	185 (67.5)
Tumor site, n (%)
Limbs	114 (41.6)
Trunk walls	52 (19.0)
Retroperitoneum	45 (16.4)
Head and neck	32 (11.7)
Intra-thoracic	11 (4.0)
Intra-abdominal	9 (3.3)
Pelvis	7 (2.5)
Para-testicular	4 (1.5)
Tumor size (cm, median [Q1; Q3]), n (%)	7.0 [3.5; 13.0]
≤5	99 (36.1)
>5 and ≤10	73 (26.7)
>10	89 (32.5)
Missing	13 (4.7)
Tumor location, n (%)
Superficial	89 (32.5)
Deep	178 (65.0)
Missing	7 (2.5)
Histology, n (%)
Liposarcoma	65 (23.7)
Leiomyosarcoma	43 (15.7)
UPSa	39 (14.2)
Dermatofibrosarcoma	31 (11.3)
Synovialsarcoma	13 (4.7)
Angiosarcoma	12 (4.4)
Rhabdomyosarcoma	10 (3.6)
Myxofibrosarcoma	10 (3.6)
Other sarcomasb	29 (10.6)
Sarcoma NOSc	22 (8.0)
Grade, n (%)d
1	32 (19.5)
2	43 (26.2)
3	70 (42.7)
Missing	16 (9.8)
Metastatic or lymph node involvement at diagnosis	35 (12.8)

^aUndifferentiated pleomorphic sarcoma.

^bIncludes four Ewing's sarcoma, three fibrous tumors, two osteosarcoma with soft-tissue sarcoma origins, one desmoplastic small-round-cell tumor, one malignant glomus tumor, one malignant rhabdoïd tumor, three epithelioid haemangioendothelioma, five low-grade fibromyxoid sarcoma, five other sarcomas, and four malignant peripheral nerve sheath tumor.

^cNot otherwise specified.

^dTumors of the following 11 subtypes were not relevant for the grade: angiosarcoma, dermatofibrosarcoma protuberans, epithelioid hemangioendothelioma, round-cell liposarcoma, malignant peripheral nerve sheath tumor, alveolar rhabdomyosarcoma, embryonnal rhabdomyosarcoma, Ewing's sarcoma, low-grade fibromyxoid sarcoma, small-round-cell desmoplastic tumor, and malignant rhabdoid tumor, leaving 161 gradable subtypes (of 274 tumors).

practices observed and adherence with guidelines

Over 70% of patients received guideline-compliant care for four of the eight diagnostic criteria (Table 2): MRI/scan before surgery for deep tumors (Delphi 1: 81.5%), tumor characteristics determined (Delphi 4: 92.7%), thoracic scan received (Delphi 5: 70.8%), and secondary histological review (Delphi 7: 75.2%). Moderate adherence was observed for two criteria: histological diagnosis before surgery for patients with deep, rapidly increasing, or >5 cm in size tumors (Delphi 2: 51.7%), and adequate biopsy technique (Delphi 3: 55.5%). The lowest adherence was observed for two criteria: only 37.1% of patients with superficial tumors received imaging before surgery (Delphi 6), and only 6.4% of patients' files were discussed at MDT meetings before biopsy (Delphi 8).

Table 2.

Adherence of patient management with diagnosis, surgery, radiotherapy, chemotherapy, and pathology (criteria by DELPHI consensus, in order per category)

