Chrisnutting-oncology.co.uk

The British Journal of Radiology, 78 (2005), 405–410 E 2005 The British Institute of Radiology Management of non-Hodgkin’s lymphoma of the thyroid: theRoyal Marsden Hospital experience 1,2K J HARRINGTON, MRCP, FRCR, 1V J MICHALAKI, MD, 3L VINI, FRCR, 1,2C M NUTTING, MD, MRCP, FRCR,1K N SYRIGOS, MD, 4R A’HERN, PhD and 3C L HARMER, FRCP, FRCR 1Head and Neck Cancer Unit, Royal Marsden Hospital, Fulham Road, London, 2Institute of Cancer Research, FulhamRoad, London, 3Thyroid Unit, Royal Marsden Hospital, Fulham Road, London and 4Department of Statistics, RoyalMarsden Hospital, Fulham Road, London, UK Abstract. A retrospective review was conducted of patients treated for thyroid non-Hodgkin’s lymphoma(TNHL) at the Royal Marsden Hospital between 1936 and 1996 to determine the effect of radiotherapy (RT) onoutcome. 91 patients were identified from the Thyroid Unit Database. There were 77 females and 14 males witha median age of 65 years (range 22–87 years). RT was delivered according to two separate policies: (1) involvedfield radiotherapy (IFRT) to the thyroid bed and cervical lymph nodes; (2) extended field radiotherapy (EFRT)covering the thyroid bed, cervical and mediastinal lymph nodes. 89 patients received RT as part of definitivetreatment following surgery, to a dose of approximately 40 Gy. 25 patients received IFRT and 64 patientsEFRT. 27 patients received cytotoxic chemotherapy. 18 patients (72%) treated with IFRT died of TNHL with amedian relapse free survival (RFS) of 10 months and a median overall survival (OS) of 21 months. In contrast,only 29 patients (46%) treated with EFRT died of TNHL with a median RFS of 76 months (p50.01 for RFSwith respect to IFRT and p50.04 for OS). Significantly more patients treated with IFRT relapsed locally (52%vs 27%). There was no difference in the rates of systemic relapse (20% vs 22%). EFRT alone for Stage I, but notfor Stage II disease, yielded acceptable rates of local control and disease free survival with doses of at least40 Gy. These historical data strongly support the addition of combination chemotherapy to the treatmentregimen in all patients with Stage II disease. Indeed, in recent years this has become the standard of care for allcases of thyroid lymphoma unless the histology is of marginal zone type (mucosa associated lymphoma tissue(MALT) lymphoma).
Thyroid non-Hodgkin’s lymphoma (TNHL) is an shift towards using a combination of chemotherapy and uncommon tumour, representing 2–8% of thyroid malig- RT in all cases other than mucosa associated lymphoma nancies and approximately 1–2% of extranodal lympho- tissue (MALT) lymphoma of the thyroid gland. As mas [1–4]. It occurs most frequently in elderly females and regards therapeutic irradiation, there remain a number has been linked to Hashimoto’s thyroiditis and prior of unresolved issues regarding the volume of tissue that therapeutic irradiation of the thyroid bed. Staging should should be treated and the dose prescription required.
include CT scanning of the neck, thorax, abdomen and There are two main options in the selection of radiation pelvis, together with bone marrow aspirate and trephine treatment fields: (1) involved field RT (IFRT) which aims to include the entire thyroid gland and local neck nodes; There exists no universally accepted standard of care for and (2) extended field RT (EFRT) which involves treating TNHL and a number of controversies exist regarding the the thyroid, neck, mediastinal and in some centres, axillary roles of surgery, radiotherapy (RT) and chemotherapy in the management of this disease [5–12]. Previously, surgery These two options have never been formally compared occupied a pre-eminent place in management and most in a randomized study and in view of the rarity of this patients underwent extensive resections. In recent years, condition; such a study is unlikely to be conducted. A a however, the appreciation that TNHL is sensitive to RT result, the choice between IFRT and EFRT has been a and chemotherapy has resulted in a move towards limited matter of clinical judgement based on prior experience.
