A double-blind, randomized trial of pyridoxine versus placebo for the prevention of pegylated liposomal doxorubicin-related hand-foot syndrome in gynecologic oncology patients

Authors

  • Vivian von Gruenigen MD,

    Corresponding author
    1. Department of Obstetrics and Gynecology, Summa Akron City Hospital, Akron, Ohio
    • Department of Obstetrics and Gynecology, Medical Director, Women's Health Services, Summa Akron City and St. Thomas Hospitals, 525 E. Market Street, Med II, PO Box 2090, Akron, OH 44309-2090
    Search for more papers by this author
    • Fax: (330) 375-7549

  • Heidi Frasure MS,

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author
  • Nancy Fusco RN, BSN,

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author
  • Robert DeBernardo MD,

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author
  • Elisa Eldermire RN, BSN,

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author
  • Susan Eaton CCRP,

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author
  • Steven Waggoner MD

    1. Department of Obstetrics and Gynecology, University Hospitals Case Medical Center, Department of Reproductive Biology, Case Western Reserve University School of Medicine; Cleveland, Ohio
    Search for more papers by this author

  • Presented at the 2009 Annual Meeting of the American Society of Clinical Oncology, Orlando, Florida, May 29 to June 2, 2009.

Abstract

BACKGROUND:

The objective of this study was to compare the incidence of hand-foot syndrome (HFS) in patients who received pyridoxine versus placebo during pegylated liposomal doxorubicin (PLD) chemotherapy for recurrent ovarian, breast, or endometrial cancer.

METHODS:

Patients received PLD 40 mg/m2 for a maximum of 6 cycles. Patients were assigned randomly to receive pyridoxine 100 mg twice daily (Group A) or placebo (Group B) and received standard HFS education. Patients completed the Functional Assessment of Cancer Therapy quality-of-life (QOL) questionnaire. The incidence of HFS as measured by common toxicity criteria was compared between groups. Analyses were conducted according to an intent-to-treat basis. Chi-square tests or Fisher exact tests were used.

RESULTS:

Thirty-four patients were enrolled (18 in Group A and 16 in Group B), and 5 patients were unevaluable for HFS assessment. Overall, 15 of 29 patients (52%) had HFS (all grades), and 10 of 29 patients (35%) had grade 2/3 events. Eight of 15 patients in Group A (53%) and 7 of 14 patients in Group B (50%) had HFS (P = .857). For grade 2/3 events, there was no difference between groups: Six of 15 events (40%) occurred in Group A, and 4 of 14 events (29%) occurred in Group B (P = .70). There was no difference in QOL scores between patients who had grade 2/3 HFS and patients who had grade 0/1 HFS. QOL analysis revealed that all patients had elevated social well being.

CONCLUSIONS:

Pyridoxine as administered in the current study did not prevent HFS in patients who received PLD. It is possible that QOL is not compromised in patients with HFS because they may have increased social well being while coping with their disease. Cancer 2010. © 2010 American Cancer Society.

The majority of patients with ovarian cancer present with advanced stage disease, and, unfortunately, many develop recurrent disease. Pegylated liposomal doxorubicin (PLD) is a common chemotherapeutic used to treat recurrent ovarian cancer.1, 2 The efficacy of PLD in the treatment of other malignancies (eg, breast cancer and endometrial cancer) is evident.3-6 However, 1 of the most common side effects of treatment is palmar-plantar erythrodysesthesia, also referred to as hand-foot syndrome (HFS).

HFS occurs in 29% to 49% of patients who receive PLD treatment.7-11 HFS is characterized by paresthesias or tingling of the palms or soles with the potential to develop erythema and edema, and it is complicated by pain, blistering, and cracking of the skin. If HFS occurs, it generally is observed after 2 or 3 cycles of treatment. The mechanism of HFS probably is related to keratinocytes and epidermal transit time. A subclinical inflammatory response at the time of redosing may increase the permeability of skin blood vessels to liposomes.12 Management of HFS may include modification of dose or dosing intervals, cooling methods, wearing loose-fitting clothes, and using topical agents such as emollients, lotions, and moisturizing creams.11, 13, 14 When extreme, HFS may necessitate cessation of therapy.