Delphi criteria by treatment stage		Consensusa	Eligible patients N	Observed patients n	Adherence (%)
Diagnosis
1	Patients with a deep tumor who received an MRIb or a scan of the tumor region before surgery	Perfect	178	145	81.5
2	Patients with a deep and/or increasing in size and/or >5 cm tumor who received histological diagnosis before surgery	Perfect	232	120	51.7
3	Patients receiving an adequate biopsy technique (percutaneous or surgical biopsy)	Perfect	254	141	55.5
4	Patients for whom tumor characteristics were determined (size and depth)	High	274	254	92.7
5	Patients who received a thoracic scan	High	274	194	70.8
6	Patients with a superficial tumor who received imaging evaluation of the tumor region before surgery	High	89	33	37.1
7	Patients for whom the sarcoma diagnosis was seen or reviewed by a reference laboratory	Good	274	206	75.2
8	Patients' records discussed in MDT meeting before biopsy	Good	157	10	6.4
Pathology
9	Patients for whom the pathology report on the surgical specimen includes histological type, grade, and surgical margin	Perfect	229	169	73.8
10	Patients who have a frozen tumor fragment	High	274	87	31.2
Surgery
11	Patients' records discussed at MDT meeting before surgery	Perfect	234	70	29.9
12	Patients unresectable (R2 risk or damaging surgery) for whom record discussed at a MDT meeting	Perfect	58	37	63.8
13	Positivity/negativity of surgical margins reported	Perfect	234	184	78.6
14	Metastatic patients whose records were discussed at MDT meetings	High	35	26	74.3
15	Patients for whom the quality of the surgery was defined at a MDT meeting	High	234	131	56.0
16	Patients whose first surgery was carried out in a specialized center	High	234	93	39.7
17	Patients whose records were discussed at MDT meetings	High	79	53	67.1
Radiotherapy
18	Patients for whom radiotherapy decision was made at a MDT meeting	Perfect	112	97	86.6
19	Patients with deep tumors and head, neck, extremities, or trunk wall locations who received conformation radiotherapy	Good	70	60	85.7
20	Patients with R1 surgery who received postoperative radiotherapy for head, neck, limbs, and trunk wall locations	Good	40	33	82.5
21	Patients with R0 surgery who received postoperative radiotherapy for deep tumors of head, neck, extremities, or trunk wall locations	Good	41	28	68.3
22	Treatment delay between surgery and radiotherapy—median [Q1; Q3]	Good	101	11 weeks [8; 17]
Chemotherapy
23	Patients for whom the chemotherapy decision was made at a MDT meeting	Perfect	74	49	66.2

^aDelphi consensus relates to the percentage of consensus achieved on the importance of criteria after consultation with national committee specialists: ‘good’ (75%–90%), ‘high’ (≥90%–100%), and ‘perfect’ (100%).

MRI: magnetic resonance imaging; MDT: multidisciplinary team.

In other words, irrespective of the relevant population considered for adherence for each criterion, we observed that almost three quarters of patients received imaging of the tumor region before surgery (72.6%) (supplementary Table S1, available at Annals of Oncology online). Less than half of patients overall received thoracic imaging or abdominal imaging before surgery (44.9% and 43.8%, respectively), but at least 88.9% of those with retroperitoneum, pelvis, and intra-thoracic or -abdominal sites received thoracic or abdominal imaging (data not shown). More than four of 10 patients did not receive a biopsy and histological diagnosis before surgery. Most patients' records were discussed in MDT meetings at some stage during their care (80.7%).

Pathology reports were generally compliant with guidelines (Table 2) (Delphi 9: 73.8%), with the least compliant aspect of this criteria being the reporting of surgical margins.

Overall, 85.3% of patients (n = 234) received surgery; 40.9% radiotherapy (n = 112) and 27% chemotherapy (n = 74) (supplementary Table S1, available at Annals of Oncology online). Initial surgery was carried out in a specialized treatment center for 93 (39.7%) of patients. Of the remaining 141 patients who received initial surgery outside of specialized centers, 67 (47.5%) were subsequently referred to a specialized center for continuing care (MDT discussion or treatment), and 74 (52.5%) received further care in a nonspecialized hospital. Overall, 92 patients had an initial R1 resection; among the 43 (47%) patients who required a second operation, 38 (88%) had the first surgery outside of specialized centers, and 5 (12%) in a specialized center. Only 29.9% of eligible patients' cases were discussed in an MDT meeting before surgery (Table 2) (Delphi 11), although adherence was better for unresectable tumors (63.8% discussion in MDT meetings, Delphi 12) and metastatic patients (74.3% discussion in MDT meetings, Delphi 14). Reporting the assessment of surgical margins (positivity or negativity) was largely compliant (78.6%, Delphi 13).