surgical intervention, usually in the form of a diagnostic Similarly, formal evaluation of the optimal radiation biopsy followed by definitive RT (or chemotherapy dose required for local control has never been undertaken.
followed by RT). However, there are opponents of this In line with treatment of nodal and non-thyroid extra- view who recommend excision or debulking of disease nodal NHL, most patients with TNHL have received followed by post-operative RT [13–22].
doses in the order of 40 Gy. Previous studies have In spite of the controversy regarding the surgical suggested that there is a threshold dose below which the management of TNHL, radical RT became the treatment chance of achieving local control is small. Tupchong et al of choice in most centres, although there has now been a [23] reported 0% local control rated for patients whoreceived less than 20 Gy. Therefore, in this retrospectiveanalysis, we have attempted to address these twoimportant questions relating to the management of Received 27 February 2004 and in revised form 28 October 2004,accepted 6 December 2004.
The British Journal of Radiology, May 2005 K J Harrington, V J Michalaki, L Vini et al Table 1. Summary of patient characteristics for 91 patientswith thyroid non-Hodgkin’s lymphoma A search of the Royal Marsden Hospital (RMH) Thyroid Database was undertaken for a period that spanned the years between 1936 and 1996.
This database contains a thorough summary of over 2000 patients presenting to RMH with thyroid malignancy over the last six decades. A total of 91 patients with histologically proven NHL were identified. Case notes were reviewed in order to collect demographic data, details of disease-related symptoms and tumour stage, surgical and radiotherapeutic management and standard outcome measures, including relapse free (RFS) and overall survival (OS). For the purposes of data collection, the results of staging investigations were accepted at face value and noattempt was made to re-evaluate the staging assigned topatients treated at different times. Similarly, the patholo- gical classification of NHL has undergone a number of The presenting symptoms are documented in Table 1.
changes during the period of time under study [24–29]. In For two patients, no presenting symptoms were recorded a previous study of 46 patients with TNHL from RMH, but the vast majority (98%) presented with neck swelling.
all of whom are included in this analysis, it was shown that Local obstructive symptoms (stridor, dysphagia and the majority (91%) presented with adverse (intermediate or high grade) histology. Similar findings were apparent for number (34%) had previously been noted to have goitre, this group of patients. No attempt has been made to review previous pathological specimens or re-classify casesaccording to the current REAL classification. Of the 91 patients who were identified, 89 were eligible for analysis In keeping with data presented for other studies, the of the effect of definitive RT. The two patients excluded, majority of patients presented with early stage disease. 35 received primary chemotherapy without RT.
had Stage I disease confined to the thyroid gland and 49had Stage II disease. Only seven patients, were found tohave more advanced Stage III or IV disease at presenta- tion. From analysis of the database, it was apparent thatthese patients presented with massive thyroid enlargement Follow-up data were obtained for all patients treated and evidence of dissemination, rather than as patients with with RT. RFS and OS times were calculated for each widely disseminated lymphoma with incidental involve- patient to the nearest month, taken from the time of ment of the thyroid gland. Systemic B symptoms were presentation to the time of first relapse (for RFS) or death/ uncommon, occurring in only two patients.
All of the patients underwent an initial surgical procedure, the details of which are presented in Table 2.
Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS, Inc., Chicago, IL), statistical software. Survival curves were obtained by the Over the 60 year period in which patients were treated, a Kaplan-Meier method and comparisons were made with number of different treatment protocols were applied the log-rank statistics. Factors found to be significant at involving use of 220–250 kV X-rays, 60Co gamma rays and univariate analysis were considered for multivariate 4–6 MV photons. The median tumour radiation dose was analysis. Cox proportional hazards models were developed 40 Gy (range 4–72 Gy), which was delivered at a dose rate for the multivariate analysis of survival and any value of of between 1.2 Gy and 2.3 Gy per day.
p,0.05 was considered to be statistically significant.
