ESTRO 36 Abstract Book

S304

ESTRO 36 2017

_______________________________________________________________________________________________

as identifying patients' care needs. With the increase in

long-term remote follow-up of patients PROMs play an

important role as they offer the opportunity to assess and

address the health concerns or health-related quality of

life (HRQOL) issues of individual patients [1]. Other

important clinical applications of PROMs include aiding

treatment choices as well as identifying high risk patients

who may have poorer long-term health-related outcomes

[2, 3]. These are all key challenges of modern oncology,

and PROMs play a strategic role in this as they enable the

tailoring of treatments according to the priorities, risks or

concerns of individual patients.

The successful application of PROMs requires a deeper

understanding of the methods for extracting information

carried within PROMs [4]. PROMs data are complex, with a

large number of variables (HRQOL, symptoms, function,

bother, performance or health concerns) measured on

different scales (with different levels, ratios or

frequencies). Symptom clusters are groups of 3 or more

correlated symptoms that occur together, and this is

stable over time [5, 6]. Symptom clusters can be easily

determined specifically to each dataset or clinical trial

[7]. This can be used as a method of grouping symptoms

for the purpose of summarising PROMs and extracting

meaningful information. The advantage of exploring

symptom clusters within a dataset is that it allows a study

specific method of grouping symptoms. Because of this

symptom clusters have the potential to improve sensitivity

and specificity to symptom grouping. Only items that are

strongly correlated, and so measure the same underlying

health concern, are included in summative scores. This

can be utilised in PROMs data modelling or clinical decision

making.

In clinical trials PROMs are often seen as a research tool

and it can be challenging to deliver real-time clinical

applications. PROMs can be difficult for patients to

complete, and missing data is another common problem

when analysing and interpreting PROMs [8]. Some of the

causes of missing data include the complexity of long and

multiple PROMs questionnaires, lack of feedback following

the delivery of PROMs, difficulty in understanding

questions or language issues (potentially associated with

migration), patients missing their appointments or

dropping-out of studies, and intermittent missingness

when patients fail to complete some of the questions. All

this contributes to the degree of missing data and in turn

a reduction in sample size, limited analytical applications

or even the risk of biased results. As treatments evolve

and the characteristics of patient populations change,

study specific approaches to analysing PROMs are

warranted. The correlation and grouping of items, missing

data, and the ceiling or floor effect in collected data

should all be investigated for each study when interpreting

and analysing PROMs. This may advance PROMs data

analysis and lead to the extraction of more relevant and

meaningful information.

1. Horwitz EM, Bae K, Hanks GE, Porter A, Grignon DJ,

Brereton HD et al. Ten-Year Follow-Up of Radiation

Therapy Oncology Group Protocol 92-02: A Phase III Trial

of the Duration of Elective Androgen Deprivation in Locally

Advanced Prostate Cancer. Journal of Clinical Oncology.

2008;26(15)

2. Weldring T, Smith SMS. Patient-Reported Outcomes

(PROs) and Patient-Reported Outcome Measures (PROMs).

Health Services Insights. 2013;6

3. Warrington L, Absolom K, Velikova G. Integrated care

pathways for cancer survivors - a role for patient-reported

outcome measures and health informatics. Acta Oncol.

2015;54(5)

4. Faithfull S, Lemanska A, Chen T. Patient-reported

Outcome Measures in Radiotherapy: Clinical Advances and

Research Opportunities in Measurement for Survivorship.

Clin Oncol. 2015;27(11)

5. Aktas A. Cancer symptom clusters: current concepts and

controversies. Curr Opin Support Palliat Care. 2013;7(1)

6. Dodd MJ, Miaskowski C, Paul SM. Symptom clusters and

their effect on the functional status of patients with

cancer. Oncol Nurs Forum. 2001;28(3)

7. Skerman HM, Yates PM, Battistutta D. Multivariate

methods to identify cancer-related symptom clusters. Res

Nurs Health. 2009;32(3)

8. Gomes M, Gutacker N, Bojke C, Street A. Addressing

Missing Data in Patient-Reported Outcome Measures

(PROMS): Implications for the Use of PROMS for Comparing

Provider Performance. Health Econ. 2016;25(5)

Symposium: Hypofractionation in prostate cancer

SP-0583 Moderate hypofractionation in prostate

cancer: what have we learnt from phase 3 trials

D.P. Dearnaley

Institute of Cancer Research, Academic Radiotherapy,

London, United Kingdom

Evidence has accumulated suggesting that prostate cancer

(PCa) may be particularly sensitive to radiation fraction

size. This has considerable implications for the delivery

of radical radiation treatments suggesting that shorter

treatments using higher dose/fraction schedules might

improve the therapeutic ratio and make treatment more

convenient for patients as well as using radiotherapy

resource more effectively. Four large randomised

controlled trials testing modest hypofractionation for

localised PCa have reported efficacy and side effect

outcomes within the last year

(1-4)

. The largest trial,

CHHiP, which included 3216 patients compared standard

fractionation (SFRT) using 2.0Gy daily fractions (f) (total

dose 74Gy) with two experimental hypofractionated

(HFRT) schedules using 3.0Gy/f (total doses of 60Gy and

57Gy)

(1)

. The trial used a non-inferiority design and

demonstrated that HFRT at 60 Gy was non–inferior to

SFRT. Five year disease control rates defined by

biochemical (PSA)/clinical failure free outcome were for

HFRT (60Gy) 90.6% (95% confidence intervals 88.5 - 92.3)

compared with SFRT 88.3% (86.0 - 90.2) (hazard ratio 0.84,

(95% CI: 0.65 – 1.07)); treatment related toxicities were

low and similar. A complementary study design was used

in the PROFIT trial

(2)

which included 1206 patients and

compared SFRT using 2.0Gy/f (total dose 78Gy) with the

same HFRT schedule of 3.0Gy/f (total dose 60Gy). HFRT

was again shown to be non-inferior to SFRT with identical

21% biochemical/clinical failure rates at 5 years. In

PROFIT gastro-intestinal side effects were increased in the

SFRT group compared with HFRT group probably due to

the higher SFRT dose given compared with

CHHiP. Intensity modulated radiotherapy methods (IMRT)

using either forward or inverse planning with a 3 part

simultaneous integrated boost were used in all patients in

the CHHiP trial. IMRT/IGRT methods were used in the

PROFIT trial. A key difference between the trials was the

use of 6 months neoadjuvant androgen deprivation

therapy (ADT) in CHHiP whilst RT alone was used in PROFIT

which probably explains the 11% higher biochemical

control rate in CHHiP. Both investigator groups suggested

that HFRT (60Gy/20f in 4 weeks) could be considered a

new standard of care. In contradistinction authors of the

HYPRO study

(3)

came to different conclusions testing dose

escalated HFRT. 804 patients received either SFRT 78Gy

in 2Gy daily fractions or HFRT giving 64Gy in 3·4Gy

fractions but importantly treating with three fractions per

week and therefore protracting overall treatment time

(OTT). The gain in tumour control was smaller than might

have been expected from radiobiological models (HFRT

80.5% vs. CFRT 77.1%) and not statistically significant. The

relatively unfavourable side effect profiles may be due to

the higher HFRT doses delivered. The trial failed to