Adherence for radiotherapy ranged from high, with 82.5% of patients with R1 surgery receiving radiotherapy for head, neck, limbs, or trunk wall tumors (Delphi 20), to moderate, with 68.3% of patients with R0 surgery receiving radiotherapy for deep tumors of the head, neck, limbs, or trunk wall (Delphi 21). Other criteria (Delphi 18 and 19) showed >85% adherence for eligible patients.

adherence for major criteria

Of the 261 eligible patients overall (13 did not receive treatment), diagnostic imaging practices (criterion 1) were compliant for 64.4% of patients; biopsy methods (criterion 2) for 67.0%, and MDT discussion within 31 days (criterion 3) for 47.1%.

After adjustment for age and sex, factors associated with less adherence for diagnostic imaging practices (criterion 1) were: tumor site (limbs, head, and neck, or trunk wall) [odds ratio (OR) 3.03, 95% confidence interval (CI) 1.18–7.78], superficial tumors (OR 3.70, 95% CI 1.80–7.61), and treatment in a nonspecialized center (OR 5.69, 95% CI 2.74–11.85) (Table 3). Factors associated with less adherence for biopsy methods (criterion 2) were: tumor site (internal trunk) (OR 4.17, 95% CI 2.21–7.86) and treatment in a nonspecialized center (OR 3.20, 95% CI 1.45–7.05). For MDT discussion within the recommended timeframe (criterion 3), factors associated with lower adherence were: superficial tumors (OR 1.77, 95% CI 1.12–2.80) and treatment in a nonspecialized center (OR 3.23, 95% CI 1.73–6.06).

Table 3.

Multivariate model for factors associated with nonadherence with overall composite criteria for soft-tissue sarcoma patients diagnosed in two French region districts between 1 November 2006 and 31 December 2007 (261 patients)^a

	Diagnostic imaging		Biopsy methods		MDT meeting within 31 days
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
Tumor site
Internal trunk	1		1b		–
Limbs, head, neck, and trunk wall	2.73 (1.20–6.23)	0.0172	0.21 (0.10–0.44)	<0.0001	–
Tumor depth
Deep	1		–		1
Superficial	3.53 (1.67–7.44)	0.0009	–		1.78 (1.07–2.96)	0.0255
Type of institution
Specialized center	1		1	0.0001	1
Others	4.80 (2.05–11.26)	0.0003	3.92 (1.96–7.81)		2.77 (1.48–5.19)	0.0015

^aThirteen patients included in previous analyses who did not receive treatment, but are not included in the multivariate model; the model was adjusted for age and sex.

^bFor descriptive purposes in the text, this OR is presented with the limbs, head, neck, or trunk wall site as the reference: OR 4.76, 95% CI 2.27–10.0.

MDT: multidisciplinary team; OR: odds ratio.

Soft-tissue sarcoma incidence

Of the 274 new cases of adult STS diagnosed in this study, 211 were eligible for the estimation of incidence over the 2007 calendar year (diagnosis made in 2006 for other patients, or they were resident outside of the two defined areas). The standardized incidence rate per 100 000 inhabitants was 3.33 based on the IARC and WPP populations (supplementary Table S2, available at Annals of Oncology online).

discussion

This prospective population-based observational study evaluates STS patient care and management across two French regions and provides robust estimation of standardized STS incidence. We used active recruitment from multiple sources to ensure completeness of recruitment, and in particular, we focused on identifying and explaining areas of low adherence with recommended practices. For this, we developed a set of 23 criteria covering initial care management in primary STS.

