Surgery Beats Liver Metastases as Effectively in Patients Over 80, Landmark Japanese Study Finds — So Why Do They Still Die Sooner?

When colorectal cancer spreads to the liver, the operating table is often the last place where a cure remains on offer. For patients in their eighties, surgeons have long hesitated to play that final card, fearing that frailty, competing illnesses and slow recovery would erase whatever benefit the operation could deliver. A sweeping new study from Japan — the largest investigation of its kind ever assembled — now delivers a strikingly two-sided verdict. Among carefully selected octogenarians, liver resection kept colorectal liver metastases at bay just as effectively as it did in younger patients. Yet the elderly still died of their disease at nearly twice the rate, and the cause was not the operation itself. It was what failed to happen afterward: when the cancer returned, patients aged 80 and older were dramatically less likely to receive any treatment at all.

Colorectal cancer is one of the leading causes of cancer death worldwide, and its most feared complication — metastasis to the liver — is precisely the scenario in which surgery has rewritten the rules over the past two to three decades. For resectable colorectal liver metastases, hepatectomy is now regarded as the single most effective treatment available, its results steadily improved by better patient selection, refined operative technique and an expanding arsenal of systemic chemotherapy regimens. Recurrence, however, remains the field’s stubborn adversary: tumors reappear in the remnant liver or elsewhere in the body even after apparently curative resection, and long-term survival often hinges on whether those recurrences are attacked aggressively with repeat surgery, chemotherapy, thermal ablation or radiation. How much of that salvage arsenal to deploy in the very old has never been firmly established, and the dilemma is sharpened by evidence that frail patients face higher operative risks while many are judged ineligible for curative surgery in the first place.

The demographic backdrop lends the question urgency. Japan has already become the world’s first super-aged society, and average life expectancy there reached 81.05 years for men and 87.09 years for women in 2022 — the longest longevity on record anywhere. Across much of the globe, life expectancy now exceeds 80 years, so the number of elderly patients carrying colorectal liver metastases is projected to climb steadily in the decades ahead. Yet frailty is common in this population, postoperative morbidity and mortality are likely to be elevated, and multidisciplinary treatment is often deliberately scaled back. Whether the operation genuinely pays off in octogenarians has been difficult to determine, because elderly and nonelderly patients differ systematically in both their backgrounds and their tumors — a confounding problem that earlier studies, most including fewer than 100 patients aged 80 or older, have struggled to disentangle.

To break that deadlock, a team led by Kiichi Sugimoto of the Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital mined the database of the Joint Committee of the Liver Metastases Survey of Colorectal Cancer, a nationwide registry spanning 145 Japanese centers with prospectively collected data on patients diagnosed between 2013 and 2017 and survival information updated in 2018 and again in 2022. From 7,828 registered patients with known survival status, the investigators excluded 3,476 who did not undergo liver resection, 518 with simultaneous extrahepatic metastasis, 69 whose disease was judged incurable at surgery and 1,356 with missing records. The final cohort comprised 2,409 patients: 2,194, or 91.1%, under the age of 80, and 215, or 8.9%, aged 80 or above — by a wide margin the largest group of operated octogenarians with colorectal liver metastases ever analyzed. The study was conducted under the Declaration of Helsinki with approval from the institutional review board at Komagome Hospital.

The two groups were strikingly unequal before any statistical correction. Elderly patients who reached the operating room carried a visibly lighter tumor burden: only 16.3% had cancer in both lobes of the liver, versus 34.0% of the young; merely 9.8% harbored four or more metastases, versus 21.3%; and metastases were less often synchronous — detected at the same time as the primary bowel tumor — and more often metachronous, appearing months or years later. Octogenarians were also far less likely to receive chemotherapy before surgery, 10.2% versus 21.1%, or after it, 19.5% versus 54.8%, and were more likely to score 3 or higher on the American Society of Anesthesiologists physical status scale, which flags severe systemic disease. Operative quality, meanwhile, was reassuringly similar between the ages: complete R0 resections, with no residual tumor left behind, were achieved in 93.0% of the elderly versus 94.9% of the young, and severe postoperative complications of Clavien–Dindo grade 3 or above occurred at statistically indistinguishable rates.

Because survival comparisons are meaningless without balance, the team turned to propensity score matching, a statistical technique designed to mimic the equality of a randomized trial using observational data. The method builds a logistic model that estimates each patient’s probability — the propensity score — of belonging to the elderly group, calculated from every clinicopathological factor that differed significantly between the groups: ASA classification, emergence time of the metastases, their distribution across the liver, their number, and pre- and postoperative chemotherapy. Patients are then paired one-to-one across groups with near-identical scores. Here the model achieved a c-statistic of 0.73, indicating acceptable discrimination; a caliper width of less than 0.01 defined acceptable matches; and standardized differences below 0.10 confirmed that all covariates were balanced in the matched sample. The procedure yielded 212 well-matched elderly–younger pairs, with all computations performed in JMP Pro 18 and survival differences quantified by Kaplan–Meier analysis, log-rank tests and Cox proportional-hazards regression.

The investigators then tracked three endpoints, each capturing a different dimension of success. Recurrence-free survival measured the interval from liver surgery to the first detectable return of cancer; cancer-specific survival counted only deaths attributable to the malignancy; and overall survival registered death from any cause. In the unmatched cohort, recurrence-free survival was statistically indistinguishable — five-year rates of 35.9% in the young versus 33.6% in the old — but cancer-specific survival, 68.2% versus 53.1%, and overall survival, 64.4% versus 46.9%, were markedly worse in the elderly. Propensity matching sharpened the picture rather than dissolving it. Matched octogenarians still controlled their disease comparably, with five-year recurrence-free survival of 33.8% versus 40.9% and a hazard ratio of 1.20 that missed statistical significance. Yet cancer-specific survival remained significantly inferior, 53.4% versus 70.5% (hazard ratio 1.77), as did overall survival, 47.3% versus 66.5% (hazard ratio 1.80). Something beyond surgical skill, tumor biology or operative risk was shortening lives.