Small case series have questioned whether pyridoxine may prevent HFS.7, 8, 10 Pyridoxine (vitamin B6) is a water-soluble vitamin that plays an important role in basic and essential cellular functions. The mechanism of action for the prevention or treatment of HFS with pyridoxine remains unclear but reportedly may be related to antagonism of the P2X purinergic receptor, which accelerates repair of the skin barrier and prevents epithelial hyperplasia.15 In addition, it is unknown whether HFS significantly affects a woman's quality of life (QOL).

Healthcare outcomes research measures what is done and what the intervention actually accomplishes. It is uncertain how PLD effects patient outcomes like HFS and QOL. Patient-related outcomes are tools completed by patients to measure many aspects of the patient's experience and may include assessment of QOL. QOL changes throughout the course of a cancer patient's journey through treatment, remission, recurrence, and at the end of life. Research tools take patient's subjective QOL and attempt to measure it objectively by including physical, functional, social, and emotional domain assessments.16 During adjuvant chemotherapy, physical and functional well being decline.17, 18 However, it is unclear how palliative chemotherapy affects a patient's QOL and which specific domains are altered.

The use of patient-related outcomes is changing the way research is conducted and medicine is practiced. These measures can be divided into 4 main groups: general surveys, disease-specific surveys, symptom-specific surveys, and therapy-specific surveys. Ideally, a treatment-specific survey should be used, for example, the Functional Assessment of Cancer Therapy-Taxane (FACT-Taxane).19 However, there is no such tool for PLD or HFS. Until more treatment-specific tools are developed, outcomes researches still must rely on the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE), which are used by nurses and physicians to grade the side effects of cancer treatment.

The primary objective of the current study was to compare the efficacy of pyridoxine versus placebo for the prevention of HFS in patients with recurrent ovarian cancer, metastatic breast cancer, or endometrial cancer who received PLD (40 mg/m2 every 4 weeks). A secondary objective was to compare the QOL of patients who experienced HFS with that of patients who did not experience HFS during chemotherapy. Finally, in this analysis, we sought to compare QOL domains in patients who had recurrent ovarian cancer with the same domains in patients who primarily received adjuvant chemotherapy and in the normative US population of adult women.

MATERIALS AND METHODS

The study was designed as a double-blind, randomized, controlled trial (Fig. 1). Randomization was stratified by cancer diagnosis (ovarian, breast, or endometrial). All patients were randomized to receive either oral pyridoxine (vitamin B6) 100 mg twice daily or a comparable placebo. The vitamin B6 dose was based on small case reports and series that were available at the time of study initiation in 2004.7, 8, 10 Institutional review board approval was obtained, and all patients signed informed consent before participating in the study.

Figure 1.

This is a chart of the study design schema. PPE indicates palmar-plantar erythrodysesthesia syndrome; HFS, hand-foot syndrome; PLD, pegylated liposomal doxorubicin; IV, intravenous; q, every; twice daily, twice daily; Q-day, daily.

Patients were treated for recurrent ovarian cancer, metastatic breast cancer, or advanced endometrial cancer at a starting chemotherapy dose of 40 mg/m2 every 4 weeks. Eligibility criteria (related to PLD treatment) included a Karnofsky performance status ≥60% and adequate bone marrow, renal, and liver function. Patients were required to have a baseline multigated acquisition scan or a 2-day echocardiogram within normal limits. Ineligible patients included those who had received previous PLD therapy, pregnant or lactating women, and women with a history of hypersensitivity reactions attributed to a conventional formulation of doxorubicin HCL or the components of PLD. Patients with a history of cardiac disease or clinical evidence of congestive heart failure were ineligible.

Patients were required to have discontinued corticosteroid therapy at least 3 weeks before enrollment, and no corticosteroids were allowed for the duration of the trial. Patients who currently were receiving any pyridoxine therapy except for that included in a multivitamin must have discontinued pyridoxine therapy at least 3 weeks before enrollment. Patients who had experienced previous HFS were not specifically excluded from the study.

