Effectiveness of Integrative Medicine in the Management of Lymphoma Survivors: An Exploratory Preference-Based Controlled Trial

Abstract

Introduction

Lymphoma has recovery rates above 60%, but many survivors experience impaired quality-of-life (QoL) requiring survivorship care. This study evaluated the role of an integrative oncology (IO) clinic in managing lymphoma survivors.

Methods

In this exploratory preference-based controlled trial, adults in remission after lymphoma treatment were allocated to two groups: those attending the IO survivorship clinic (intervention) and those declining (control). The intervention included complementary medicine, spiritual, and social support, delivered weekly for up to 6 months in addition to standard follow-up. The primary outcome was QoL improvement (EQ-5D-5L index). Secondary outcomes included symptom relief (MYCAW), cognitive function, and perceived control.

Results

Twenty-nine patients were enrolled: 15 in the intervention and 14 in the control group. Over the first 3 months, a significant time × group interaction in EQ-5D-5L scores favored the intervention (p = 0.005), reflecting superior QoL trajectory. MYCAW concerns also improved significantly in the intervention group across 6 months (p = 0.005 and p = 0.03). At 3 months, FACT Cog-Oth scores were significantly higher in the intervention arm (p = 0.01), indicating better “other” cognitive functions (e.g., memory, clarity, confusion). To note, adherence to IO mainly decreased after 3 months.

Conclusion

In this exploratory preference-based study, an IO survivorship clinic for lymphoma survivors was associated with improvements in QoL, especially for adherent patients. Given the preference-based design, these findings should be interpreted cautiously and viewed as hypothesis-generating rather than confirmatory. Further studies are warranted to evaluate long-term benefits and sustainability of this approach.

Background

Lymphoma is the fifth most prevalent type of cancer and the most frequent hematological malignancy [1]. Due to scientific progress and the introduction of novel therapies and better supportive care in the last decades, its overall survival has increased [2]. Consequently, the population of lymphoma “survivors” is steadily increasing, shifting clinical focus beyond disease control toward comprehensive survivorship care.

Cancer survivorship care encompasses more than the management of residual symptoms or quality-of-life (QoL) impairment. Contemporary survivorship frameworks emphasize a broader scope that includes health optimization, surveillance for relapse and secondary malignancies, prevention and management of late treatment-related toxicities, psychosocial well-being and support for reintegration into social, occupational, and family life [3, 4]. Despite these structured recommendations, many lymphoma survivors continue to experience physical and psychological sequelae that are insufficiently addressed in routine follow-up.

Long-term and late effects among lymphoma survivors are well documented. For example, psychologic effects and even post-traumatic stress disorder-related symptoms have been described among non-Hodgkin lymphoma survivors [5, 6]. In addition, decreased physical and cognitive functioning seems to be associated with negative life changes [7]. Chemotherapy-induced peripheral neuropathy is a frequent symptom in such patients, even years after treatment, and is associated with impaired QoL [8], cognitive impairment [9], and decreased social reintegration [10]. Finally, three follow-up studies of the German Hodgkin Study Group have found cancer-related fatigue as a major and debilitating symptom experienced by Hodgkin lymphoma survivors, independently of the type of treatment they received [10–12]. These unmet needs persist despite adherence to evidence-based survivorship guidelines [3, 4], highlighting gaps between recommended care and patient-experienced outcomes.

Integrative oncology (IO) is a “patient-centered, evidence-informed field of cancer care that utilizes mind and body practices, natural products, and/or lifestyle modifications from different traditions alongside conventional cancer treatments, and aims to optimize health, QoL, and clinical outcomes across the cancer care continuum and to empower people to prevent cancer and become active participants before, during, and beyond cancer treatment” [13]. It seems that lymphoma survivors use the cited modalities at a rate higher than the general population [14], often seeking relief from persistent symptoms and greater perceived control over their health [15], with high satisfaction [16].

