ESTRO 36 Abstract Book

S18

ESTRO 36 2017

_______________________________________________________________________________________________

France

Centre Hospitalier Lyon Sud, Service Hématologie, Lyon,

France

Hôpital Saint Louis, Hematology and BMT-, Paris, France

Nijmegen Medical Centre, Heamtology, , The

Netherlands

Erasmus MC Cancer Institute- University Medical Center,

Hematology, Rotterdam, The Netherlands

CHU de Lille and Université de Lille II., Hematology and

BMT- INSERME U955, Lille, France

Chaim Sheba Medical Center, Hematology division BMT

and cord blood bank, Tel-Hashomer, Israel

Purpose or Objective

Total-body irradiation (TBI) has an historical established

role in preparative regimens used before allogeneic

transplant in both acute lymphoblastic leukemia (ALL) and

acute myeloid leukemia (AML). The most popular

myeloablative conditioning consists of 12Gy delivered in 6

fractions (2Gy twice daily for 3 days) in combination with

cyclophosphamide. This schedule of treatment delivery is,

however, time-consuming and became less popular in the

radiation oncology community in the era of development

of new technologies. The aim of the SARAZIN study was to

analyze the impact of the modified myéloablative

fractionated TBI regimens as compared to the standard 6

fractions-schedule on outcome of patients undergoing

allotransplant for ALL and AML.

Material and Methods

We retrospectively compared myeloablative TBI regimens

of 3126 patients registered in the EBMT database

transplanted between 2000 and 2014 for ALL (n=1783) or

AML (n=1343). Pre-transplant chemotherapy consisted

mainly of cyclophosphamide (Cy) in 92% and 97% of ALL

and AML patients, respectively. TBI was delivered as

either 12Gy in 6 fractions (group 1; ALL, n=1362 and AML,

n=857), or single dose TBI (STBI) (group 2; ALL, n=54 and

AML, n=79), or 9-12Gy in 2 fractions (group 3; ALL, n=173

and AML, n=256), or 12Gy in 3-4 fractions (group 4; ALL,

n=194 and AML, n=151). The majority (70%-79%)* of ALL

and AML (57%-79%) patients were grafted in 1

complete

remission (CR1). The rate of transplants from unrelated

donors was higher in ALL (24%-50%) vs AML (20%-37%).

Results

The median follow-up was 61 months and 85 months in the

ALL and AML patients, respectively. At 5 years, leukemia

free survival (LFS), overall survival (OS), relapse incidence

(RI) and non-relapse mortality (NRM) were 46.5%, 50.4%,

29%, 24.5% in ALL and 45.7%, 48%, 30.4% and 23.8%,

respectively. LFS at 5y in AML and ALL patients were

respectively: 48% and 45%, 32% and 45%, 45% and 53%, 42%

and 50% in the 4 TBI groups (p=0.082 for AML and p=0.32

for ALL). Additionally, for both AML and ALL, no statistical

significance was found between the 4 TBI groups for OS

(p=0.82 in ALL; p=0.11 in AML), RI (p=0.29 in ALL; p=0.23

in AML) and for NRM (p=0.58 in ALL; p=0.12 in AML).

In multivariate analyses of TBI schedules, comparing the

different schedules to the standard 12Gy in 6 fractions

(group 1 vs group 2; group 1 vs group 3; group 1 vs group

4), fractionation was not found as independent prognostic

factor neither in ALL nor in AML patients for LFS, OS, RI or

NRM.

Conclusion

The SARASIN study showed that using a TBI dose of 12Gy

as pre allogeneic transplantation, fractionation has no

impact on relapse or survival neither in ALL, nor in AML

patients. The reduction of the number of fractions even in

this rather high TBI dose level is not associated with

increased risk of NRM. Altogether, our data suggests that

12Gy could be delivered safely in less than 6 fractions.

This may lead to increase TBI availability as pre

transplantation conditioning regimens in acute leukemia

patients.

PV-0043 Radiotherapy to the mediastinum in

Hodgkin's lymphoma: Is B-VMAT the only arc solution?

C. Hanna

, C. Featherstone

, S. Smith

NHS Greater Glasgow and Clyde, Clinical Oncology,

Glasgow, United Kingdom

NHS Greater Glasgow and Clyde, Radiotherapy planning

and imaging, Glasgow, United Kingdom

Purpose or Objective

Patients treated for Hodgkin’s lymphoma have a long life

expectancy. It is therefore essential to restrict dose to

organs at risk (OARs) to reduce long-term toxicity when

using consolidation radiotherapy treatment. Smaller

volumes and lower doses have made an impact but it is

also crucial that the correct radiotherapy technique is

chosen. "Butterfly" arc therapy (B-VMAT) is a technique

that has gained popularity with the claim of reducing dose

to heart, lung and breast tissue. This project aims to

assess if B-VMAT offers any advantage over 3D conformal

or alternative arc techniques as a solution to reduce long-

term toxicity in these patients.

Material and Methods

Review of case records and planning CT scans for 8

patients with mediastinal lymphoma treated with

radiotherapy in the past 12 months. We produced a list of

"aspirational" dose constraints (DCs) (Table 1) after review

of current literature. These DCs were used to compare five

different planning techniques (Figure 1) for each patient

(Varian/Eclipse Version 13). Mean OAR dose values were

calculated for each technique and a paired t-test was used

to compare conformity of B-VMAT to the other techniques.

Results

Patients were aged between 20-42 years; 4 male and 4

female.

Heart Dmean of 15Gy was met for all plans except one

(AP:PA) and 10Gy was mostly achievable. B-VMAT and

ARC-A were the optimal plans in terms of OAR dose except

for one patient when there was most overlap between

heart and PTV. In this case the ARC-F was superior.

Lung-PTV Dmean of 12Gy was achieved in all but one plan

(AP:PA). Arc therapies achieved better V20 doses

compared to 3DCRT. Arc therapy, in general, did not

generate high V5s except for ARC-F which failed to meet

the V5 constraint for all patients.

Breast doses were similar for arc and 3D conformal plans

except when using ARC-F which was inferior. Dmean of

2Gy was always met. V4 <5% was met for all plans except

ARC-F.

Table 1 displays mean OAR doses for all 8 patients for each

technique. ARC-A and B-VMAT both consistently out-

perform 3D-CRT, ARC-F and hybrid without a substantial

increase in dose bathing. Conformation number (CN) was

significantly better for ARC-F, significantly worse for

AP:PA and equivalent for ARC-A and hybrid when

compared to B-VMAT. Although superior or equivalent in

conformality compared to B-VMAT, both ARC-F and hybrid

techniques were nevertheless inferior in reducing OAR

doses.