PLD (Centocor Ortho Biotech Inc., Horsham, Pa) therapy was initiated at 40 mg/m2 intravenously every 4 weeks over 1 hour. Pyridoxine was supplied as 100-mg tablets, and placebo tablets that were indistinguishable from the active drug also were provided. Pyridoxine or matching placebo were given orally the first morning before the PLD infusion and twice daily thereafter. All nursing education was standardized by using a PLD education checklist and literature provided by Centocor Ortho Biotech Inc. Patients were instructed to bring in their medication containers and pill calendar at each visit to check for compliance.

Patients were monitored weekly with telephone calls from the research nurse and monthly at clinic visits for overall (including pyridoxine) and specific PLD-related toxicities using version 3.0 of the NCI CTCAE.20 Adverse events like HFS, hematologic toxicities, and stomatitis were managed by dose modification or dose delays and adjustments. Any patient whose chemotherapy treatment was delayed was evaluated on a weekly basis until adequate hematologic and nonhematologic parameters were met. If delays >2 weeks occurred, then patients were taken off study. Patients were instructed to continue taking pyridoxine/placebo every day despite delays in treatment. HFS was managed by lengthening the dosing interval and dose reduction as noted (Table 1). The treating physician was informed only with regard to which regimen the patient was assigned if there was a serious adverse reaction or if the patient experienced grade 2 or 3 HFS. Patients with grade 2 or 3 HFS were unblinded; and, if the patient was taking placebo, then they were given pyridoxine starting on Day 1 of the next planned therapy. If the patient was taking pyridoxine, then they were removed from the trial and managed at the discretion of the treating physician.

Table 1. Hand-Foot Syndrome Toxicity Grading and Dose Modifications
Toxicity Grade: Hand-Foot SyndromeDose Adjustment
  1. PLD indicates pegylated liposomal doxorubicin.

Grade 1: Mild erythema, swelling, or desquamation that does not interfere with daily activitiesRedose without reduction
Grade 2: Erythema, desquamation, or swelling that interferes with, but not precluding, normal physical activities; small blisters or ulcerations <2 cm in greatest dimensionDelay dosing up to 2 wk or until resolved to grade 0-1; if, after 2 wk, there is no resolution, then PLD should be discontinued; when resolved to grade 0-1, if <2-wk delay, then unblind; if patient is receiving pyridoxine, then remove from trial; if patient is receiving placebo, then initiate pyridoxine on D1 of subsequent dose without reduction
Grade 3: Blistering, ulceration, or swelling that interferes with walking or normal daily activities; cannot wear regular clothingDelay dosing up to 2 wk or until resolved to grade 0-1; if, after 2 wk, there is no resolution, then PLD should be discontinued; when resolved to grade 0-1, if <2-wk delay, then unblind; if patient is receiving pyridoxine, then take off trial; if patient is receiving placebo, then initiate pyridoxine on D1 of subsequent dose, reduce to level 1, and return to original dose interval
Grade 4: Diffuse or local process that causes infectious complications, or a bed-ridden state, or hospitalizationDelay dosing up to 2 wk or until resolved to grade 0-1; if, after 2 wk, there is no resolution, then PLD should be discontinued; when resolved to grade 0-1, if <2-wk delay, then unblind; if patient is receiving pyridoxine, then take off trial; if patient is receiving placebo, then initiate pyridoxine on D1 of subsequent dose, reduce to level 1, and return to original dose interval

QOL was measured with the FACT-G questionnaire after the third course of PLD before the patient was seen by the treating physician and before chemotherapy was administered. The questionnaire was read to the patient by the research nurse in the case of illiterate patients; otherwise, it was completed by the patient herself. The FACT-G version 4 is a 27-item core questionnaire that evaluates the domains of physical, functional, family-social, and emotional well being.16 Questions are answered on a 5-point Likert scale (0 = not at all, 1 = a little bit, 2 = somewhat, 3 = quite a bit, and 4 = very much), and a subscale was computed when ≥50% of the items had been answered. Women with ovarian cancer completed the FACT-Ovarian (FACT-O), an additional 12 questions specific to ovarian cancer. Reliability, validity, and sensitivity to change of the FACT-G have been demonstrated in a variety of settings, and relative scores can be compared with normative data.21 Basen-Engquist et al demonstrated reliability and validity of the FACT-O instrument in ovarian cancer patients with internal consistency reliability for the subscales ranging from 0.74 to 0.92 and with test-retest reliability between 0.72 and 0.88.22

