Liver transplantation (LT) for non-resectable colorectal liver metastases (nCRLM) has evolved from an experimental approach to a potentially curative treatment in a highly selected group of patients with liver-limited disease [1]. The pioneering SECA study from Oslo University Hospital demonstrated long-term overall survival (OS) in patients transplanted beyond traditional resection criteria, thereby establishing LT as a viable oncological surgical strategy [2, 3]. Comparable outcomes have subsequently been reported from other centers, including the Rochester protocol in the United States [4], demonstrating the reproducibility of this concept in independent cohorts.

Accurate patient selection remains critical, as disease recurrence occurs in most cases. Conventional morphologic and biochemical markers alone provide only an indirect approximation of tumor biology or behaviour, as they rely on surrogate measures such as carcinoembryonic antigen levels, tumor growth rate, response to chemotherapy, and the presence or absence of extrahepatic disease.

[¹⁸F]FDG PET/CT provides quantitative information on tumor glucose metabolism, which may reflect tumor aggressiveness in addition to providing the well-established volumetric parameter, metabolic tumor volume (MTV), for quantifying metabolic burden and predicting prognosis in colorectal cancer [5]. In a seminal study, Grut et al. [6] demonstrated that pre-transplant MTV was a strong predictor of survival after LT in nCRLM. These findings were subsequently confirmed in a larger national cohort [7] and validated externally by Wehrle et al. [8], thereby establishing MTV as a reproducible prognostic biomarker in transplant oncology [9].

Recently, the TRANSMET trial [10] demonstrated that patients randomized to LT plus chemotherapy against chemotherapy alone obtained a substantial transplant benefit thereby providing level 1 evidence for LT in nCRLM. In a recent multicentre study, Vitale et al. identified female sex as an independent predictor of inferior survival after LT for nCRLM, despite otherwise comparable clinicopathologic features [11]. MTV was not part of the risk stratification in this study.

The aim of this study was therefore to explore potential sex-related differences in the prognostic value of pre-transplant MTV measured by [¹⁸F]FDG PET/CT.

This study confirms that [¹⁸F] FDG PET/CT-derived metabolic tumor volume is a robust predictor of survival following LT for nCRLM and demonstrates a significant interaction between sex and MTV, indicating that the adverse prognostic impact of high MTV (>70 cm³) differs between males and females. Female recipients appear to have a lower tolerance for high metabolic tumor burden, with more frequent non-pulmonary recurrences, resulting in poorer outcomes despite otherwise similar baseline characteristics.

The prognostic importance of MTV has been consistently demonstrated in prior work from our group [8, 9] and subsequently validated internationally [10]. The present findings expand this evidence by formally demonstrating a sex-dependent interaction between MTV and survival in multivariable analyses. In a recent multicentre cohort, not including PET-derived MTV data, Vitale et al. reported that female sex was independently associated with poorer outcomes after LT [11].

Sexual dimorphism in cancer is based on the observation that differences in the incidence and mortality of non sex-related cancers cannot be explained solely by sociobehavioral (gender related) factors, but may instead be related to biological differences, including sex chromosomes [12]. Variations in health-seeking behavior, lifestyle factors, and environmental exposures may contribute to these disparities, but are insufficient to explain the full extent of sex-related differences observed in cancer epidemiology [13]. For most cancers unrelated to reproductive function, both incidence and prevalence tend to be modestly higher in males than in females, with the exceptions of thyroid and gallbladder cancer. Similarly, cancer-related mortality is generally higher in males than in females [13, 14].

Sex hormones may play a role in the initiation and progression of CRC, with age-dependent effects. While premenopausal females generally exhibit superior survival compared with age-matched males, post-menopausal females often experience poorer outcomes, potentially due to diagnosis at more advanced stages or intrinsically more aggressive tumor characteristics, such as right-sided tumor location and BRAF mutations [11, 15]. Sex-specific patterns of adipose distribution and lipid metabolism may further influence cancer risk. Visceral fat-associated inflammation is a recognized driver of carcinogenesis and may contribute to the increased incidence of several cancers after menopause, when females tend to shift towards greater visceral adiposity [13].