Evidence supporting IO interventions is growing, particularly in cancer survivorship populations dominated by solid tumors such as breast and lung cancers. Systematic reviews and randomized trials suggest that modalities including acupuncture, mind-body interventions, physical activity, and selected nutritional approaches may alleviate cancer-related fatigue, pain, chemotherapy-induced peripheral neuropathy, mood disturbances, and other survivorship-related concerns, with a favorable safety profile [17–20]. However, these studies typically evaluate individual symptoms or single modalities and rarely address survivorship as a multidimensional, longitudinal process.

In hematologic malignancies, the existing literature on complementary and integrative medicine (CIM) approaches has focused predominantly on patients undergoing active treatment for lymphoma, leukemia, and myeloma, where IO interventions have been explored mainly for symptom control and supportive care during therapy [21–23]. In contrast, data specifically examining IO approaches in the posttreatment survivorship phase of hematologic cancer (particularly lymphoma) remain scarce. Prior survivorship studies in lymphoma largely emphasize symptom prevalence and long-term burden, while structured evaluations of CIM interventions in this population are limited and primarily observational [14–16]. As a result, there is a lack of prospective clinical data assessing whether an integrative survivorship model can meaningfully enhance patient-reported outcome measures (PROMs) beyond standard hematology follow-up. To address this gap, we conducted an exploratory preference-based controlled trial evaluating the effectiveness and safety of an IO approach for QoL improvement in lymphoma survivors.

Methods

Study Design and Setting

This was an exploratory preference-based comparative effectiveness controlled clinical trial aiming to assess whether an IO approach is more effective than a conventional-only approach in managing symptoms of lymphoma survivors. It was conducted at the Hematology Institute of Bnai Zion Medical Center in Haifa, Israel. Trial methods and results were reported according to the Consolidated Standards of Reporting Trials (CONSORT) 2025 guidelines [24], adapted to non-randomized (preference-based) studies (for all online suppl. data, see https://doi.org/10.1159/000551325).

Study Population

Eligible participants were recruited from the Hematology Institute at Bnai Zion Medical Center in Haifa, Israel, if they (1) were diagnosed with a lymphoproliferative disease, (2) were aged 18 years or older, (3) received chemotherapy, biological treatment, or both for treating the lymphoproliferative disease (no limitation from treatment prospective), (4) were defined in remission by the haemato-oncologist, (5) finished lymphoma treatment for less than 1 year (maintenance therapy was authorized), (6) were able to respond to questionnaires, and (7) signed informed consent. Exclusion criteria included (1) age less than 18 years old (children or adolescents) and (2) finishing lymphoma treatment since more than 1 year.

Group Assignment

Patients chose to participate in one of two study groups. Patients not willing to come regularly to the clinic for IO treatments were assigned to the control arm and received conventional follow-up only, according to NCCN guidelines for cancer survivors [4]. Patients willing to attend the IO clinic were recruited to the IO arm in addition to conventional care as above.

Blinding

Due to the preference assignation (participants deciding themselves their assigned group), the study was not blinded to patients or investigators.

Intervention

All patients were followed by a hemato-oncologist. The frequency and type of visits and exams were determined by NCCN guidelines [4], patients’ symptoms, and physician’s clinical judgment. Patients recruited to the intervention arm received, on top of the defined conventional medicine follow-up, IO intervention including: (1) emotional treatments: supportive, non-psychotherapeutic counseling sessions delivered by trained social workers focusing on emotional distress related to cancer survivorship, coping strategies, adjustment to remission, fear of recurrence, and communication within family and social contexts; (2) spiritual guidance: sessions addressing existential concerns, meaning-making, values, sources of hope, and personal belief provided by trained spiritual care practitioners; (3) CIM: acupuncture, herbal supplements, mind-body, touch and/or movement therapies such as reflexology or shiatsu; or combinations of these. Predefined treatment protocol including the type and frequency of these interventions was adapted to each patient’s symptoms and preferences and defined by the integrative team in coordination with the patient, based on evidence-based data, and included attendance to the IO program at least once weekly and treatment with at least one of the CIM modalities. The duration of the intervention was 6 months from recruitment. Adherence with the different interventions was recorded and objectively defined as (1) full adherence: at least one weekly IO session; (2) partial adherence: less than one weekly IO session; (3) full nonadherence: no IO session in the same period.