Table 2. Initial surgical procedure performed on the 91 patients with thyroid non-Hodgkin’s lymphoma The median age of the patients was 65 years (range 22– 87 years). There was no significant difference between the median ages of the female (median 65 years, range 27–87 years) and male (median 63 years, range 22–85 years) patients. There were 77 females and 14 males, representing The British Journal of Radiology, May 2005 The majority (61 of 89) received a radiation dose of Table 3. Significant factors on univariate analysis of overall 40 Gy or more. Treatment was usually delivered using parallel-opposed anterior and posterior fields, although, direct anterior, anterior oblique or lateral fields wereemployed occasionally. Spinal cord shielding was not used.
No attempt has been made to assess the effect of technique on outcome. 25 patients received IFRT to the thyroid bed and cervical lymph nodes and 64 patients received EFRT to the thyroid bed, cervical and mediastinal lymph nodes.
In this latter group, there were no data detailing the number of patients in whom the superior mediastinal nodes, as opposed to all mediastinal nodes, were treated.
The axillary nodes were not treated routinely.
27 patients received cytotoxic chemotherapy under a wide variety of circumstances. Two received definitive IFRT, involved field radiotherapy; EFRT, extended field radio- combination chemotherapy without RT. Five patients received induction chemotherapy prior to radical RT, eightreceived adjuvant chemotherapy following RT and 14 The type of surgical procedure performed had a significant received chemotherapy following relapse after RT. Clearly, impact on the outcome of subsequent RT as demonstrated this experience reflects the state of what was considered optimal practice at the time that the patients presented. Incurrent practice, it is likely that the majority of patientswould receive combination chemotherapy followed by 18 of 25 patients (72%) treated with IFRT died of Regimens comprised cyclophosphamide, vincristine and TNHL. The median RFS was only 10 months and the prednisolone (COP), cyclophosphamide, doxorubicin, vin- median OS was 21 months. In contrast, only 29 of 64 patients (45%) treated with EFRT died of TNHL. The mitoxantrone, cyclophosphamide, etoposide, bleomycin median RFS was 76 months, while the median OS has not and vincristine (PMitCEBO). Analysis of the role of yet been reached. The 5-year survival data according to chemotherapy in this cohort of patients was not possible disease stage stratified by radiation field (IFRT or EFRT) because of the small numbers and the different regimens are illustrated in Figures 1 and 2. Patients who received employed. Eight of 13 patients treated initially with EFRT had higher survival rates (p,0.005), (log rank test).
chemotherapy as part of a planned combined approach For patients treated with IFRT, 52% (20% local alone and were long-term survivors. However, results for relapsed 32% combined local and distant) had evidence of loco- disease were generally poor, with only two of eight regional failure at the time of death, compared with only patients relapsing after RT successfully salvaged with 26% (9% local alone plus 17% local and distant), of patients treated with EFRT. There was no significant difference inthe rates of local control between two groups (20% forIFRT vs 22% for EFRT). The patterns of failure of patients dying following IFRT and EFRT are detailed in Table 5.
The overall 5-year survival rate was 48%, with a median RFS of 34 months. Univariate analysis of OS according to patient variables and treatment factors is shown in When the survival data for patients who received Table 3. Adverse prognostic factors were as follows: ,40 Gy were compared with those who received 40 Gy advanced stage, surgical procedure (only biopsy vs moredebulking procedure), extent of radiation field, radiation Table 4. Significant independent factors on multivariate analy- dose of 40 Gy or more, and the presence of stridor at diagnosis. A number of these factors have been implicatedpreviously [29–32]. Analysis of the effect of dysphagia, dysphonia and previous goitre on the OS revealed no significant impact. Independent predictive factors obtained from multivariate analysis are shown in Table 4.
Analysis of the extent of surgical resection (biopsy alone vs more extensive resection), revealed that only in the last decade was there a marked change towards less extensive procedures. Thus, prior to 1990 biopsy was performed in 30 patients compared with 43 who underwent an extensiveresection. After 1990, 13 patients underwent biopsy IFRT, involved field radiotherapy; EFRT, extended field radio- compared with only 5 who had a more extensive resection.
