Original article

© Schattauer 2009

Clinical outcomes of adjuvant external-beam radiotherapy for differentiated thyroid cancer Results after 874 patient-years of follow-up in the MSDS-trial M. Biermann1,*; M. K. Pixberg2,*; B. Riemann2,*; A. Schuck3,*; A. Heinecke4,*; K. W. Schmid5,*; N. Willich3,*; H. Dralle6,*; O. Schober2,* *for the MSDS study group; 1Department of Radiology, Haukeland University Hospital, Bergen, Norway; 2Department of Nuclear Medicine, Münster University, Germany; 3Department of Radiooncology, Münster University, Germany; 4Department of Biometrics, Münster University, Germany; 5Institute of Pathology and Neuropathology, University Hospital of Essen, West German Cancer Center, University of Duisburg-Essen, Germany; 6Department of General Surgery, University Halle-Wittenberg, Halle, Germany

Keywords Thyroid cancer, external beam radiotherapy, radioiodine treatment, survival

Summary Aim: Evaluate the clinical benefit of external beam radiotherapy (RTx) for locally invasive thyroid carcinoma with follicular cell differentiation (DTC). Patients, methods: The Multicentre Study on Differentiated Thyroid Cancer (MSDS) was planned as a prospective multicenter trial on the benefit of adjuvant RTx in locally invasive DTC (pT4; UICC 1997) with or without lymph node metastases and no known distant metastases. All patients were treated with thyroidectomy, 131I-therapy, and TSH-suppression and were randomized to receive additional RTx or not. In 4/2003 the trial became a prospective cohort study after only 45 of then 311 patients had consented to randomization. 351 of 422 patients met the trial’s inclusion criteria. Age was 48 ± 12 years (mean ± SD). 25% were men. Tumours were papillary in 90% and follicular in 10%. Of 47 patients randomized or allocated to RTx, 26 actually received RTx. Results: Mean followup was 930 days. In an actual treatment analysis, 96% (25/26) of the RTx-patients

Correspondence to: Martin Biermann, MD, Associate Professor Nuclear Medicine/PET-Centre, Department of Radiology, Haukeland University Hospital and Section for Radiology, Department of Surgical Sciences, University of Bergen, 5021 Bergen, Norway Tel. +47/55 97/76 43, Fax +47/55 97/76 02 E-mail: [email protected]

reached complete remission (CR) vs. 86% in the non-RTx patients. Recurrences occurred in 0 vs. 3 % of patients: 6 reoperated for regional lymph node metastases, 1 tracheal invasion treated with tracheoplasty, 1 local invasion necessitating laryngectomy, 2 distant metastases (1 lung, 1 lung + bone). Serious chronic RTx toxicity occurred in 1/26 patients. Conclusion: The MSDS trial showed low mortality and recurrence rates and a weak benefit of RTx in terms of local control that did however not reach statistical significance. Routine RTx in locally invasive DTC can no longer be recommended.

Schlüsselwörter Schilddrüsenkarzinom, perkutane Strahlentherapie, Radioiodablation, Überleben

Zusammenfassung Ziel: Nachweis des klinischen Nutzens der adjuvanten perkutanen Strahlentherapie für lokal organüberschreitendes Schilddrüsenkarzinom mit Follikelzelldifferenzierung (DTC). Patienten, Methoden: Die Multizentrische Studie Differenziertes Schilddrüsenkarzinom (MSDS) war ge-

Klinische Resultate der adjuvanten perkutanen Strahlentherapie bei differenziertem Schilddrüsenkarzinom – Ergebnisse der MSDS-Studie nach 874 Patientjahren Nachbeobachtungszeit Nuklearmedizin 2009; 48: 89–98 doi: 10.3413/nukmed-0221 received: October 25, 2008 accepted in revised form: November 26, 2008 prepublished online: March 23, 2009

plant als prospektive randomisierte Multizenterstudie zum Nutzen der adjuvanten perkutanen Radiatio bei lokal überschreitendem DTC (pT4; UICC 1997) mit oder ohne Lymphknotenmetastasen und ohne Fernmetastasen. Die Behandlung umfasste Thyreoidektomie, Radioiodablation und TSH-Suppressionsterapie sowie randomisiert die zusätzliche perkutane Radiatio. 4/2003 wurde die Studie als prospektive Kohortenstudie weitergeführt, nachdem lediglich 45 von 311 Patienten in die Randomisierung eingewilligt hatten. 351 von 422 erfüllten die Einschlußkriterien. Das mittlere Alter betrug 48 ± 12 Jahre (± Standardabweichung) bei einem Anteil männlicher Patienten von 25%. 90% der Tumoren waren papillär, 10% follikulär. Von 47 Patienten, die der Radiatio zugewiesen worden waren, wurden 26 bestrahlt. Ergebnisse: Die mittlere Nachsorgeperiode ist 930 Tage. 96% (25/26) der bestrahlten Patienten erreichen eine komplette Remission gegenüber 86% der nicht bestrahlten. Rezidive ereigneten sich bei keinem der bestrahlten und 3% der nicht-bestrahlten Patienten: 6 Rezidivoperationen wegen regionaler Lymphknotenmetastasen, 1 Trachealinvasion (behandelt mit Tracheoplastie) und 1 Lokalinvasion (Laryngektomie) und 2 Patienten mit Fernmetastasen (1 Lunge, 1 Lunge und Skelett). Bei 1/26 Patienten führte die Radiatio zu schwerer chronischer Toxizität. Schlussfolgerung: Die MSDS-Studie ergab niedrige Letalität und Rezidivraten. Eine geringgradige Verbesserung der lokalen Kontrolle durch die Radiatio war statistisch nicht signifikant. Die routinemäßige Radiatio kann bei lokal invasivem DTC nicht länger empfohlen werden. Nuklearmedizin 3/2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