Training and Quality Control

CIM practitioners (acupuncturists, naturopaths, mind-body and touch therapists), social workers, and spiritual guiders who administered the intervention had at least 5 years of clinical experience in their discipline among patients with oncological or hemato-oncological conditions. A Data Safety Monitoring Board constituting of five experts controlled the safety and quality of the intervention as well as data collection on a yearly basis. Dropout, withdrawal, treatment adherence, and uncollected data were recorded.

Outcomes

All outcomes were self-reported and included the following:

Primary Endpoint

The primary outcome was the effectiveness of an IO approach on improving QoL of lymphoma survivors. The QoL was described using EQ-5D-5L, which includes 5 dimensions of health (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression), each of them describing 5 levels of problems (“no,” “slight,” “moderate,” “severe,” and “extreme” problems or “unable to”). This questionnaire has been chosen since it is easy to use, validated, and translated into Hebrew, Russian, and Arabic, not disease-specific, and permits calculation of utility scores [25]. Patients filled the questionnaire at recruitment, 1 month, 2 months, 3 months, and 6 months post-recruitment (end-of-treatment in the IO arm). These timepoints were selected to characterize both early trajectories and short-term sustainability of QoL changes during the high-adjustment period following completion of lymphoma therapy. We compared between groups the trajectory of EQ-5D-5L index-based summary score which is a preference-based score applying the preferred model as a scoring algorithm for EQ-5D-5L health state utilities and reflects QoL [26].

Secondary Endpoints

The effect of the intervention on different concerns/symptoms experienced by lymphoma survivors was evaluated as a secondary outcome by the Measure Yourself Concerns and Wellbeing (MYCAW) questionnaire that was filled monthly until 6 months from recruitment (end-of-treatment) due to its individualized, symptom-focused nature and sensitivity to short-term change, making it suitable for monitoring evolving concerns over time. The MYCAW questionnaire was appropriate due to its validity, translation into relevant languages, and appropriateness to CIM studies [27].

The effect of the intervention on cognitive functions was evaluated at recruitment, 3 months and 6 months post-recruitment by the validated Functional Assessment of Cancer Therapy-Cognitive Function (FACT-cog) questionnaire, version 3 [28]. This questionnaire was available in Hebrew, Russian, and Arabic, and specifically evaluated 4 dimensions of cognitive dysfunction in cancer survivors: Perceived Cognitive Impairments (Cog-PCI), Perceived Cognitive Abilities (Cog-PCA), impact on Quality-Of-Life (Cog-QOL), and Others (Cog-Oth), giving each of them a different score. Indeed, as previously noted, impaired cognitive functioning is current among lymphoma survivors and has been associated with negative life changes such as impaired financial situation [7]. The perception of disease control was assessed as well by filling out the 4-questions-Perceived Personal Control (PPC) scale [15] once a month in both arms, as previous studies showed the association of IO use with a better PPC which is related to more positive mental functioning among lymphoma survivors [15].

Safety and Adverse Events

A checklist with acupuncture adverse events based on the AcupAE questionnaire [29] was used to evaluate acupuncture-associated safety events after each treatment. For non-acupuncture CIM therapies, the practitioner directly questioned and evaluated the patient after each intervention for possible safety events. Severe adverse events requiring IO treatment cessation included anaphylactic shock, pneumothorax, massive bleeding, or infection in the acupoint area requiring systemic antibiotic treatment.

Criteria for Discontinuation

Participants were discontinued from the study if they voluntarily withdraw informed consent, for safety reasons or due to significant noncompliance with the study protocol as judged by the principal investigator. Reasons for discontinuation were recorded, and patients withdrawn from the study were included in the intention-to-treat analysis.