The British Journal of Radiology, May 2005 K J Harrington, V J Michalaki, L Vini et al Table 5. Survival according to stage of disease and extent ofirradiated volume IFRT, involved field radiotherapy; EFRT, extended field radio- Analysis of the data from this large series of patients with TNHL treated over a prolonged period of time in asingle centre provides valuable insights in to the manage-ment of this uncommon disease. RT has traditionallyplayed a major role in treatment of TNHL and thefindings of this study have direct bearing on its use.
Multivariate analysis has highlighted the extent ofradiation field as an independent prognostic factor. RFS Figure 1. Kaplan-Meier curve showing overall survival of and OS rates were significantly better for patients treated Stage I patients stratified by radiation field (extended field with EFRT compared with IFRT; this difference was not radiotherapy vs involved field radiotherapy).
seen in Stage II disease. This effect seemed to be mediatedat the level of the thyroid bed and adjacent lymph nodes in or more, radiation dose was not significant in multivariate the neck and mediastinum because 52% of patients treated analysis of survival (hazard ratio 0.64, 95% confidence with IFRT experienced locoregional failure compared with interval (CI): 0.33–1.24, p50.18). However, the confidence only 27% of patients treated with EFRT. There was no intervals are wide and it is not possible to rule out an difference between the two groups in terms of systemic effect. Radiation dose was significant in multivariate relapse. Therefore, the poor results of IFRT in all disease analysis for local control (hazard ratio 0.42, 95% CI: stages mean that this approach should be abandoned in 0.18–0.94, p50.04). At 5 years, only 7 of 27 (26%) patients favour of more extensive radiation fields that encompass treated at the lower dose level were alive, compared with the mediastinal nodes. For patients with Stage I disease, 35 of 62 (56%) treated to 40 Gy or more (p,0.01, log rank definitive EFRT yielded excellent rates of local control and test). However, the first group included six patients who survival and represented the treatment of choice. However, received doses of less than 30 Gy; none achieved local more recently, combination chemotherapy followed by RT control and their maximal survival was 6 months. The has become the standard of care for this disease and given effect of radiation dose on outcome is presented in the data from this study, EFRT should be considered as Figure 2. Kaplan-Meier curve showing overall survival of Figure 3. Kaplan-Meier curve showing overall survival of Stage II patients stratified by radiation field (extended field patients according to the radiation dose received (,40 Gy vs radiotherapy vs involved field radiotherapy).
The British Journal of Radiology, May 2005 Similarly, analysis of outcomes at radiation doses below 5. Ansell SM, Grant CS, Habermann TM. Primary thyroid 40 Gy versus 40 Gy or more provides clear indications for lymphoma. Semin Oncol 1999;26:316–23.
clinical practice. A threshold dose of 40 Gy or more was 6. Bisbee AC, Thoeny RH. Malignant lymphoma of the thyroid shown to be associated with a significantly improved local following irradiation. Cancer 1975;35:1296–9.
7. Burke JS, Butler JJ, Fuller LM. Malignant lymphomas of the control outcome. This is in keeping with previous data [23] thyroid: a clinical pathologic study of 35 patients including and we suggest a standard dose of 40 Gy in 20 fractions ultra structural observations. Cancer 1977;39:1587–602.
8. Derringer GA, Thompson LD, Frommelt RA, Bijwaard KE, The data on extent of surgical resection indicate a Heffess CS, Abbondanzo SL. Malignant lymphoma of the significantly improved outcome in patients who underwent thyroid gland: a clinicopathologic study of 108 cases. Am J a debulking procedure (defined as anything more than a biopsy) prior to RT and concur with those reported in 9. Ha CS, Shadle KM, Medeiros LJ, et al. Localized non- previous studies [33, 34]. Patients who had total macro- Hodgkin lymphoma involving the thyroid gland. Cancer scopic removal of tumour had the highest rate of local control and long term survival (p,0.001). No significant 10. Holmes HB Jr, Kreutner A, O’Brien PH. Hashimoto’s differences were seen between lobectomy, subtotal or total thyroiditis and its relationship to other thyroid diseases.