It is estimated that the incidence of HFS for all toxicity grades is 49%.7-11, 23 A decrease ≥50% in the incidence of HFS in patients receiving pyridoxine would be of clinical significance. A sample size of 27 patients per group was chosen, because this would allow us to detect a difference between an HFS incidence of 49% and 11.5% (α = .05; 2-sided power = 0.80). An interim analysis was conducted after 30 patients were enrolled and had evaluable HFS assessment data. Because of decreased accrual and similar HFS rates between groups, the trial was closed for enrollment after the interim analysis was completed.

Data were analyzed according to the treatment group assigned (intent-to-treat) even if patients were noncompliant and stopped taking the assigned intervention. Demographic and treatment history variables (primary cancer treatment, stage of disease, number of previous chemotherapy cycles) were described and compared by using t tests, chi-square tests, or Fisher exact tests between groups. The relative risk (RR) with appropriate 95% confidence intervals (CIs) were used to describe the incidence of HFS between groups. Comparisons also were made between patients who experienced grade 2/3 HFS versus grade 0/1 HFS. QOL data during chemotherapy are described and compared between patients who had grade 0/1 HFS and grade 2/3 HFS. In addition, an exploratory analysis was done comparing QOL data from our cohort of women with recurrent ovarian cancer, QOL data from women with newly diagnosed ovarian cancer on Gynecologic Oncology Group (GOG) protocols 152 and 172 who were receiving intravenous paclitaxel and cisplatin chemotherapy, and QOL data obtained from the normative US population of adult women.18, 24, 25 The breakdown of domains in those who received intraperitoneal chemotherapy on the GOG 172 trial is not available.

RESULTS

Thirty-four patients were enrolled, including 18 women who were randomized to Group A (pyridoxine) and 16 women randomized to Group B (placebo). The mean patient age (±standard deviation) was 64 ± 9.6 years (range, 45-81 years), and 56% of women were aged ≥65 years (Table 2). The majority of patients (85%) were Caucasian. Twenty-five patients (73.5%) were treated for recurrent ovarian/peritoneal cancer, 8 patients (23.5%) were treated for endometrial cancer, and 1 patient was treated for breast cancer; all patients received single-agent PLD therapy. The majority of patients (56%) had received 1 prior chemotherapy regimen, and 44% had received ≥2 prior regimens. No patients in this study had prior HFS. Socioeconomic and performance status were similar between groups. The mean number of PLD chemotherapy cycles administered was 3.5 (range, 1-6 cycles) and did not differ between groups. Mean compliance for taking the assigned intervention (pyridoxine vs placebo) was 92% versus 97%, respectively. The number of chemotherapy cycles administered and compliance did not differ between groups.

Table 2. Patient Demographics Between Groups, N = 34
CharacteristicNo. of PatientsPa
Group A: Pyridoxine, n = 18Group B: Placebo, n = 16Total
  • SD indicates standard deviation; PLD, pegylated liposomal doxorubicin.

  • a

    Student t test or Fisher exact test.

Age: Mean ± SD, y62.9 ± 9.465.9 ± 11.064.3 ± 0.2.401
Cancer type   .999
 Ovarian/peritoneal121325 
 Endometrial448 
 Breast011 
Race   .323
 Caucasian161329 
 African American134 
 Other101 
Mean no. of chemotherapy cycles of PLD administered (range)3.7 (1-6)3.4 (1-6)3.5 (1.6).643
No. of previous chemotherapy regimens   .632
 111819 
 24610 
 ≥3325 
Socioeconomic status   .291
 Private insurance13821 
 Medicare5712 
 Medicaid011 
Karnofsky performance status, %   .704
 90-10011819 
 70-895611 
 60-69224 