89

90

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

The benefit of external beam radiotherapy therapy (RTx) in addition to surgery, radioiodine (131I) therapy (RIT) and thyroidstimulating hormone (TSH) suppressive therapy for differentiated thyroid cancer (DTC) has long been a subject of debate (1). As early as 1982 a retrospective cohort study comparing two patient cohorts treated at the two university hospitals in Munich in Germany concluded that a prospective randomized trial was needed (19). Based on the results of two German series that showed an impressive benefit of RTx in addition to surgery and RIT (12, 19) multimodal therapy including RTx was considered as standard in patients with locally invasive thyroid cancer (13) and routinely applied well into the mid-1990s (14). The interdisciplinary guidelines of the German cancer society from 1996

marked a reversal of this trend (16) as more evidence accumulated that multimodal therapy without routine use of RTx could yield excellent results (20, 24). The following is the first report on outcomes of the Multicentre Study on Differentiated Thyroid Cancer (Multizentrische Studie Differenziertes Schilddrüsenkarzinom; MSDS) (3).

Patients, material, methods MSDS Trial MSDS was planned as a prospective randomized trial on the clinical benefit of adjuvant external beam radiotherapy (RTx) (3) for locally invasive differentiated thyroid carcinoma

(DTC) free of distant metastases corresponding to TNM stages pT4 pN0/1/x M0/x R0/1 (34) and pT3/4 pN0/1/x M0/x R0/1 (33) respectively, in Germany, Austria, and Switzerland.

Design The methodology of the trial has been described in detail (3, 4). MSDS was designed as a comprehensive cohort study (26) with randomization and observation arms (씰Fig. 1). Patients were enrolled at the time of the first ablative radioiodine therapy (RIT). Inclusion criteria were papillary or follicular carcinomas (in this study collectively named differentiated thyroid carcinomas, DTC) with local extension beyond the thyroid capsule – UICC pT4 (34)/pT3b (32, 33) – irrespective of nodal status (pN0/1/x (33, 34)) and free of distant metastases (M0/x (33, 34)), age between 18 (incl.) and 70 years (excl.) at the time of initial surgery, completion of primary surgical therapy with R0 (no tumour residues) or R1 (microscopic residues) resection, Karnofsky index ≥ 70 %, freedom from distant metastases at the time of initial RIT, and informed patient consent. Cases were seen by reference pathologists and reclassified if necessary. Patients who agreed to randomization were randomized into treatment arms A (RTx) or B (no RTx) at the time of the first 131 I-scintigraphy 3–4 months after initial RIT. In centres not participating in randomization, patients were assigned to either treatment arm by the referring physicians. The power calculation underlying the trial was based on the assumption of a cumulative recurrence rate of 19% in the non-irradiated arm over a 5-year recruitment and a 3-year follow-up period (1, 12). It demanded a sample size of 2 × 250 randomized patients in order to detect a 50% reduction in recurrence rate by RTx with 80% power (Fisher’s exact test, p < 0.05, two-tailed) (1). The study protocol was approved by the ethics review board at Münster University on 16 June 1997. The trial was conducted according to the Declaration of Helsinki.

Treatment protocol Fig. 1 MSDS study protocol

The MSDS treatment protocol was in accordance with the current treatment guidelines (2, 16, 28). It included

Nuklearmedizin 3/2009

© Schattauer 2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

●

●

●

total thyroidectomy (TT) with central lymphadenectomy (6, 7), RIT to ablate the thyroid remnant (7, 8), and TSH-suppressive therapy with Lthyroxine (TSH < 0.1 mU/l) (3).

RIT was to be administered under endogenous TSH-stimulation after 4 weeks’ withdrawal of L-thyroxine using standard activities of 3–4 GBq with a post-therapeutic whole-body scintigraphy (WBS) (9). If scintigraphic 131I uptake by the thyroid remnant was visible in the WBS (≥ 200 MBq; ≥ 48 h) 3–4 months after RIT under TSH-stimulation, a second fraction of RIT with 4–10 GBq was to be given immediately after the WBS (7, 8). At the time of each WBS, TSH, serum thyroglobulin (Tg), Tg-recovery (TgR) and anti-Tg-antibodies (TgAB) and blood cell count were required as well as an ultrasound (US) examination of the neck. For patients in treatment arm A, RTx was begun after documented complete elimination of cervical 131I uptake in a diagnostic WBS, if necessary after repeat RIT (1, 3, 19, 27) in accordance with the current German radiotherapy guidelines (28). RT included the thyroid bed with a dose of 59.4 Gy and 66.6 Gy after R0 and R1 resection, resp., and the regional lymph nodes of the neck and upper mediastinum including the posterior cervical chain from the mandible and mastoid process to the tracheal bifurcation with a dose of 50.4 Gy and 54.0 Gy in pN0 and pN1/x disease, resp. Fractionation was 1.8 Gy/d 5 days a week. 3-D or quasi 3-D planning according to IRCU 50 was mandatory (27). Treatment plans were reviewed by the trial’s reference radiotherapist before start of therapy. Acute and chronic toxicity was monitored with yearly follow-up visits within the first 5 years following RTx (27).

Fig. 2 Patient recruitment in the MSDS trial: The map shows the participating centres in Germany, Austria, and Switzerland. The bubble size corresponds to the number of analyzable patients. The inset shows number of reported patients (n = 422) to the trial over time. The randomization arm was terminated on 1 April 2003.

Definitions A patient was defined to be in complete remission (CR) when there was no 131I uptake in a WBS, serum thyroglobulin (Tg) under TSH-stimulation – endogenous after thyroid hormone withdrawal or after stimulation with recombinant TSH (rTSH) – was eliminated below the institutional threshold (usually 1–2 ng/ml), and other imaging studies (regular cervical ultrasound and an initial computed tomography (CT) of the thorax as part of the protocol, other modalities on clinical suspicion) were negative. A recurrence was defined as the occurrence of new disease after a patient had reached CR or as the occurrence of disease in a region that had previously undergone localized therapy (e. g. a lymph node metastasis in a cervical compartment that had been subject to lymph node dissection before or a local recurrence after thyroidectomy).