Sample Size Calculation and Power Analysis

The trial was planned as a two-arm parallel study with ED-5Q-5L as a primary outcome. Power was estimated for the time × group interaction using a repeated-measures framework (linear mixed-effects). Since no prior study evaluated similar outcomes in lymphoma survivors and given the exploratory nature of the current study, we considered a medium effect size to be appropriate. Assuming a two-sided α = 0.05, within-person correlation ρ = 0.50, and no violation of sphericity, a medium standardized interaction effect (Cohen’s f = 0.25) required a minimum sample size of 28 patients (14 per arm) to achieve 80% power.

Statistical Methods

Data analysis was performed using IBM SPSS Statistics software, version 22. Demographic and clinical data were analyzed at baseline to measure the balance among the study groups. Quantitative variables were described using mean and standard deviation or median with interquartile range depending on their distribution. Qualitative variables were described using frequency and percentage distributions. For comparing normally distributed variables between our study groups, we used the t test for independent samples. For comparing quantitative variables that do not distribute normally between our study groups, we used the Mann-Whitney test. For comparing qualitative variables between our study groups, we used the independent chi-square test and Fisher’s exact test. Comparison of continuous variables between the study groups was performed using repeated-measures ANOVA or linear mixed models for missing data. Sensitivity analyses restricted to early follow-up (T0–T3) and to per-protocol participants were performed and compared to intention-to-treat analysis. All comparisons were two-sided with significance level set at p < 0.05.

Results

Baseline Characteristics

A total of 42 patients were screened for eligibility. Twenty-nine of them (69%) met eligibility criteria and were recruited between June 2023 and October 2024. Among them, 15 patients were willing to come to the hematology unit at least weekly for IO intervention, and the remaining 14 were assigned to the control group. Eleven (73%) of the patients assigned to IO arm completed 6 months of treatment per-protocol. Reasons for withdrawal included relapse in 1 (25%) patient at 2 months of follow-up and incompliance in the remaining 3 (75%) patients that stopped treatment between 2 and 3 months post-recruitment (Fig. 1). Mean age was 59 ± 15 years, and 16 (55%) patients were women. Baseline demographic and clinical characteristics were similar in patients from the two study arms, except for a higher rate of previous CIM use in the IO (93%) as compared to the control (43%) group (p = 0.005) as shown in Table 1. Patients’ comorbidities are presented in online supplementary Table 1.

Fig. 1.

CONSORT 2025 flow diagram and study scheme.

Table 1.

Baseline characteristics

Characteristics	IO	Control	p value
	N = 15	N = 14
Agea	59.5 (12.3)	58.5 (17.1)	0.86
Gender (female), n (%)	9 (60)	7 (50)	0.72
Origin, n (%)
Israeli Jewish	5 (33)	6 (43)	​
Ex-USSR	7 (47)	3 (21)	0.44
Arab	3 (20)	5 (33)	​
Family status (married), n (%)	12 (80)	10 (71)	0.48
Occupation, n (%)
Independent/employed	6 (40)	7 (50)	​
Unemployed/housewife	3 (20)	4 (29)	0.51
Retired	6 (40)	3 (21)	​
ECOG, n (%)
0	6 (40)	5 (36)	​
1	8 (53)	6 (43)	0.51
2	1 (6.7)	3 (21)	​
Lymphoma type, n (%)
Aggressive	8 (53)	6 (43)	0.76
Indolent	7 (47)	8 (57)	​
Time from diagnosis (mo)a	23.7 (24.0)	26.7 (28.3)	0.76
Time from remission (mo)a	6.5 (8.3)	4.7 (8.9)	0.59
Number of comorbiditiesa	2.3 (1.8)	2.1 (1.9)	0.86
Lines of therapya	1.3 (0.7)	1.9 (1.3)	0.20
Types of prior therapy, n (%)
Chemotherapy	14 (93)	14 (100)	1.00
Targeted therapy	15 (100)	13 (93)	0.48
ASCT	2 (13)	1 (7.1)	1.00
CART	1 (6.7)	3 (10)	0.33
Duration of therapy (mo)a
Chemotherapy	5.3 (2.0)	6.5 (2.3)	0.16
Targeted therapy	6.6 (2.6)	7.2 (5.5)	0.70
Time from EOT (years)a
Chemotherapy	1.4 (1.5)	1.4 (1.7)	0.96
Targeted therapy	0.7 (0.7)	1.1 (1.6)	0.42
ASCT	1.2 (0.4)	1.0	0.62
CART	0.5	1.4 (1.4)	0.67
Previous use of CIM, n (%)	14 (93)	6 (43)	0.005
Baseline questionnairesa
EQ-5D index score	0.53 (0.39)	0.41 (0.42)	0.59
MYCAW-1	5.1 (0.7)	4.5 (1.8)	0.35
MYCAW-2	5.1 (0.7)	4.6 (1.1)	0.11
FACT-cog total	77.7 (30.3)	80.5 (37.1)	0.82
PPC	9.1 (4.7)	7.4 (5.1)	0.36