Five patients (3 in Group A and 2 in Group B) were not evaluable for HFS assessment (because of reactions to chemotherapy during the first course of chemotherapy). Overall, 52% of patients developed HFS (all grades), including grade 2 or 3 events in 35% (Table 3). No grade 4 events were observed. In Group A, 8 of 15 patients (53%) developed HFS versus 7 of 14 patients (50%) in Group B (RR, 1.07; 95% CI, 0.536-2.16; P = .857). For grade 2/3 events, there was no difference between groups: Six of 15 events (40%) occurred in Group A, and 4 of 14 events (29%) occurred in Group B (RR, 1.40; 95% CI, 0.50-3.94; P = .70). The occurrences of all adverse events by category between groups are listed in Table 4. The most common side effects for all grades in both groups included blood/bone marrow symptoms (overall, 62%), gastrointestinal symptoms (overall, 55%), and constitutional symptoms (overall, 45%). The most common grade 3/4 toxicity was bone marrow suppression (overall, 10%). PLD dose reductions occurred in 2 patients in Group A, including 1 patient who had grade 4 anemia requiring transfusion and 1 patient because of grade 3 stomatitis. In patients who received placebo (Group B), PLD treatment was delayed in 3 patients for grade 2 deep vein thrombosis, grade 3 increased alkaline phosphatase, grade 3 hematologic parameters, and grade 3 HFS. Five patients who were receiving pyridoxine and 5 patients who were receiving placebo had disease progression and died during PLD therapy.

Table 3. Incidence of Hand-Foot Syndrome Between Groups, n = 29
Hand-Foot SyndromeGroup A: Pyridoxine (n = 15)Group B: Placebo (n = 14)Total No. (%)
Grade 1235 (17)
Grade 2336 (21)
Grade 3314 (14)
All grades8715 (52)
Table 4. Incidence of Adverse Events (Excluding Hand-Foot Syndrome) Experienced by Patients Between Groups, n = 29
Adverse EventGroupaNo. of Events (%)
All GradesGrade 3/4
  • a

    Group A, pyridoxine (n = 15); Group B, placebo (n = 14).

Allergy/immunologicA0 (0)0 (0)
 B1 (7)1 (7)
Blood/bone marrowA8 (53)1 (7)
 B10 (71)2 (14)
Constitutional symptomsA7 (47)0 (0)
 B6 (43)1 (7)
GastrointestinalA11 (73)2 (13)
 B5 (36)0 (0)
Hemorrhage/bleedingA1 (7)1 (7)
 B0 (0)0 (0)
Hepatic/pancreasA0 (0)0 (0)
 B2 (14)0 (0)
InfectionA3 (20)1 (7)
 B1 (7)0 (0)
LymphaticsA2 (13)0 (0)
 B2 (14)0 (0)
Metabolic/laboratoryA1 (7)0 (0)
 B2 (14)1 (7)
Musculoskeletal/soft tissueA0 (0)0 (0)
 B2 (14)0 (0)
NeurologyA4 (27)1 (7)
 B1 (7)0 (0)
PainA5 (33)0 (0)
 B2 (14)0 (0)
Pulmonary/upper respiratoryA3 (20)1 (7)
  4 (29)0 (0)
Renal/genitourinaryA2 (13)0 (0)
 B3 (21)0 (0)
VascularA1 (7)1 (7)
 B1 (7)1 (7)

There were no differences in global or domain QOL scores after Cycle 3 either between groups or between the patients with grade 2/3 HFS versus grade 0/1 HFS (Table 5). Overall, functional (score, 18.7) and emotional (score, 18.7) domains were the most compromised during therapy. There was a decline in the physical domain for those with grade 2/3 HFS; however, it was not statistically significant. Less than half of patients (4 of 10 patients; 40%) with grade 2/3 HFS reported being bothered by side effects of treatment, as indicated by scores ≥2 (somewhat or higher) on the FACT-G physical domain question regarding treatment side effects.

Table 5. Domain Quality-of-Life Scores During Chemotherapy With Pegylated Liposomal Doxorubicin
DomainQuality-of-Life Domain Scores: Mean ± SDPa
OverallHand-Foot Syndrome
Grade 0/1Grade 2/3
  • SD indicates standard deviation; HFS, hand-foot syndrome; FACT-G, Functional Assessment of Cancer Therapy-General.

  • a

    Student t test.