Follow-up Data management 131

I WBS, cervical ultrasound (US) and serum-Tg under endogenous or exogenous (recombinant TSH, rTSH) TSH-stimulation were performed three months and one year after the last ablative 131I therapy and thereafter at 2-year intervals. Out-patient followup visits (TSH, Tg, US) were scheduled at half-year intervals (3).

Case report forms (CRF) were sent to the trial’s data centre after each follow-up visit. Data were then entered into the trial’s central client server database under Oracle 8i/9i (Oracle Corp., Redwood Shores/CA) and SuSE Linux (3, 4). After each entry of documentation, a status fax was automatically gener-

ated and faxed back to the participating centre by the database. The fax confirmed all incoming documentation and listed the patient’s pertinent data. These included each 131 I activity, the date of its application and the uptake, Tg, TgR, TgAB, TSH and blood count as well as a list of documentation errors requiring correction or clarification (4). In compliance with Good Clinical Practice (GCP), the database maintained a complete audit trail for each entry of information (4). Histology reports and surgical notes were sent to the trial’s reference surgeon for review. Histological specimens were sent to the trial's reference pathologist for confirmation of the histological diagnosis.

Biometrics and data analysis For the present analysis, primary observations stored in several SQL tables in a relational data model were transformed and aggregated into a single star view, which was then queried by a single SQL*Plus script under Oracle 9i. Query results, and in particular outliers, were then validated against the primary documentation. The data of the star view was imported into SAS 9.1 for Windows (SAS Institute, Inc.) for statistical analysis. Student’s t-test and Fisher’s exact test were used for hypothesis testing (p < 0.05, two-tailed) and calculation of confidence intervals for continuous and categorical vari-

© Schattauer 2009

Nuklearmedizin 3/2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

91

92

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

Fig. 3 Patient flow, follow-up (FU) and outcomes in the MSDS trial

ables, respectively. Survival analysis was performed by computing product-limit estimates (log rank test, p < 0.05). Plots were generated with Sigmaplot 10.0 for Windows (Systat Software, Inc.).

Results Recruitment From January 1, 2000 until 31 December 2004, n = 422 patients from 45 centres in Germany, Austria, and Switzerland were assigned to the trial (씰Fig. 2). Of these, 17 patients lacked documentation (씰Fig. 3). 8 patients withdrew their consent, including 1 patient who had already undergone randomization. Of the remaining 397 patients, 46 did not meet the trial’s inclusion criteria because of the reasons listed in 씰Table 1 (“excluded” patients). The present report focuses on the remaining 351 patients who met the trial’s inclusion criteria (“included” patients).

Treatment assignment The trial was randomized until 31 March 2003. At that time, only 35 out of 279 patients had undergone randomization. Advised by the trial’s independent Data Monitoring and Safety Committee (DSMC), the MSDS steering committee decided continue MSDS as a purely observational prospective cohort study (3). Of the 36 patients, 17 were randomized to receive RTx (Fig. 3). Of the 17, only 6 patients actually received RTx. Of the 18 patients randomized to no RTx, none received RTx. Of the 316 patients in the observational arm, 20 received RTx by the assignment of the treating centre (Fig. 3).

Patient characteristics Mean age of the 351 included patients at the time of the initial operation was 48 ± 12 years [20–69] (mean ± standard deviation; range) (씰Tab. 2). All thyroid carcinomas were histologically documented to be locally invasive corresponding to TNM categories pT4 (34)/pT3b (32). 4 out of 80 cases that were classified as pT3 according to the revised

TNM classification (33) after 1 Jan 2003 showed no local invasion – pT3a according to the later TNM supplement (32) – and were excluded (Tab. 1). 90% had papillary cancer, and the remaining follicular cancer. Maximum tumour diameter was 2.0 ± 1.6 cm [0.2–10]. 24 % of tumours represented PTC with a maximum diameter ≤ 1 cm (Tab. 2). Only 26 of the 351 patients received RTx after WBS, 6 through randomization and 20 through allocation by their physicians (Fig. 3). The characteristics of the various subgroups in the trial are summarized in Table 2. With the exception of a higher proportion of patients with microscopically incomplete resections (R1) in arms A/RTx+ there were no major differences between arms A and B and irradiated and non-irradiated patients even though some comparisons reached statistical significance (Tab. 2). There were no significant differences between the two groups of the 35 randomized patients (not shown).

Primary therapy Four patients had had previous subtotal thyroid resections with a benign histology be-

Nuklearmedizin 3/2009

© Schattauer 2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

tween 15 and 21 years before their first thyroid cancer operation. 35% of patients underwent one operation before reaching CR, 59% two operations, and 7% three or more operations (Tab. 2). Of all included patients, 27% were examined with preoperative fine-needle biopsy (FNB), 40% with intraoperative frozen section and 49% with either and 18% with both methods. FNB was reported to show malignancy in 40%, potential malignancy in 37% and was false-negative in 18% of cases, frozen section showed malignancy in 66%, potential malignancy in 9% and was false-negative in 25%. Transient and/or permanent recurrent laryngeal nerve palsy (RLNP) was reported in 19% of included patients. Patients received a mean cumulative 131I activity of 6.5 ± 5.0 GBq in order to reach CR after a median of 187 days (Tab. 2).

Outcomes Under the above treatment regimen, 306 of the 351 included patients reached CR: 25 of the 26 irradiated patients (96%) and 281 of the 325 non-irradiated patients (86%) (Tab. 2). The last reported remission status for the 1 patient in the RTx+ arm is partial remission (PR) (4%) and for the patients in the RTx-

Tab. 1 Reasons for exclusion from the trial (“excluded” patients, n = 41) reason for exclusion

n pts. PTC ≤ 1 cm RTx recurrence

poorly differentiated thyroid carcinoma (incl. 1 patient > 70 yrs.)