The female immune system appears to have a higher activation threshold for T-cell mediated immune responses, partly driven by estrogen, potentially reflecting immunological adaptations related to pregnancy [16]. Sex-related differences regarding immune system responses have been shown to influence disease onset, symptom presentation, patterns of progression and treatment outcomes across a range of conditions, including autoimmune disorders, infectious diseases, cancers and vaccine efficacy [17]. Also, the liver is an estrogen-sensitive organ [18], and experimental data suggest that estrogen may modulate the microenvironment of liver metastases in a sex-specific manner [19]. However, these proposed mechanisms remain speculative and are not directly supported by data in the present study.

Interestingly, population studies have shown that females are about 25% less likely to undergo surgery for liver metastases, and in this cohort, the all-cause mortality is significantly higher in females compared to males. Since the LT patients all had nCRLM, this may indicate that the gender related effects on outcome are related to biological differences in tumor biological behavior [20].

Given a limited availability of liver grafts, optimizing patient selection is of clear clinical relevance. In the present study, males more frequently presented with pulmonary metastases (two-thirds of recurrences), which are known to have relatively indolent course and are often amenable to pulmonary resection with prolonged SAR [21, 22]. In contrast, approximately two-thirds of females experienced non-pulmonary recurrence, a pattern associated with poorer survival outcomes [22].

The current analysis shows that males with high MTV had an acceptable median OS of 59 months. Thus, males who otherwise demonstrate favorable characteristics during pre-transplant evaluation may still be considered for inclusion despite high MTV (>70 cm³), although this should be interpreted with caution given the limited sample size. The ROC analysis identified optimal MTV cut-off values for predicting 5-year OS of 76 cm³ in females and 66 cm³ in males. Accordingly, the currently used, non-sex-specific cut-off value of 70 cm³ appears appropriate, with no clear indication at present for separate MTV thresholds for males and females.

Several limitations merit consideration. This was a retrospective study with a limited sample size, reflecting the rarity of LT for nCRLM; consequently, the statistical power for sex-stratified analyses should be regarded as modest. Exact information on pre- and postmenopausal status among female patients was unavailable. However, given the median age of the female patients was 58 years and that all patients received chemotherapy, it is likely that a majority were postmenopausal at the time of LT. PET/CT acquisition protocols and scanner technology evolved during the inclusion period, which may have introduced variability in SUV quantification, although this is unlikely to have substantially affected MTV measurements. The 70 cm³ MTV threshold is empirically derived and should not be considered as an absolute. Minor variations in preoperative chemotherapy and immunosuppressive regimens may also have influenced outcomes. Finally, as with all observational studies, unmeasured confounding factors cannot be excluded. Nonetheless, the use of uniform national selection criteria and long-term follow-up adds robustness to the observed associations.

Introducing different inclusion and exclusion criteria based on sex would be premature and is not the intention of this study. Some recent papers, however, indicate that relevant gender specific differences are clinically relevant [20] and the findings from the current study underscore that biological sex should be considered as one of several factors when planning and individualizing treatment strategies aimed at achieving the best possible outcomes.

In summary, pre-transplant [¹⁸F]FDG PET/CT-derived MTV remains a robust prognostic biomarker after LT for nCRLM. Female patients exhibit poorer OS and SAR at high MTV levels. These findings suggest that sex may modify the prognostic impact of MTV but should be considered hypothesis-generating and require validation in larger cohorts. The underlying mechanisms remain to be elucidated.

Female patients had shorter OS than males, including when the analysis was restricted to patients with high MTV. This may be explained by the post-transplant recurrence patterns, with females developing more unfavourable non-pulmonary recurrences. These findings are hypothesis-generating and should be interpreted with caution and require validation in larger cohorts before clinical implications can be drawn.