FACT-G84.6 ± 9.684.9 ± 10.284.4 ± 9.5.916
Physical23.5 ± 3.925.2 ± 2.921.6 ± .2.057
Functional18.7 ± 4.117.6 ± 5.720.0 ± .4.312
Emotional18.7 ± 4.818.1 ± .319.4 ± 2.5.547
Social23.7 ± 1.924 ± 1.823.4 ± 2.1.522
Ovarian subscale31.3 ± 4.7   

An exploratory line-item analysis was performed on the physical and functional QOL domains to strategize for future intervention trials. With regard to physical difficulties, nausea (14%) was the most common followed equally by pain (7%) and trouble meeting the needs of family (7%). In the functional domain, sleep disruption (7%), enjoyment (7%), and being content (7%) all were compromised equally.

Global QOL scores were higher for patients enrolled on palliative PLD chemotherapy (global QOL score, 84.6) than for patients enrolled on the GOG 152 and GOG 172 studies who received adjuvant intraperitoneal chemotherapy71 and patients who received adjuvant intravenous chemotherapy (global QOL score, 80.9). It is noteworthy that global scores for patients enrolled on palliative PLD were higher than scores for the normative population (global QOL score, 79.6). However, their social domain scores were markedly higher compared with scores from the US population of adult women (Table 6).

Table 6. Quality of-Life Domain and Total Scores in the US Population of Adult Women During Adjuvant Chemotherapy as Reported Before Cycle 4 in Gynecologic Oncology Group Study 152 (GOG 152), GOG 172, and the Current Study
 Quality-of-Life Domain Score: Mean ± SD
DomainUS Population of Adult Women, NormativeGOG 152/GOG 172 IV Arms (n = 361)GOG 172 IP Arm (n = 148)Current Study
  1. SD indicates standard deviation; IV, intravenous; IP, intraperitoneal; FACT-G, Functional Assessment of Cancer Therapy-General.

Physical well being22.1 ± 5.420.1 ± 5.423.5 ± 3.9
Functional well being18.3 ± 6.917.4 ± 5.918.7 ± 4.1
Social well being19.8 ± 6.824.8 ± 3.423.7 ± 1.9
Emotional well being19.4 ± 5.118.6 ± 4.018.7 ± 4.8
FACT-G79.6 ± 18.680.9 ± 14.171.0 ± 14.984.6 ± 9.6

DISCUSSION

The objective of palliative chemotherapy in the setting of recurrent cancer is to improve the quantity of life and QOL for patients. It is imperative to measure patient-related outcomes, such as side effects of therapy, to appropriately evaluate treatment regimens when efficacy results are similar among various chemotherapeutic options. Fifty-two percent of patient's in the current prospective study had HFS, and 35% had grade 2/3 events. There was no difference in the incidence of HFS between patients who received prophylactic pyridoxine and patients who received placebo. Surprisingly, only 40% of patients with grade 2/3 HFS reported being bothered by side effects of PLD treatment, which speaks to the resilience of this patient population. Patients who were receiving pyridoxine or placebo had a similar occurrence of overall adverse events and dose reductions, and pyridoxine did not appear to compromise the efficacy of PLD therapy.

HFS and other forms of rashes are potential side effects in several chemotherapy and biologic therapeutics. The incidence of HFS with PLD that we observed in the current study was high normal compared with the incidence reported in the literature.7-11, 23 Steroids were an exclusion criteria in this study, which may be a factor in our HFS incidence. Some clinicians give pretreatment steroids to prevent chemotherapy anaphylaxis, and this may suppress some treatment-induced drug rashes. Regional cooling methods using ice packs at the wrists and ankles and the consumption of iced liquids during PLD infusion and for 24 hours after chemotherapy also have been suggested as means for reducing the frequency and severity of HFS and are well tolerated.13, 14 However, others consider cooling mechanisms an actual risk factor for HFS.26

The mechanism of action of HFS with certain chemotherapy and biologic therapeutics is unclear. With regard to PLD, histologic patterns include HFS and morbilliform eruptions with atypical squamous proliferations that exhibit epidermal dysmaturation.27 Others postulate that the exact clinical presentation may depend on pre-existing skin diseases.28 The patterns of cutaneous adverse effects of biologics, such as sorafenib, may be mixed. In addition, biologic HFS reactions may be clinically distinct from the well known chemotherapy-induced HFSs.29