12

5

secondary cancer (incl. 1 pt. with previous RTx)

7

patient age at the time of initial operation

5

1

distant metastases at the time of initial RIT

5

2

no extrathyroidal invasion (pT3a UICC 2002 (32))

4

no extrathyroidal invasion (pre UICC 2002)

3

serious medical condition interfering with treatment

3

extrathyroidal primary

2

deviation from the study protocol predating enrolment

2

1

R2-resection with macroscopic tumour residues

1

1

suspicion of distant metastases

1

no thyroid cancer

1

total

1

2

1

46

3

2

9

N pts.: number of patients (multiple reasons per patient), PTC ≤ 1 cm: number of patients with papillary thyroid cancer ≤ 1 cm, RTx: actual external beam radiotherapies, recurrrence: number of recurrences (see Table 4 for details). 1 RTx was performed immediately after the first radioiodine therapy and thus constituted a protocol violation in a patient who otherwise would have been eligible.

arm PR in 25 (8%), stable disease in 6 (2%), progressive disease in 8 (2%), death (suicide) in 1 (0.3%) and missing in 8 (2%). Recurrences and deaths according to treatment arms are listed in 씰Table 3. There

were no significant differences in recurrencefree survival between Arms A and B (intention-to-treat analysis; 씰Fig. 4A). Because of the high numbers of crossovers between Arms A and B (Fig. 3), an actual treatment

Tab. 2 Patient characteristics: The tables list the pertinent characteristics of the various subgroups in the trial: patients meeting and not meeting the trial’s inclusion criteria (included and excluded), and, among the included patients, treatment arms A and B (intention-to-treat analysis), patients who did, and who did not, receive external beam radiotherapy (RTx +/-; actual treatment analysis), and patients who did, and who did not, suffer a recurrent event (recurr. +/-). group

n

age (years)

male (%)

PTC (%)

≤ 1cm TD (%) (cm)

included

351

48±12

25*

90*

24*

excluded 046

52±16

42*

57*

A (RTx+) 046

48±10*

26

B (RTx-)

305

46±12*

RTx +

026

RTx -

pNx (%)

R+ (%)

>2 OP 131I (%) (GBq)

CR (%)

time to CR (days)

rec. (%)

2.0±1.6* 39

30

16

07

6.5±5.0

87*

187 (141–335)

003* (2–6)

07*

3.6±2.9* 33

22

26

10

8.3±9.2

50*

152 (129–418)

020* (9–34)

91

15

2.3±2.5

43

35

30*

07

5.5±5.2

89

165 (146–293)

004 (0.5–15)

25

90

25

1.9±1.5

38

29

14*

07

6.7±5.5

87

187 (141–342)

003 (1–6)

47±10

23

88

15

1.9±0.9

50

27

42*

08

5.1±2.5* 96

165 (147–279)

000 (0–13)

325

46±12

25

90

24

2.0±1.5

38

30

14*

06

6.7±5.1* 86

187 (140–342)

003 (2–6)

recurr. +

011

41±14

45

82

09

2.7±1.6

73*

18

16

18

7.6±3.9

88

414 (132–619)

(100)

recurr. -

340

48±12

24

91

24

2.0±1.5

38*

30

18

06

6.5±5.0

73

184 (142–322)

00(0)

pN1 (%)

age: patient age at the time of the first operation (mean ± standard deviation), male: man, PTC: papillary thyroid cancer, ≤ 1 cm: PTC ≤ 1 cm, TD: maximum tumour diameter, pN1/x: lymph node metastases in a pathological specimen, R+: incomplete resection (R1/2), > 2 OP: proportion of patients receiving more than 2 operations before reaching complete remission (CR) or the first recurrent event (whatever comes first), 131I: cumulative 131I-activity given until time of CR or the first recurrent event, CR: proportion of patients reaching complete remission, time to CR: days until reaching CR (median, interquartile range), Rec.: proportion of patients with documented recurrence or death with 95% confidence interval in brackets. An asterisk (*) denotes a statistically significant difference within each comparison pair (t-test for continuous variables, Fisher’s exact test for categorical variables; Savage one way analysis for time to CR; p < 0.05 two-sided).

© Schattauer 2009

Nuklearmedizin 3/2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

93

94

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

Tab. 3 Recurrent and fatal events in patients meeting the trial’s inclusion criteria: The tables list the individual patients in which death or recurrence is reported.

case TD (cm) pN

R

Rnd.

arm

RTx

CR

days

diagnosis

therapy

outcome

suicide (pt. in CR)

-

died

1

3.0

0

0

+

B

-

+

1032

2

6.8

1

1

-

A

-

-

370

recurrence larynx, upper mediastinum

laryngectomy, RTx

renewed recurrence

3

1.1

x

0

-

A

-

+

395

LNM K1a (2/10)

OP

CR

4

3.8

1

0

-

B

-

-

248

LNM K1a + K2

OP

PR

5

2.6

1

1

-

B

-

-

281

LNM, invasion v. subclavia

OP, vascular interponate

CR

6

1.7

1

0

-

B

-

+

290

LNM K3, met. sternum

OP, RTx

CR

RIT

CR

131I-uptake

7

3.5

1

1

-

B

-

+

404

pulmonary

8

2.0

1

0

-

B

-

+

536

LNM K1b (1/4) K2 (2/19)

OP

CR

9

1.0

1

1

-

B

-

+

842

LNM K1b (2/11)

OP

CR

10

2.5

x

0

-

B

-

+

1013

pulmonary met. (CXR)

?

