The impact of pre‐operative intravenous iron on quality of life after colorectal cancer surgery: outcomes from the intravenous iron in colorectal cancer‐associated anaemia (IVICA) trial

Anaemia is associated with a reduction in quality of life, and is common in patients with colorectal cancer . We recently reported the findings of the intravenous iron in colorectal cancer‐associated anaemia (IVICA) trial comparing haemoglobin levels and transfusion requirements following intravenous or oral iron replacement in anaemic colorectal cancer patients undergoing elective surgery. In this follow‐up study, we compared the efficacy of intravenous and oral iron at improving quality of life in this patient group. We conducted a multicentre, open‐label randomised controlled trial. Anaemic colorectal cancer patients were randomly allocated at least two weeks pre‐operatively, to receive either oral (ferrous sulphate) or intravenous (ferric carboxymaltose) iron. We assessed haemoglobin and quality of life scores at recruitment, immediately before surgery and at outpatient review approximately three months postoperatively, using the Short Form 36, EuroQoL 5‐dimension 5‐level and Functional Assessment of Cancer Therapy – Anaemia questionnaires. We recruited 116 anaemic patients across seven UK centres (oral iron n = 61 (53%), and intravenous iron n = 55 (47%)). Eleven quality of life components increased by a clinically significant margin in the intravenous iron group between recruitment and surgery compared with one component for oral iron. Median (IQR [range]) visual analogue scores were significantly higher with intravenous iron at a three month outpatient review (oral iron 70, (60–85 [20–95]); intravenous iron 90 (80–90 [50–100]), p = 0.001). The Functional Assessment of Cancer Therapy – Anaemia score comprises of subscales related to cancer, fatigue and non‐fatigue items relevant to anaemia. Median outpatient scores were higher, and hence favourable, for intravenous iron on the Functional Assessment of Cancer Therapy – Anaemia subscale (oral iron 66 (55–72 [23–80]); intravenous iron 71 (66–77 [46–80]); p = 0.002), Functional Assessment of Cancer Therapy – Anaemia trial outcome index (oral iron 108 (90–123 [35–135]); intravenous iron 121 (113–124 [81–135]); p = 0.003) and Functional Assessment of Cancer Therapy – Anaemia total score (oral iron 151 (132–170 [69–183]); intravenous iron 168 (160–174 [125–186]); p = 0.005). These findings indicate that intravenous iron is more efficacious at improving quality of life scores than oral iron in anaemic colorectal cancer patients.


Introduction
Colorectal malignancy is often associated with anaemia, with a reported incidence of up to 40% in newly diagnosed cases [1]. The aetiology of this anaemia is frequently due to iron deficiency secondary to chronic blood loss (absolute iron deficiency) or impaired utilisation of iron stores (iron sequestration, and functional iron deficiency) [2,3]. In addition, treatment of the underlying colorectal cancer using surgery or chemotherapy can lead to a worsening of anaemia in these patients.
It is recognised that anaemia causes a variety of symptoms including fatigue, lethargy and dyspnoea [4]. It has been shown that reducing haemoglobin (Hb) levels are associated with decreasing quality of life (QoL) scores in the context of malignancy, and hence it has been proposed that reversal of this anaemia will improve cancer-related QoL [4].
Furthermore, in relation to operative cases, there has been a recent focus on standardising end points in peri-operative medicine, with cancer-related QoL emerging as a key patient-centric end point [5].
Iron replacement therapies such as oral iron are associated with deleterious side-effects including abdominal pain, constipation and diarrhoea. Treatment non-adherence rates attributed to such side-effects have been reported to be in the region of 40% [6]. In addition, absorption pathways and access to oral iron supplementation may be impaired in patients with malignancy [4]. Newer intravenous iron preparations have been developed which are proposed to offer safer, better tolerated and more efficacious treatment of iron deficiency anaemia [7].
We aimed to compare the QoL scores of colorectal cancer patients who were randomly allocated to receive either oral or intravenous iron as pre-operative treatment for their anaemia, in order to review if either treatment conferred an advantage in terms of improving QoL scores.