That something surfaced when the researchers asked what happened after recurrence struck. Among 1,329 younger patients whose cancer returned, 86.4% received treatment — repeat surgery, chemotherapy, radiofrequency or microwave ablation, radiation, or combinations thereof. Among the 130 octogenarians with recurrence, only 61.5% did, a gap the authors describe as the study’s most consequential finding. It widened with the disease’s reach. When relapse remained confined to the remnant liver, roughly nine in ten younger patients were treated, versus 79.1% of the elderly; for extrahepatic recurrence the figures fell to about 85 to 90% versus 55.8%; and when cancer appeared at both sites, barely half of the old — 50.0% — received any therapy at all, while approximately 40% went entirely untreated. Surgical resection of recurrent disease, the option with the greatest curative potential, was used particularly sparingly in octogenarians, who more often received chemotherapy alone — or nothing.

The analysis also identified who fell through this gap. Among the elderly, two factors independently predicted receiving no treatment for recurrence: synchronous metastases nearly tripled the odds (odds ratio 3.00), and omitting postoperative adjuvant chemotherapy raised them more than fivefold (odds ratio 5.22) — suggesting that patients who had declined, or been denied, intensive treatment early on tended to opt out again later. In younger patients the predictors were different: poorer anesthesiological fitness (odds ratio 2.28) and anatomical rather than partial liver resection (odds ratio 2.68). Within the elderly group itself, depth of tumor invasion, lymph node involvement, synchronous timing, bilobar distribution and large tumor size all independently worsened outcomes, and — perhaps counterintuitively — preoperative chemotherapy doubled the hazard of cancer-specific death, most plausibly a marker of aggressive biology rather than harm from the drugs. Even among treated patients, octogenarians still fared worse than the young, which the authors attribute to unrecorded differences in chemotherapy regimens and cycle numbers.

The authors are candid about the caveats. The study was retrospective; it could not compare operated octogenarians with those denied surgery, because the reasons for withholding operations — general condition, patient preference, oncological context — were not systematically recorded. Chemotherapy regimens varied across institutions and were negotiated case by case between physician and patient; details of second-line and later treatments for recurrence were unavailable; and the cohort was exclusively Japanese, meaning validation in European and American populations is needed before the conclusions are generalized. One intriguing nuance emerged: an ASA score of 3 or higher, a powerful predictor of death among the young, lost all prognostic value in the elderly — almost certainly because so many frail octogenarians had been filtered out before surgery, a selection bias the team acknowledges. Still, the central message is difficult to ignore. In the operating room, age 80 need not be the barrier clinicians fear, since recurrence-free outcomes equalized once case selection was accounted for. The inequity arises later, in the quiet decisions about whether a recurrence is worth treating in a very old person. As super-aged societies multiply worldwide, the researchers conclude, physicians should weigh those limits — and the lives they may cost — before the first incision is ever made.

Subject of Research: Long-term oncological outcomes of surgical resection for colorectal liver metastases in patients aged 80 years and older compared with younger patients, analyzed with propensity score matching in a nationwide Japanese cohort.

Subject of Research: Medicine

Article Title: Surgical Resection for Colorectal Liver Metastasis in Elderly Patients Aged ≥ 80: A Retrospective Nationwide Cohort Survey in Japan With Propensity Score Matching

Article References: Sugimoto, K., Saiura, A., Nojiri, S., Kobayashi, H., Honda, G., Kawai, K., Hasegawa, K., & Takahashi, K. (2026). Surgical Resection for Colorectal Liver Metastasis in Elderly Patients Aged ≥ 80: A Retrospective Nationwide Cohort Survey in Japan With Propensity Score Matching. Annals of Gastroenterological Surgery, 10(4), 1107-1119. https://doi.org/10.1002/ags3.70213

Image Credits: AI Generated

DOI: 10.1002/ags3.70213

Keywords: colorectal liver metastasis, liver resection, elderly patients, octogenarians, propensity score matching, recurrence-free survival, cancer-specific survival, overall survival, treatment for recurrence, adjuvant chemotherapy, nationwide cohort study, Japan

Cite Scienmag News
APA MLA Chicago

Ophelia Keating. (August 30, 2026). Liver Surgery Viable for Colorectal Cancer Patients 80 and Older, Study Finds. Scienmag. https://scienmag.com/liver-surgery-viable-for-colorectal-cancer-patients-80-and-older-study-finds/

Copy citation Download RIS

Tags: colorectal cancer liver metastasis treatmentcolorectal cancer prognosis in elderlycolorectal cancer prognosis in patients over 80effectiveness of liver resection in patients over 80effectiveness of liver surgery in patients over 80elderly cancer surgery risks and benefitsfactors influencing cancer recurrence in elderlyimpact of age on colorectal cancer survivalimpact of age on liver cancer treatment outcomesJapanese study on colorectal liver metastasesJapanese study on liver metastasesliver metastasis surgery in elderly colorectal cancer patientsliver surgery for elderly colorectal cancer patientsmortality rates in elderly liver metastasis patientsoutcomes of liver resection in octogenarianspost-operative treatment disparities in elderly cancer patientspost-surgery cancer recurrence in elderlyrisks and benefits of liver surgery in older adultssurgical decision-making in octogenarianssurgical outcomes in octogenarians with colorectal cancersurvival rates after liver surgery in older adultstreatment disparities for elderly cancer patientstreatment gaps for elderly colorectal cancer patients