PrD

11

1.8 (m)

1

0

-

B

-

+

1070

local paratracheal recurrence

tracheoplasty with muscle flap

CR

case: case number, TD (cm): maximal tumour diameter in cm ((m) = multifocal), pN: lymphonodal staging (33, 34), R: radicality of resection (33, 34), Rnd.: randomization arm (+/-), arm: treatment arm (A = intended RTx, B = no RTx), RTx: actual external beam radiotherapy (+/-), CR: reached complete remission before recurrence (+/-), days: days between 1st operation and recurrence, diagnosis, therapy, outcome: last known outcome. PrD = progressive disease; remaining abbreviations: see Table 1

analysis was carried out (Fig. 4B). Again, there were no statistically significant differences between the RTx+ and RTx– arms. The estimate for 5-year recurrence-free survival in the RTx– arm is 94% (95% confidence-interval 89–97) and the cumulative event rate 3% (2–6), while no events occurred in the RTx+ arm. The baseline characteristics of patients with and without recurrences are listed in Table 2. 8/11 (73%) patients with recurrences were nodal positive versus 37% of the remaining patients (p < 0.05). If tumours ≤ 1 cm are excluded from the analysis, the 5-year recurrent-free survival in the RTx – arm is 93% (86–96) with a cumulative event rate of 4% (2–7%).

Adverse events Acute toxicity under RTx was reversible in all patients (27). One case of serious (RTOG IV*) chronic radiotoxicity was reported in a patient requiring tracheostomy for chronic laryngeal oedema, which results in a 4% (95%-confidence interval 0.1–20%) cumu*

Results have been presented at the Annual Meeting of the European Association in Copenhagen on 16 October 2007.

lative rate for serious (RTOG III-IV*) chronic toxicity. In this patient, a Reinke oedema of the larynx had been present before RTx.

Excluded patients The 46 patients not meeting the trial’s inclusion criteria generally had poorer prognostic factors with higher age, a higher proportion of male patients and a larger tumour diameter (Tab. 2). Most importantly, it included 2 patients with distant metastases at the time of initial staging and 12 patients with poorly differentiated carcinoma. The latter group was responsible for 5 of the 9 clinical events in the group of the excluded patients, including 2 tracheal invasions requiring tracheotomy (씰Tab. 4). Compared with the patients meeting the study’s inclusion criteria the group of patients excluded from the trial had an adverse prognosis with an estimated event-free 5-year survival rate of 55% (24–78; p < 0.0001) (Fig. 4D) and a cumulative event rate of 20% (9–34).

Discussion The MSDS trial represents the largest multiinstitutional cohort of high-risk (20) DTC patients worldwide second to the North American National Thyroid Cancer Treatment Cooperative Study (29). In the years of 2001 and 2002 the trial recruited approx. 50% of the eligible incident population (3, 14). The rationale underlying adjuvant RTx in addition to RIT is that locoregional microscopic tumour residues and micrometastases to the regional lymph nodes are frequent especially in locally invasive carcinoma (21), that they cause clinically relevant recurrences (12) and distant metastases (19) if left untreated, and that they can be sterilized by RTx but not by RIT (1). Lesions may be too small so that they are not hit by the beta-radiation taken up in the tumour tissue in the course of RIT or may be of so low differentiation that they do not have 131I uptake (1). This “therapeutic gap” may be closed by RTx (25). The main result of the MSDS trial is that the recurrence rate in patients with locally invasive DTC receiving multimodal treatment including surgery, RIT and TSH-suppressive therapy but excluding RTx is, after a mean follow-up of nearly three years, significantly

Nuklearmedizin 3/2009

© Schattauer 2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

Fig. 4 Event-free survival in the MSDS-trial A) Intention-to-treat-analysis (Arm A – RTx+ v. Arm B – RTx-) of patients meeting the trial’s inclusion criteria (“included” pts.). Product-limit estimate (KaplanMeier) calculated with SAS 9.1. Vertical bars denote censored patients. B) Actual treatment analysis (RTx +/-) of the same cohort. C) Actual treatment analysis (RTx +/-) of the patients not meeting the trials inclusion criteria (“excluded” pts.). None of the comparisons in panels A-C reached statistical significance (n. s., log rank test). D) Comparison of included and excluded patients (p < 0.0001; log rank test).

lower than expected on the basis of retrospective series from the 1990s (12). The assumption underlying the trial’s power calculation was a cumulative recurrence rate of 19% after a 5-year recruitment and a 3-year follow-up period (1). The recurrence rate observed in the trial over this period is 3% (2–6%) with an estimated 5-year recurrence-free survival of 95% (90–97%). This rate corresponds to recently published single-institution data of patients with locally invasive thyroid cancer

patients (10, 15). Even though the data suggest a slight benefit of RTx – none of the 26 RTx-patients included in the trial had a recurrence even though a high proportion of patients (42%) had microscopically incomplete resection – this benefit is counterbalanced by severe (RTGO ºIV) chronic toxicity in 1 irradiated patient (4%; 95% CI 0.1–20%). Chronic toxicity observed in the MSDS-trial corresponds to a similar rate of events in a large prospective trial for RTx of head and

neck cancer using a comparable dose regimen (18). Toxicity is thus in a similar order of magnitude as the treatment benefit observed in the MSDS cohort. Based on these data from the MSDS trial, routine RTx in addition to surgery, RIT and TSH-suppressive therapy in patients with locally invasive DTC cannot be recommended. This is in line with our recent metaanalysis (1). Of the 7 reports showing a benefit of RTx in terms of local control (5, 11, 12, 17, 30, 31)

© Schattauer 2009

Nuklearmedizin 3/2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

95

96

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

Tab. 4 Known recurrences in patients not meeting the trial’s inclusion criteria: One death among the 17 patients lacking documentation (Fig. 3) is included as case no. 10. case TD (cm) pN1 1

4.0

2

0

R

PDTC Rnd.