ASCT, autologous stem cell transplant; CIM, complementary and integrative medicine; EOT, end-of-treatment; mo, months; MYCAW, Measure Yourself Concerns and Wellbeing; PPC, Perceived Personal Control; USSR, Union of Soviet Socialist Republics.

^aMean (standard deviation).

Quality-of-Life

Linear mixed-effects models were estimated using restricted maximum likelihood with a compound symmetry covariance structure for within-participant repeated measures, using all available observations under a missing-at-random assumption (full-information estimation). Across the T0–T3 period (first 3 months after recruitment), a linear mixed-effects model with random intercepts adjusting simultaneously for sociodemographic and clinical covariates showed a time × group interaction of EQ-5D-5L index score (p = 0.005) favoring IO compared to control arm. In the all-timepoints model (T0–T6), the time × group interaction was not significant (p = 0.281). Collectively, these models indicate a consistent overall advantage of the IO arm in QoL (EQ-5D-5L index score), with early-period divergence (T0–T3) but no evidence of differing trajectories across the full 6-month follow-up (Fig. 2a). Residual diagnostics did not suggest major departures from model assumptions. Outcomes are presented in online supplementary Table 2.

Fig. 2.

Evolution of quality-of-life (EQ-5D) and patients’ concerns (MYCAW) overtime in both study arms. a EQ-5D-5L. b First concern (MYCAW-1). c Second concern (MYCAW-2).

To note, the first 3 months of study were characterized by higher compliance for IO treatments in the intervention group which decreased afterward (Fig. 3); therefore, the significant interaction observed in the T0–T3 period coincides with adherence trends. Missing outcome data arose primarily from declining adherence and study withdrawal. Analyses using linear mixed-effects models assume that data are missing at random, conditional on observed variables. While this assumption cannot be formally verified, baseline characteristics and baseline PROMs were similar between participants with complete and incomplete follow-up. To assess robustness, sensitivity analyses restricted to early follow-up (T0–T3), when adherence was highest, and to per-protocol participants were performed and yielded results consistent in direction with the primary intention-to-treat analysis. Patients’ adherence per CIM modality is presented in online supplementary Table 3. When sub-analyzing the different parts of EQ-5D-5L questionnaire, we could see that although no significant differences were noted at baseline, the mobility, self-care, pain/discomfort, and anxiety/depression scores were lower in the IO arm as compared to control starting at 2 months post-recruitment (2.1 ± 0.9 vs. 3.1 ± 0.7, p = 0.005; 1.5 ± 0.8 vs. 2.5 ± 1.3, p = 0.03; 2.5 ± 0.7 vs. 3.4 ± 1.3, p = 0.04; 1.9 ± 0.8 vs. 3.1 ± 1.3, p = 0.009 respectively), reflecting a better QoL in these domains in patients treated with IO.

Fig. 3.