Methods
We conducted this multicentre study in accordance with the Declaration of Helsinki, with full ethical approval from the National Research and Ethics Service, East Midlands, Nottingham. We registered the study with both the MHRA and Clinical Trials.Gov. We obtained written informed consent from all study participants.
We have previously reported the methods as part of a trial comparing blood transfusion rates of anaemic colorectal cancer treated with pre-operative oral and intravenous iron [8]. Anaemic colorectal cancer patients with non-metastatic disease were randomly allocated preoperatively in a 1:1 fashion using variable block allocation, stratified by sex and age, to receive either oral iron (ferrous sulphate 200 mg twice daily until surgery) or intravenous iron (ferric carboxymaltose -Ferinject ™ ; Vifor Pharma, Glattbrugg, Switzerland) dosed by weight and haemoglobin in accordance with the summary of product characteristics. Treatment allocation was un-blinded owing to the change in stool colour associated with oral iron supplementation. To minimise the risks of including patients with non-iron deficiency anaemia, those with the following conditions were not included: metastatic disease; preexisting haematological disease; renal failure; and those currently undergoing chemotherapy. All patients were included at recruitment and surgery. Only those who underwent resectional surgery and attended the outpatient follow-up were included at outpatient review.
Quality of life assessments and haemoglobin measurements were performed at the following timepoints: recruitment before iron administration; on the day of surgery before intervention; and at their outpatient followup visit between two and three months following discharge.
If an outpatient appointment was expedited due to a complication, this appointment was not used for trial purposes, and review was delayed until the subsequent appointment falling within the correct 2-3-month postoperative period. This was to ensure that all reviews occurred at a comparable postoperative time-point.
The QoL measures we used included the EuroQoL 5dimension 5-level (EQ5D5L) [9] questionnaire and the modified Short Form 36 v1 (SF36) [10] as overviews of general well-being. These were augmented with the Functional Assessment of Cancer Therapy -Anaemia (FACT-An) questionnaire [11]. This validated questionnaire assesses specific quality of life concerns related to anaemia and fatigue in cancer patients.
The EQ5D5L questionnaire has been widely used in cancer and cancer-related anaemia studies [12]  of the intravenous iron group patients received the drug and did not receive oral iron. This study's oral iron treatment protocol was adhered to by 50 out of 55 (91%) patients who did not have the date of surgery moved; no patients randomly allocated to oral iron received intravenous iron.
Patients were well-matched across groups (Table 1).
Postoperative recovery was similar across both groups.
There were no significant differences in complication rates   Table 3)    Table 4.

Discussion
We found that intravenous iron resulted in a faster clinically evident increase in QoL scores than oral iron, and may be more efficacious at improving QoL scores in anaemic colorectal cancer surgical patients. The differences seen were most profound over a longer duration from initiation of treatment, which is expected given the lag between intravenous iron administration and response, as noted in previous trials [19]. Despite this, the significant clinical effect of intravenous iron was also evident after short periods of pre-operative optimisation. This benefit of intravenous iron was not solely limited to the specific symptomatology of anaemia but was also evident across generic measures of well-being. We believe that it is most likely that the  [18,23].
This threshold of discrimination for changes in healthrelated quality of life has been validated clinically [23], and is based on the psychological assessment of the limits of human discrimination [24]. Measures of Vitality (SF36), Functional Well-Being and specific scores of anaemia symptomatology (FACT-AN) still met this higher threshold of MCD, and hence changed significantly even over the short period from recruitment to day of surgery. These components would appear to be closely linked to anaemia and thus also Hb levels ( Table 2).
The QoL tools employed requested patients to report their QoL over periods ranging from 1-4 weeks.
Considering that the initial time period of iron treatment from recruitment to surgery was in the order of 3 weeks, there would have been a degree of overlap of the pretreatment period when QoL was reported on the day of surgery, leading to perhaps an underestimation of the   clinically-relevant increases in scores were seen with intravenous iron at the point of surgery, few scores were significantly higher than those within the oral iron group. We also acknowledge that current recommendations suggest at least a three-week timeframe for haemoglobin incrementation with intravenous iron administration [25].
Likewise, oral iron was given over a similar timescale, and in both cases the treatment effect may have been underestimated. However, our trial was designed to be pragmatic and applicable to current cancer treatment timelines. It must be acknowledged that in some European countries the timing of surgery for malignancy is subject to legal regulations [26]. Consequently, a balance between timely cancer treatment, haemoglobin improvement and quality of life is needed. Therefore, the constraints of the clinical timeline were factored into the trial design.
It must be re-emphasised that in the current randomised controlled trial, the study was powered to detect a difference in transfusion rates, and not for QoL as the primary outcome measure. QoL outcomes were specified as secondary outcomes in the original trial design, and we, therefore, acknowledge that further studies powered to analyse QoL outcomes are required to validate these findings. Furthermore, due to difficulties in concealing oral iron administration from patients due to stool discoloration [27], the study was not blinded by design. This does leave our study vulnerable to the placebo effect, and the influence of patient beliefs on QoL perception. In addition, questionnaires were distributed at follow-up between two and three months following surgery. Quality of life can change over time, and this may have, therefore, influenced patient-reported scores. There were, however, no significant differences in mean time to postoperative follow-up between the two groups.
Haemoglobin was positively correlated with six subscales across all three QoL questionnaires. Although we found that the correlation was modest, it is important to acknowledge that there are a multitude of factors influencing QoL. Therefore, it could be argued that over this timescale, the improvements in QoL scores seen with Hb changes as a discrete factor emphasise the clinical importance of treating anaemia in this patient group. On review of the entire cohort, it appeared that as Hb increases approached the 30 g.l À1 increment, small improvements in QoL scores were evident, which rose to a moderate clinical effect when changes exceeded this mark. This could be used as a target for both future research and in clinical practice to guide therapy in this patient population.