Arm

RTx

1

+

A

-

-

CR

days 78

0

1

+

-

A

-

1265

1

1

-

-

B

-

60

4

1

0

+

-

B

-

5

1

0

-

-

B

-

1

0

-

-

B

-

3

6

1.8

2.5 (m)

+

diagnosis

therapy

local recurrence

tracheostomy 131I-neg.),

pulmonary met. ( bone met. (1st rib)

progression of locoregional metastases

OP

PR

106

local recurrence

tracheostomy

died after 134 d due to vascular invasion

158

pulmonary metastases

180

LNM K2 (1/10) + K3 (3/11) 131I-uptake

7

5.5

1

0

-

-

B

-

+

452

new roid bed

8

3.0

1

0

+

-

B

-

+

769

LNM K1a (5/6), K2 (5/13)

0

1

+

-

B

-

+

1515

9

outcome

10

in the thy-

died after 158 d OP RIT OP

lacking documentation died

not enrolled in trial

abbreviations: see legend of Table 2; PDTC: poorly differentiated thyroid carcinoma

or survival (5, 11, 19, 31, 35), none is representative of current medical practice at specialized institutions in the industrialized world.

Limitations The trial has a number of limitations. The first and most important limitation is the lack of a sufficient number of randomized patients (3) and the high number of crossovers to the non-irradiated arms of the trial. This is apparently due the diminished acceptance of RTx among physicians and patients. While 113 of 258 patients (44%) with locally invasive DTC received RTx in a nationwide survey conducted in 1996 (14), only 26 of 351 included patients (7%) underwent RTx in the MSDS trial. Too few patients gave initial consent to randomization, and of the 17 patients randomized to receive RTx, 11 crossed over to the non-RTx arm. Thus the original aim of the trial, to establish internal validity of the trial results on the base of a randomized intervention, has not been reached. The second limitation concerns the external validity of the MSDS data. Patient documentation in the MSDS was very resource-intensive and required dedication by the participating institutions. This may have lead to a selection of centres of excellence. Our clinical

experience from one of Germany’s largest nuclear medicine referral centres with regular secondary and tertiary consultations suggest that results the MSDS trial may not be generalized to all institutions, particularly those treating a low volume of patients. The comparison of the patients included in the trial with the excluded patients (Fig. 4D) implies that the trial’s inclusion criteria – DTC of any size with any degree of extrathyroidal extension based on the 1997 UICC classification (34) – may have contributed to the selection of a collective with a more favourable prognosis. Single-institution series from the 1990s had demonstrated local invasion to be a major independent risk factor (12, 20). Recent retrospective data from Japan suggest that only massive extrathyroidal extension to the subcutis or cervical viscera including the recurrent laryngeal nerve, but not minimal extension to the perithyroidal soft tissue (pT3b in the 2002 UICC classification (32) is an independent risk factor for local or distant recurrence (15). On the other hand, poorly differentiated cancer carried a poor prognosis with 5 out of 12 recurrences (42%; 15–72%) over the trial’s follow-up period. In this group of patients RTx might be beneficial (1) even though the MSDS trial provides no conclusive data on this.

A third limitation of the trial may lie in the high proportion (24%) of small cancers ≤ 1 cm, all of papillary differentiation, in patients meeting the inclusion criteria. In this subgroup, only 1 out of 83 patients (1%; 0.03–7%) had a recurrence (Tab. 3). This is in line with recent retrospective data from Münster University Hospital (22, 23). However, all comparisons in Figure 4 remain valid if patients with a tumour diameter ≤ 1 cm are excluded from the analysis (not shown), so that the overall conclusions are unaffected. A fourth limitation of the MSDS data presented is the – until now – limited followup of the trial cohort of a mean 930 days. However, all the major trials showing a benefit of RTx in locally invasive (12, 19) or incompletely resected (31) DTC showed clear trends in the outcome data already after a comparable duration of follow-up. This, and the fact that it is possible to demonstrate a statistically significant difference in recurrence-free survival between arms in the MSDS trial, give us the justification to perform this interim analysis in accordance with the plans outlined in the study protocol. A fifth limitation of the trial may be underreporting of events. This is a concern with all multicenter trials. We have tried to address this issue by developing a sophisticated system for quality control, which in-

Nuklearmedizin 3/2009

© Schattauer 2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

cluded regular status faxes to the participating centres (3), and by presenting interim reports at a succession of national meetings. Some unevenness of quality of follow-up and documentation between centres is unavoidable, but will only affect results if there are differential errors between treatment arms. A sixth limitation of the trial has been limited adherence to the then current treatment guidelines and the trial’s treatment protocol (2). This not only applies to nuclear medicine but is exemplified by the unacceptably high number of repeat operations needed to achieve CR due to suboptimal preoperative staging. However, if relevant, suboptimally administered surgical and radionuclide treatment should increase the potential benefit of RTx. Therefore, this effect does not affect our conclusions.

Conclusion Mortality and recurrence rate in the MSDS trial have been low. Data show a weak benefit of RTx in terms of local control and chronic RTx toxicity in a similar order of magnitude even though none of the observed effects reached statistical significance. Based on the MSDS data, routine RTx in patients with locally invasive DTC can no longer be recommended. Further follow-up of the MSDScohort is planned.