Patients’ monthly compliance in IO arm.

Symptomatic Improvement

According to MYCAW questionnaire, main patients’ concerns included weakness/fatigue/insomnia in 16 (30%) of cases, musculoskeletal pain in 10 (18%) of cases, neuropathy in 9 (17%), and neurocognitive concerns in 5 (9%) of them (Fig. 4). We analyzed MYCAW ratings overtime using a linear mixed-effects model with fixed effects for group, categorical time and their interaction, and a random intercept for participants. The first concern rated 0–6 by patients (MYCAW-1) showed high baseline scores with no group difference (5.1 ± 0.7 vs. 4.5 ± 1.8, p = 0.35) but had differential trajectories (significant time × group interaction, p = 0.005) during the study time: improving by −2.41 points (95% CI: −3.44 to −1.38; p < 0.001) in IO arm vs. −0.33 points (95% CI −1.79 to 1.13; p = 0.66) in control group from recruitment to 6 months with a between-group difference in change of −2.08 points (95% CI −3.87 to −0.29; p = 0.02), favoring the IO arm (Fig. 2b). Similar results were observed for the second concern (MYCAW-2) that showed high baseline scores with no group difference (5.1 ± 0.7 vs. 4.6 ± 1.1, p = 0.11) but significant time × group interaction (p = 0.03), indicating that longitudinal trajectories differed between arms consistent with a greater and earlier improvement in the IO arm (Fig. 2c). To note, statistically significant differences between study arms of both MYCAW-1 and MYCAW-2 started at 3 months post-recruitment (2.9 ± 1.6 vs. 4.5 ± 1.5, p = 0.01 for MYCAW-1; 2.7 ± 1.1 vs. 4.9 ± 1.6, p = 0.001 for MYCAW-2).

Fig. 4.

Patients’ concerns according to MYCAW questionnaire.

Cognitive Function

No significant difference was noted in the FACT-cog total score at baseline, at 3 months, or at 6 months. However, we noted that the FACT Cog-Oth score was significantly higher than control at 3 months of evaluation (12.9 ± 5.5 vs. 7.1 ± 4.5, p = 0.01), reflecting an improvement in “other” cognitive functions (e.g., remembering information, speaking or thinking clearly, being confused) in the IO arm after 3 months of treatment, while no difference was noted at the “6 months” point (Fig. 5a).

Fig. 5.

Evolution of different scales overtime in both study arms. a Cognitive function (FACT-cog scale). b PPC scale. IO, integrative oncology; Oth, Other; PCA, Perceived Cognitive Abilities; PCI, Perceived Cognitive Impairment; PPC, Perceived Personal Control; QoL, quality-of-life; T0, screening; T1, 1 month; T2, 2 months; T3, 3 months; T6, 6 months.

Perception of Disease Control

Although no between-group differences in PPC domains were statistically significant at baseline, by 2 months post-recruitment, patients in the IO group reported higher (better) scores for control over emotional responses to cancer compared with the control arm (2.6 ± 1.0 vs. 1.7 ± 0.9, p = 0.04), suggesting a potential positive effect of IO on emotional regulation (Fig. 5b).

Safety

No safety events were reported throughout the study course.

Discussion

In this preference-based exploratory study of lymphoma survivors within the first year of remission, adding an IO survivorship clinic to standard hematology follow-up appeared safe and beneficial for key PROMs. IO may contribute to improvement in QoL, patient concerns, and appears associated with some dimensions of cognitive function and perceived disease control, mainly during the first 2–3 months, coinciding with higher adherence, while outcomes between 3 and 6 months generally resembled the control group.