Although pyridoxine was not preventative at 100 mg twice daily, it may be beneficial at higher doses or with other biologic or chemotherapeutic drugs.30 Azad et al used combination therapy with sorafenib and bevacizumab in a phase 1 study.31 The incidence of HFS in that study was 76%, and the investigators used pyridoxine in doses up to 400 mg administered twice daily. They subsequently noted increases in HFS with the cumulative use of these agents using maximum doses of pyridoxine of 800 mg daily. Other chemotherapeutic agents that are used to treat ovarian cancer may induce a drug rash, including gemcitabine and capecitabine.32 In addition, there appears to be a higher incidence of cutaneous adverse events with other biologic agents that are used to treat ovarian cancer. These include cetuximab, lenalidomide, and tanomastat.33-35

The HFS rash appeared to be tolerable in most patients, because it did not disrupt patient-rated QOL. However, it is unclear whether the tools used in the current study were adequate to capture the specific problem of rash, because the line item in the physical well being domain relates to overall side effects of treatment and does not specifically address HFS. CTCAE are used by nurses and physicians to grade the side effects of cancer treatment, whereas patient-related outcomes tools like QOL questionnaires are completed by patients. Therefore, CTCAE are not a validated patient-related outcomes; and, in the current study, there was no correlation. The NCI classifies HFS according to 3 grades, whereas the World Health Organization (WHO) uses 4 grades; however, it is doubtful that the WHO definition would have changed the correlation, because the side effect remained.36 Because of the increased prevalence of treatment-induced, cutaneous side effects with the increased use of chemotherapeutic and biologic agents, it is imperative to develop patient-related outcomes simultaneously with the drugs.

The course of QOL during palliative chemotherapy has not been well defined. When comparing global QOL scores, patients on palliative chemotherapy for ovarian cancer in the current study had higher scores than those published in the normal population.25 However, on further analysis, the higher scores were caused by markedly higher social domain scores among patients with ovarian cancer as also reported among patients with ovarian cancer who had newly diagnosed disease.37 Most important, when comparing global QOL scores for patients on palliative chemotherapy with the scores for patients on adjuvant intravenous and intraperitoneal therapy, the scores were compromised most for patients who were receiving intraperitoneal therapy.

There are concerns to be taken into consideration when interpreting findings from a relatively small number of evaluable patients. A limitation of the current study was a lack of racial heterogeneity, which may reduce the generalizability of the results. Another limitation was the inability to assess and compare all adverse events because of the low power of the study. With regard to preventing HFS with pyridoxine, it may still have some efficacy at higher doses. Clinicians may consider adding intravenous steroids irrespective of any other supportive care measure before infusion with PLD to prevent HFS; or, it is possible that pyridoxine will not be effective. However, it may be effective in certain symptoms, such as tingling, and not in others, such as pain. Developing better tools to measure the intricacies of drug rashes will assist in measuring patient outcomes.

With regard to measuring QOL, assessments were missing for those patients who did not complete at least 3 cycles of chemotherapy because they were taken off study for disease progression. The exploratory QOL analysis of patients with recurrent ovarian cancer on palliative chemotherapy was limited by the number of patients; however, more randomized phase 3 clinical trials for recurrent ovarian cancer are being performed and are including QOL and symptom indices. Therefore, as the literature grows, more comparisons will be made to plan interventions to improve QOL in this compromised patient population.

Pyridoxine as administered in this study did not prevent HFS in patients who were receiving PLD. Many patients with ovarian cancer are treated with numerous chemotherapeutics agents over the course of their disease. It is anticipated that these patients will receive more biologic therapeutics over time. In addition, it is expected that there will be more pairings of chemotherapy with biologic agents. It is unclear whether the mechanisms of action of HFS from different agents are related enough to warrant similar prevention and treatment. Therefore, it will be imperative to continue exploring cutaneous adverse events in these agents along with prevention strategies.

CONFLICT OF INTEREST DISCLOSURES

Research funding was provided by Centocor Ortho Biotech Inc. (Horsham, Pa).

Ancillary