Acknowledgement The MSDS trial was funded by the Deutsche Krebshilfe e. V. (Project No. 70–2294) and has been registered at http://clinicaltrials.gov (ID NCT00144079). MSDS study group Study chairman: Prof. Dr. Dr. O. Schober.* Vice chairman: Prof. Dr. H. Dralle (HalleWittenberg).* Study coordinators: Prof. Dr. M. Biermann,* Prof. Dr. B. Riemann.* Reference centers: Biometrics: Dr. A. Heinecke,* Prof. Dr. W. Köpcke.* Pathology: Prof. Dr. K. W. Schmid* (Essen). Surgery: Prof. Dr. H. Dralle (Halle)*. Radiotherapy: Prof. Dr. A. Schuck,* Prof. Dr. N. Willich.* Clinical monitoring: Prof. Dr. Chr. Reiners,* Dr. M. Geling (Würzburg). Independent Data Monitoring and Safety Committee: Prof. Dr. H. W. Hense (Münster; chairman), Prof. Dr. H. Jürgens,* Prof. Dr. L. Kiesel, PD Dr. P.-A. Löschmann

(Münster). Members of the MSDS steering committee are indicated by an asterisk (*). MSDS study participants (nuclear medicine; number of patients contributed to the current analysis in parentheses): B. Riemann, Michaela K. Pixberg, M. Weckesser, O. Schober, Münster (55); M. Dietlein, C. Kobe, H. Schicha, Köln (44); H. Schlemmer, H. Lerch, Wuppertal (31); U. Dörr, H. Bihl, Stuttgart (30); Jana Grimm, Traute Mende, Halle (23); Ursula Nestle, D. Hellwig, C. Alexander, C. Kirsch, Homburg/Saar (19); Birgit Schryen, S. Ladner, C. Reiners, Würzburg (18); Jessika Schmitt-Makula, T. Zajic, E. Moser, Freiburg (15); A. Szikszai, C. Eilles, Regensburg (14); Waltraud Eichhorn, P. Bartenstein, Mainz (13); Christina Greb, Nicole Bena-Boupda, Dagmar Steiner, R. Bauer, Gießen (10); Peggy Fostitsch, R. Steinke, Magdeburg (8); Sandra Rosenbaum, R. Görges, A. Bockisch, Essen (8); F. Gottschalk, Frankfurt/O. (8); Cornelia Puskás, J. Peter, F. Maul, Karlsruhe (8); G. Meisetschläger, K. Scheidhauer, M. Schwaiger, München (8); J. Müller, Eva Weckesser, W. Knapp, Hannover (8); A. Becherer, R. Dudczak, Wien (7); J. Borkopp, Gerlinde Thiemann, B. Theophil, Dortmund (7); Birte Diekmeyer, Elisabeth Ostwald-Lenz, H. Wieler, Koblenz (7); A. Schneider, B. Hackmann, B. Piotrowski, Hamm (6); Gerhild Kumnig, P. Lind, Klagenfurt (6); P. Groth, C. Schümichen, Rostock (5); A. Szikszai, A. Schwarz, E. Fiedler, T. Kuwert, Erlangen (4); Gabriele Holl, R. Schöneich, Potsdam (4); H. Steinert, G. von Schulthess, Zürich (4); C. Glosemeyer, Katrin Weigel, J. Kropp, Ch. Dschietzig, Cottbus (4); P. Panholzer, Sigrid Pöcher, Elfriede Rechberger, W. Langsteger, Linz (4); Vika Müller, M. Clausen, Hamburg (3); A. Glasow, J. Werhand, P. Lindner, Hildesheim (3); R. Niemann, Angelika Deutschmann, G. Klaushenke, Oldenburg (3); Nicole Mittemeyer, G. Frese, J. Feldkamp, J. Herrmann, Bielefeld (3); J. Diener, F. Grünwald, G. Hör, Frankfurt (2); R. Nastvogel, H. Philipp, M. Schmidt, Bamberg (2); K. Kley, H. Müller, Düsseldorf (1); M. Schmücking, R. Baum, Bad Berka (1); H. Huber, Wilhelmine Maschek, Linz (1); M. Fischer, E. Bell, Dernbach (1); M. Tosch, R. Aurisch, Mönchengladbach (1); Isabel Lauer, M. Bähre, E. Richter, Lübeck (1); H. I. Lux, H. Tonscheid, K. Hering, Dortmund (1); M. Hacker, P. Knesewitsch, K. Hahn, München (1); R. Thiele, Schwäbisch Hall (1); J. Hof-

fend, Sabine Haufe, U. Haberkorn, Heidelberg (1); K. Biermann, S. Ezziddin, K. von Mallek, H. Biersack, Bonn (1);

Conflict of interest The authors declare that they have no conflict of interest.

References 1. Biermann M, Pixberg M, Schuck A et al. External beam radiotherapy. In: Biersack H-J, Grünwald F (eds). Thyroid cancer. Heidelberg: Springer 2005; 139–161. 2. Biermann M, Pixberg MK, Dörr U et al. Guidelines on radioiodine therapy for differentiated thyroid carcinoma. Impact on clinical practice. Nuklearmedizin 2005; 44: 229–237. 3. Biermann M, Pixberg MK, Schuck A et al. Multicenter Study Differentiated Thyroid Carcinoma (MSDS): Diminished acceptance of adjuvant external beam radiotherapy for differentiated thyroid carcinoma. Nuklearmedizin 2003; 42: 244–250. 4. Biermann M, Schober O. GCP-compliant management of the Multicentric Study Differentiated Thyroid Carcinoma (MSDS) with a relational database under Oracle 8i. Inform Biom Epidemiol Med Biol 2002; 33: 441–459. 5. Chow SM, Law SC, Mendenhall WM et al. Papillary thyroid carcinoma: Prognostic factors and the role of radioiodine and external radiotherapy. Int J Radiat Oncol Biol Phys 2002; 52: 784–795. 6. Deutsche Krebsgesellschaft. Malignant goiter. In: Junginger T, Hossfeld DK, Müller R-P (eds). Guidelines for the diagnosis and therapy of tumors of the gastrointestinal tract and the thyroid gland. Stuttgart: Thieme 1999; 204–213. 7. Deutsche Krebsgesellschaft. Malignant thyroid tumors. Concise interdisciplinary guidelines 2000. Germering: Zuckschwerdt 2000; 92–106. 8. Dietlein M, Dressler J, Eschner W et al. Procedure guideline for iodine-131 whole-body scintigraphy for differentiated thyroid cancer. Nuklearmedizin 1999; 38: 213–214. 9. Dietlein M, Dressler J, Farahati J et al. Guidelines for radioiodine therapy (RIT) in differentiated thyroid cancer. Nuklearmedizin 1999; 38: 221–222. 10. Eichhorn W, Tabler H, Lippold R et al. Prognostic factors determining long-term survival in well-differentiated thyroid cancer: an analysis of four hundred eighty-four patients undergoing therapy and aftercare at the same institution. Thyroid 2003; 13: 949–958. 11. Esik O, Nemeth G, Eller J. Prophylactic external irradiation in differentiated thyroid cancer: a retrospective study over a 30-year observational period. Oncology 1994; 51: 372–379. 12. Farahati J, Reiners C, Stuschke M et al. Differentiated thyroid cancer. Impact of adjuvant external radiotherapy in patients with perithyroidal tumor infiltration (stage pT4). Cancer 1996; 77: 172–180. 13. Georgi P, Emrich D, Heidenreich P et al. Radioiodine therapy of differentiated thyroid cancer. Recommendations of the Therapy Study Group of the German Society of Nuclear Medicine. Nuklearmedizin 1992; 31: 151–153.