Among the 15 patients in the IO arm, 11 (73%) completed treatment, one relapsed at 2 months, and three withdrew between 2 and 3 months. Adherence declined after 3 months, limiting reliable follow-up beyond this period. Prior studies of patients with cancer on chemotherapy similarly found that two-thirds maintained adherence, while lower adherence was associated with lack of prior CIM use [30]. Similarly in our cohort, patients choosing to come to the IO group had more CIM exposure than the control group before enrollment (93% vs. 43%). However, this point does not explain the decreased adherence in the IO group after 2–3 months of treatment. Declining compliance may reflect symptom and QoL improvements (or lack of effect) reducing perceived need for weekly visits, return to normal life roles (e.g., work, family), or the requirement for survivors – unlike patients on active therapy [30] – to come specifically to the survivorship clinic. These issues warrant qualitative studies evaluating the reasons for adherence or nonadherence to IO programs among lymphoma survivors that may allow to adapt IO interventions to this specific population. In addition, it might be that the program of once weekly was highly intensive and the exact timetable intervention required further adaptation.

Before the intervention, main concerns were generally rated as severe and included weakness, fatigue, insomnia, musculoskeletal and neuropathic pain, and neurocognitive difficulties, consistent with findings in solid tumor survivors [31]. Similar concerns have also been reported during lymphoma treatment [32], but although our small sample size limits broad interpretation, it appears that lymphoma survivors experience greater neuropathy and musculoskeletal pain and fewer gastrointestinal complaints, compared to those described during active treatment [32]. Therefore and given that most IO lymphoma studies to date have focused on patients undergoing active therapy [22], further research should specifically address IO interventions in lymphoma survivorship focusing on special needs in this population.

QoL improved by EQ-5D-5L during the first 3 months, especially from month 2, a period characterized by higher engagement with the IO program, before plateauing to the control trajectory at 3–6 months. While this temporal overlap suggests a possible association between intensity of participation and PROMs, the study design does not allow attribution of greater benefit to adherence itself. Expectancy and therapeutic alliance may have contributed. As seen in patients under active treatment [33], IO effectiveness seems greater in adherent patients, although the current study does not prove it for sure. Encouraging symptomatic survivors to remain engaged may accelerate recovery and reduce chronic issues. However, any relationship between engagement with IO services and observed improvements should be regarded as descriptive and hypothesis-generating. Larger studies should investigate the relationship between treatment adherence and the effectiveness of IO, while qualitative research should explore the reasons underlying reduced compliance with IO among some lymphoma survivors.

Patients’ concerns also seemed to improve in the IO group: specific EQ-5D-5L domains (mobility, self-care, pain/discomfort, anxiety/depression) and MYCAW concerns (weakness/fatigue/insomnia, musculoskeletal/neuropathic pain, neurocognitive issues) decreased significantly in the whole study time. Furthermore, when applying the threshold of a 2-point reduction as clinically meaningful [27], symptomatic benefit was already evident from the first month of IO treatment. These findings parallel IO benefits during active lymphoma therapy [22].

FACT-cog overall trajectories did not differ, but “other” cognitive domains (memory, clarity, confusion) appeared to improve at 3 months in favor of IO. “Chemo-brain” is common in lymphoma survivors, as shown in a recent longitudinal cohort study [34]. While this has long been an unmet need, some studies have suggested the effectiveness of selected CIM modalities [35]. Our preliminary findings support these observations although this point should be interpreted cautiously due to multiple testing, small sample size, and lack of sustained effect.

Perceived disease control regarding emotional responses also seemed to improve at 2 months. This point should be interpreted cautiously considering a single timepoint difference does not support sustained benefit. Prior work with the same instrument (PPC) linked CIM use to better mental health-related QoL through enhanced perceived control [15]. As perceived control fosters resilience [36], well-being, and psychological adjustment in patients with cancer [37], confirmation of these results would justify encouraging IO for lymphoma survivors.

The IO intervention was intentionally multimodal and individualized, precluding attribution of observed effects to specific components. Potential mechanisms likely differ across modalities and may include symptom-targeted effects (e.g., acupuncture for pain or neuropathy), stress- and arousal-modulating effects of mind-body and supportive interventions, and broader psychosocial mechanisms such as enhanced self-efficacy, coping, and emotional regulation. In addition, increased clinical attention and therapeutic alliance within a structured survivorship framework may have contributed to perceived benefit. The present study was not designed to disentangle modality-specific or mechanistic effects; therefore, findings should be interpreted as reflecting the impact of an integrative care model rather than individual therapies. Future studies should incorporate component analyses, dose-response assessments, and mechanistic endpoints to clarify which elements are most relevant for lymphoma survivorship.