Surg Gynecol Obstet 1977;144:887–90.
thyroidectomy, although seven of nine long-term survivors 11. Pledge S, Bessell EM, Leach IH, et al. Non-Hodgkin’s (58–129 months) underwent total thyroidectomy. Despite lymphoma of the thyroid: retrospective review of all patients these findings, it is unlikely that the trend towards diagnosed in Nottinghamshire from 1973 to 1992. Clin Oncol performing open biopsy to make a tissue diagnosis will be reversed. The increasing role of combination chemo- 12. Skarsgard ED, Connors JM, Robins RE. A current analysis therapy and consolidating RT is likely to strengthen the move towards more conservative surgical intervention [32, 13. Baskal N, Erdogan G, Kamel AN, Dagci SS, Akyar S, Ekinci The precise role of combination chemotherapy in C. Localized non-Hodgkin’s lymphoma of the adrenal and TNHL remains to be defined, although in recent years it thyroid glands. Endocrinol Jpn 1992;39:269–76.
has become part of the standard management of this 14. Butler JS Jr, Brady LW, Amendola BE. Lymphoma of the thyroid. Report of five cases and review. Am J Clin Oncol disease. This study demonstrates the generally poor results obtained with RT alone, even with EFRT, in patients with 15. Heimann R. Primary malignant lymphomas of the thyroid. A anything more than Stage I disease and provide strong brief review. Acta Otorhinolaryngol Belg 1987;41:727–35.
support for the use of chemotherapy as part of the primary 16. Ifrah N, Rohmer V, Saint-Andre JP, Jardel H, Boasson M, treatment. The ability of chemotherapy to ‘‘debulk’’ the Bigorgne JC. Primary lymphoma of the thyroid gland.
tumour prior to starting RT (in an analogous fashion to Diagnostic and therapeutic discussion. Apropos of 4 cases.
surgical debulking) represents a further attraction. Indeed, Ann Med Interne (Paris) 1988;139:344–8.
despite the fact that low-grade MALT lymphomas have a 17. Kuma K, Matsuzuka F. Diagnosis and therapy of malignant low distant recurrence rate [26, 28], a review of the thyroid lymphoma. Nippon Naika Gakkai Zasshi 1997;86: published literature suggests that the addition of chemo- therapy to RT significantly reduces both distant and 18. Tennvall J, Cavallin-Stahl E, Akerman M. Primary localized non-Hodgkin’s lymphoma of the thyroid: a retrospective clinicopathological review. Eur J Surg Oncol 1987;13:297– In conclusion, the following recommendations are made for the management of TNHL. IFRT yields unacceptable 19. Tsutsui K, Shibamoto Y, Yamabe H, et al. A radio- rates of disease free survival and OS and can not be recommended. EFRT to a dose of 40 Gy is effective for Hodgkin’s lymphoma: prognostic factors and re-evaluation of treatment modality. Radiother Oncol 1991;21:83–90.
For Stage II–IV disease, EFRT is associated with poor 20. Friedberg MH, Coburn MC, Monchik JM. Role of surgery in outcome; these patients should receive induction chemo- stage IE non-Hodgkin’s lymphoma of the thyroid. Surgery therapy followed by consolidation EFRT to a dose of 21. Rasbach DA, Mondschein MS, Harris NL, Kaufman DS, Wang CA. Malignant lymphoma of the thyroid gland: aclinical and pathologic study of twenty cases. Surgery1985;98:1166–70.
22. Woolner LB, McConahey WM, Beahrs OH, Black BM.
Primary malignant lymphoma of the thyroid. Review of 1. Evans TR, Mansi JL, Bevan DH, Dalgleish AG, Harmer CL.
forty-six cases. Am J Surg 1966;111:502–23.