© Schattauer 2009

Nuklearmedizin 3/2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.

97

98

Biermann et al.: External beam radiotherapy for differentiated thyroid cancer

14. Hölzer S, Steiner D, Bauer R et al. Current practice of radioiodine treatment in the management of differentiated thyroid cancer in Germany. Eur J Nucl Med 2000; 27: 1465–1472. 15. Ito Y, Tomoda C, Uruno T et al. Prognostic significance of extrathyroid extension of papillary thyroid carcinoma: massive but not minimal extension affects the relapse-free survival. World J Surg 2006; 30: 780–786. 16. Junginger T, Hartel W. Guidelines for the therapy of malignant thyroid tumors. Mitteilungen der Deutschen Gesellschaft für Chirurgie 1996; 3/1996. 17. Kim TH, Yang DS, Jung KY et al. Value of external irradiation for locally advanced papillary thyroid cancer. Int J Radiat Oncol Biol Phys 2003; 55: 1006–1012. 18. Lee D, Cosmatos D, Marcial V. Results of an RTOG phase III trial (RTOG 85–27) comparing radiotherapy plus etonidazole with radiotherapy alone for locally advanced head and neck cancer. Int J Radiat Oncol Biol Phys 1995; 32: 567–576. 19. Leisner B, Degelmann G, Dirr W et al. Therapeutic results of malignant goiter 1960–1980. Significance of percutaneous radiotherapy in well differentiated carcinomas. Dtsch Med Wochenschr 1982; 107: 1702–1707. 20. Lerch H, Schober O, Kuwert T, Saur HB. Survival of differentiated thyroid carcinoma studied in 500 patients. J Clin Oncol 1997; 15: 2067–2075.

21. Machens A, Holzhausen HJ, Dralle H. The prognostic value of primary tumor size in papillary and follicular thyroid carcinoma. Cancer 2005; 103: 2269–2273. 22. Rahbar K, Hutzenlaub V, Fischer RJ et al. Risk-profile and outcome of small papillary and follicular thyroid carcinomas (≤1 cm). Nuklearmedizin 2008; 47: 188–193. 23. Riemann B, Schober O. Are „high-risk“ patients with differentiated thyroid cancer „healthy patients“? Nuklearmedizin 2008; 47: 179–180. 24. Saur HB, Lerch H, Schober O. The survival rate of patients with a differentiated thyroid carcinoma without primary percutaneous irradiation of the neck area. Strahlenther Onkol 1996; 172: 306–311. 25. Sautter-Bihl ML, Herbold G, Heinze HG, Bihl H. Postoperative external beam radiation in differentiated thyroid cancer: situations in which radioiodine is insufficient? Some dosimetric considerations with Monte Carlo techniques. Strahlenther Onkol 1991; 167: 267–272. 26. Schmoor C, Olschewski M, Schumacher M. Randomized and non-randomized patients in clinical trials: experiences with comprehensive cohort studies. Stat Med 1996; 15: 263–271. 27. Schuck A, Biermann M, Pixberg MK et al. Acute toxicity of adjuvant radiotherapy in locally advanced differentiated thyroid cancer: First results of

28.

29.

30.

31.

32. 33. 34. 35.

the Multicenter Study Differentiated Thyroid Cancer. Strahlenther Onkol 2003; 179: 821–827. Seegenschmiedt MH, Sautter-Bihl ML, Schmitt G et al. Guidelines in radiooncology. External beam radiotherapy of malignant struma. AWMF-Guideline registry no. 052/001, 1998. Sherman SI, Brierley JD, Sperling M et al. Prospective multicenter study of thyroid carcinoma treatment: initial analysis of staging and outcome. National Thyroid Cancer Treatment Cooperative Study Registry Group. Cancer 1998; 83: 1012–1021. Simpson WJ, Panzarella T, Carruthers JS et al. Papillary and follicular thyroid cancer: impact of treatment in 1578 patients. Int J Radiat Oncol Biol Phys 1988; 14: 1063–1075. Tsang RW, Brierley JD, Simpson WJ et al. The effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer 1998; 82: 375–388. Wittekind C. TNM Supplement: A commentary on uniform use (UICC). 3rd ed. 2003. Wittekind C, Meyer HJ, Bootz F. TNM classification of malignant tumors. 6th ed. Berlin: Springer 2002. Wittekind C, Wagner G. TNM classification of malignant tumors. 5th ed. Berlin: Springer 1997. Wu XL, Hu YH, Li QH et al. Value of postoperative radiotherapy for thyroid cancer. Head Neck Surg 1987; 10: 107–112.

Nuklearmedizin 3/2009

© Schattauer 2009

Downloaded from www.nuklearmedizin-online.de on 2017-01-29 | IP: 37.44.207.189 For personal or educational use only. No other uses without permission. All rights reserved.