No adverse events of IO were observed, consistent with CIM reports in patients with lymphoma [22]. Alongside primary evidence of effectiveness, this supports the safety of these modalities in lymphoma survivors. Moreover, the IO lymphoma survivorship clinic was based in the Hematology Institute of the Medical Center and treatments were regulated and administered by experienced CIM practitioners, which probably contributed to safety [38]. This point calls for broader implementation of IO survivorship clinics in public healthcare systems, including coverage by national health insurances, to make these services accessible to patients from lower socioeconomic backgrounds or remote areas.

This study has several important limitations. Most notably, the preference-based design limits causal inference: patients who chose to participate in the IO survivorship clinic likely differed systematically from those who declined, introducing selection bias. The marked imbalance in prior CIM use (93% vs. 43%) suggests potential confounding by motivation, health beliefs, and expectancy, which may have influenced PROMs independently of the intervention. Regression to the mean, placebo and expectancy effects, and increased clinical attention within the IO setting may therefore partially account for early improvements. Differential adherence represents an additional source of bias. Declining participation after approximately 3 months likely attenuated later between-group differences and complicates interpretation of sustained effects. Although sensitivity analyses supported the robustness of early findings, residual confounding cannot be excluded. Other limitations include modest sample size, reliance on self-reported outcomes without objective endpoints, and a single-center design limiting generalizability. Indeed, the study was conducted in a single Israeli tertiary center with an established integrative medicine service, included a high proportion of prior CIM users, and focused on patients in early remission limiting external validity. Results may not extrapolate to other healthcare systems, populations with lower CIM familiarity, or long-term lymphoma survivors. Accordingly, findings should be interpreted as exploratory and hypothesis-generating, supporting signal detection rather than definitive evidence of efficacy.

Future Directions

Larger, multicenter studies should incorporate randomization or analytic bias mitigation, adherence supports beyond 3 months, and longer follow-up. Adding objective endpoints (activity, neurocognitive testing), patient-centered measures (fatigue interference, social participation, return-to-work), and economic evaluations (EQ-5D cost-utility, service use) will clarify downstream impact and scalability. Component and moderator analyses (e.g., acupuncture vs. mind-body; neurotoxic exposure, subtype, psychosocial context) may guide personalized IO in hematology survivorship.

Conclusions

In this exploratory preference-based study, an IO survivorship clinic for lymphoma survivors was safely implemented within a hematology outpatient setting and was associated with early, meaningful improvements in QoL. The waning of effects with reduced adherence underscores the importance of structured, intensive early IO and strategies to sustain participation.

Statement of Ethics

The study protocol was reviewed and approved by the Bnai Zion Medical Center Institutional Review Board in accordance with the Helsinki Declaration (0050-23-BNZ), and it was prospectively registered at www.clinicaltrials.org (NCT05982223) on 16 July 2023. Written informed consent was required for patients participating in the study.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

No funding was received for completing the study.

Author Contributions

Ilana Levy Yurkovski, Samuel Attias, Elad Schiff, Sally Awad, and Tamar Tadmor designed study protocol. Ohad Cohen-Naznin, Esther Anahory, Yael Gross-Geva, Anat Oved, Niki Bleiweiss, and Sharon Caspi-Biran provided intervention (integrative oncology treatments). Ilana Levy Yurkovski first drafted the manuscript. All authors reviewed the manuscript.

Funding Statement

No funding was received for completing the study.

Data Availability Statement

The data that support the findings of this study are subject to controlled access due to participant privacy. Restrictions apply to the availability of the data and so are not publicly available but are available from corresponding author on reasonable request.