Primary non-Hodgkin’s lymphoma of the thyroid with bone 23. Tupchong L, Hughes F, Harmer CL. Primary lymphoma marrow infiltration at presentation. Clin Oncol 1995;7:54–5.
of the thyroid: clinical features, prognostic factors, and 2. Heimann R, Vannineuse A, De Sloover C, Dor P. Malignant results of treatment. Int J Radiat Oncol Biol Phys 1986;2: lymphomas and undifferentiated small cell carcinoma of the thyroid: a clinicopathological review in the light of the Kiel 24. Anscombe AM, Wright DH. Primary malignant lymphoma of the thyroid—a tumour of mucosa-associated lymphoid tissue: review of seventy-six cases. Histopathology 1985;9:81– 3. Souhami L, Simpson WJ, Carruthers JS. Malignant lym- phoma of the thyroid gland. Int J Radiat Oncol Biol Phys 25. Fonseca E, Sambade C. Primary lymphomas of the thyroid gland: a review with emphasis on diagnostic features. Arch 4. Vini L, Harmer C. Thyroid. Treatment of Cancer. In: Price P, Sikora K, editors. Treatment of cancer (3rd edn). London: 26. Isaacson PG. Lymphoma of the thyroid gland. Curr Top The British Journal of Radiology, May 2005 K J Harrington, V J Michalaki, L Vini et al 27. Kossev P, Livolsi V. Lymphoid lesions of the thyroid: review 35. Blair TJ, Evans RG, Buskirk SJ, Banks PM, Earle JD.
in light of the revised European-American lymphoma Radiotherapeutic management of primary thyroid lymphoma.
classification and upcoming World Health Organization Int J Radiat Oncol Biol Phys 1985;11:365–70.
classification. Thyroid 1999;9:1273–80.
36. Kanetake H, Toda M, Kawamoto Y. Prognostic factors in 28. Laing RW, Hoskin P, Hudson BV. The significance of MALT primary lymphoma of the thyroid—a review of 74 cases.
histology in thyroid lymphoma: a review of patients from the Nippon Jibiinkoka Gakkai Kaiho 1993;96:1105–11.
BNLI and Royal Marsden Hospital. Clin Oncol 1994;6: 37. Belal AA, Allam A, Kandil A, et al. Primary thyroid lymphoma: a retrospective analysis of prognostic factors and 29. Logue JP, Hale RJ, Stewart AL, Duthie MB, Banerjee SS.
treatment outcome for localized intermediate and high grade Primary malignant lymphoma of the thyroid: a clinicopatho- lymphoma. Am J Clin Oncol 2001;24:299–305.
logical analysis. Int J Radiat Oncol Biol Phys 1992;22: 38. Doria R, Jekel JF, Cooper DL. Thyroid lymphoma. The case for combined modality therapy. Cancer 1994;73:200–6.
30. Sasai K, Yamabe H, Haga H, et al. Non-Hodgkin’s 39. Gospodarowicz MK, Sutcliffe SB, Brown TC, Chua T, Bush lymphoma of the thyroid. A clinical study of twenty-two RS. Patterns of disease in localized extranodal lymphomas.
31. Singer JA. Primary lymphoma of the thyroid. Am Surg 40. Leedman PJ, Sheridan WP, Downey WF, Fox RM, Martin FI. Combination chemotherapy as single modality therapy for 32. Wirtzfeld DA, Winston JS, Hicks WL, Loree TR. Clinical presentation and treatment of non-Hodgkin’s lymphoma of the thyroid gland. Ann Surg Oncol 2001;8:338–41.
41. Vigliotti A, Kong JS, Fuller LM, Velasquez WS. Thyroid 33. Pyke CM, Grant CS, Habermann TM, et al. Non-Hodgkin’s lymphomas stages IE and IIE: comparative results for lymphoma of the thyroid: is more than biopsy necessary? radiotherapy only, combination chemotherapy only, and multimodality treatment. Int J Radiat Oncol Biol Phys 34. Tsang RW, Gospodarowicz MK, Sutcliffe SB, Sturgeon JF, Panzarella T, Patterson BJ. Non-Hodgkin’s lymphoma of thethyroid gland: prognostic factors and treatment outcome. ThePrincess Margaret Hospital Lymphoma Group. Int J RadiatOncol Biol Phys 1993;27:599–604.
The British Journal of Radiology, May 2005

Source: http://www.chrisnutting-oncology.co.uk/docs/thyroid-lymphoma.pdf