Patient characteristics were relatively similar to previous reports [7, 20]. Globally, although practices were relatively compliant (70% and over for 10 criteria), we identified major areas for improvement (under 50% adherence for five criteria). Of the nine undisputable criteria for standards of care (perfect consensus and reported in all recent STS clinical guidelines, e.g. [21]), adherence was particularly low for three: receiving initial histological diagnosis before surgery, adequacy of the biopsy technique (around 50% for both), and discussion in an MDT meeting before surgery (<30%). These areas represent a crucial need for improvement in practices, particularly given that over 4 in 10 patients receive surgery before receiving a histological diagnosis. This figure may be partially represented by a number of inadvertent initial operations, or ‘whoops surgeries’. These unplanned excisions of unsuspected STS lesions originally considered to be benign were estimated to represent up to 40% of all initial operations for STS [22] and have been reported to have negative impacts on patient survival [23], as recently reported for the lack of adherence to locoregional treatment and chemotherapy guidelines [20].

Adherence with general surgical standards was concerning with over one in four patients not receiving any imaging before surgery, showing no improvement since a previous report [24]. Further, over 50% did not receive initial surgery in a specialized treatment center, although receiving treatment at high-volume specialized centers has been clearly identified as improving local control [25] and survival, irrespective of grade [26–28]. This represents an important area for educational interventions aimed at increasing sarcoma awareness and knowledge for surgeons and raising patients' awareness of the importance of seeking specialized care.

This low adherence for imaging and nonspecialized center procedures emphasizes the need for a correct sarcoma diagnosis from the outset, so that patients can be directed early to centers with significant experience dealing with these rare malignancies, especially when coupled with issues involving the lack of, or delay for discussion in MDT meetings.

Our data indicate that MDT meetings are happening, with more than three quarters of patients' records discussed at some stage in an MDT meeting, similar to rates reported in the literature [29].

Finally, the multivariate analysis reveals that patients particularly at risk of low adherence are those receiving treatment outside of a specialized center, those with superficial tumors, and those with tumors of the internal trunk for biopsy methods and those with limbs, head, neck, and trunk wall tumors for diagnostic imaging.

Our incidence results provide a good estimation of adult STS with multisource active recruitment. Using a comparable population of adults over 20 years, the adult-only incidence rate is similar to the rate reported in another French region [7]. Indeed, as recently reported [30], a collaborative method including complete cancer registry and pathology laboratory data is recommended to obtain better incidence estimation for sarcoma.

Some limitations of this observational study should be mentioned. These regions have extremely active networks for sarcoma and rare tumors, which may result in an overestimation of adherence for certain results if extended nationally across other regions with less active networks. An observation bias can also not be ruled out. When interpreting the lack of MDT meetings for patients initially treated in nonspecialized centers, it should be kept in mind that MDT meetings are not physically held at the nonspecialized centers, who refer patients to specialized centers for MDT sarcoma discussion. This association may therefore represent a delay before MDT discussion rather than a lack of discussion. In addition, these practices date from 2007, just after the implementation of sarcoma guidelines. Changes have since taken place in practices with secondary histological review being recommended at the time of study, but mandatory since 2010 at the national level (www.rreps.org), and MDT reports now systematically registered (www.netsarc.org).

conclusion

This report on practices in STS care, using multiple source data, offers a robust description of practices requiring improvement. Future interventions should focus on biopsy methods for internal trunk and superficial tumors, diagnostic imaging for limbs, head, neck, and trunk wall superficial tumors, or timely MDT discussion, in particular for patients treated outside of specialized centers. However, if, and when, diagnosis is correct, patient care appears relatively compliant with CPGs.

funding

Hospital-based Cancer Clinical Research Program 2006, by the French National Cancer Institute (Inca); Infosarcoma 2011 (French Patient Association); CONnective Tissues Cancers NETwork project, by European Union Seventh Framework Programme (FP7/2007-2013, contract FP-018806 the Network of Excellence CONTICANET) and The French League against Cancer Aquitaine Charentes 2007.

disclosure

The authors have declared no conflicts of interest.