A protein with a history in lung cancer. Krebs von den Lungen-6 (KL-6) was originally discovered as a lung cancer-associated antigen. It is a specific molecular form of a protein called MUC1 (mucin 1), which is found on the surface of epithelial cells lining the lungs and other organs. When cancer cells or damaged lung tissue produce KL-6, it is released into the bloodstream, where it can be measured with a standard blood test.
MUC1 promotes cancer progression. MUC1 is an oncogenic protein that promotes tumor growth, invasion, spread to distant sites (metastasis), and formation of new blood vessels (angiogenesis) in lung cancer. It also suppresses immune responses against cancer by increasing PD-L1 expression - a key mechanism cancer uses to evade immune attack. This biological activity makes elevated KL-6 a plausible signal of more aggressive cancer behavior.
The need for simple prognostic biomarkers. Current prognostic monitoring for lung cancer relies heavily on tumor tissue biopsies and expensive genetic or protein profiling tests. These are often unavailable through repeated testing due to invasiveness and cost - particularly in resource-limited settings. A simple blood test like serum KL-6 measurement could provide prognostic information that guides treatment decisions without requiring repeated tissue sampling.
Conflicting prior evidence. While some earlier studies showed that elevated pretreatment KL-6 levels were associated with worse outcomes in lung cancer patients, other studies found no significant association. This inconsistency motivated the authors to conduct a systematic review and meta-analysis pooling all available studies to reach a more definitive conclusion about KL-6's prognostic value.
Comprehensive literature search. The researchers searched four major medical databases - PubMed, Web of Science, Embase, and Cochrane Library - from database inception through June 23, 2025. Search terms covered KL-6, its synonyms (Krebs von den Lungen-6, MUC-1, mucin-1), and lung cancer terms. The study was pre-registered with PROSPERO (the international registry for systematic reviews), ensuring methodological transparency.
Strict inclusion criteria. Only studies that enrolled actual lung cancer patients and specifically evaluated pretreatment serum KL-6 as an independent predictor of survival were included. Studies were excluded if KL-6 was measured after treatment began, if a different form of MUC1 was measured, or if they had insufficient data for statistical analysis. Two independent reviewers screened all studies, resolving disagreements through discussion.
Statistical approach. Pooled hazard ratios (HRs) were calculated to quantify the relationship between high KL-6 levels and survival. The HR measures how much more quickly high-KL-6 patients experience disease progression or death compared to low-KL-6 patients. Heterogeneity across studies was assessed using I-squared statistics. Subgroup analyses examined whether the relationship differed by detection method, presence of interstitial lung disease, treatment type, and study design.
Study quality assessment. Each included study was independently assessed for bias risk using the Quality in Prognosis Studies tool, covering seven domains including patient selection, measurement of the prognostic factor, outcome measurement, and statistical analysis quality. Seven studies had low overall bias risk; the remainder had moderate bias risk. This quality framework ensures the meta-analysis conclusions are grounded in reliable underlying studies.
Progression-free survival results. Five studies involving 620 patients examined the relationship between KL-6 and progression-free survival (PFS) - the time until cancer worsens or a patient dies. Patients with high pretreatment KL-6 levels had significantly shorter PFS, with a pooled hazard ratio of 1.89 (95% CI: 1.46 to 2.44, P less than 0.001). This means high-KL-6 patients had approximately 89% higher risk of disease progression or death at any given time compared to low-KL-6 patients.
Overall survival results. Twelve studies covering 1,519 patients evaluated KL-6 and overall survival (OS). The pooled analysis showed that elevated KL-6 was associated with significantly shorter OS, with a hazard ratio of 1.76 (95% CI: 1.37 to 2.26, P less than 0.001) - representing a 76% higher mortality risk at any given time. The finding was consistent enough to be meaningful despite moderate heterogeneity across studies (I-squared = 51.9%).
The 500 U/mL threshold. The majority of included studies used 500 U/mL as the cut-off value for classifying patients as high or low KL-6. Pooling these studies confirmed that patients with KL-6 levels above 500 U/mL had significantly worse prognosis for both PFS and OS. This threshold aligns with what is routinely used in Japanese clinical practice as the upper limit of normal.
Consistent across treatment types. Subgroup analysis based on treatment regimens - including surgery, chemotherapy, targeted therapy with gefitinib, and immune checkpoint inhibitors - showed no significant differences between subgroups for PFS, suggesting that KL-6's prognostic value is not restricted to any particular treatment context.
The ILD complication. KL-6 is already an established biomarker for interstitial lung disease (ILD) - a group of conditions that cause progressive lung scarring (fibrosis). ILD affects 2.4% to 10.9% of lung cancer patients as a comorbid condition, and it is itself a major negative prognostic factor. This creates an important ambiguity: in patients who have both lung cancer and ILD, does elevated KL-6 reflect cancer aggressiveness, ILD severity, or both?
The subgroup analysis result. The meta-analysis directly addressed this question by stratifying patients by the presence or absence of ILD. In lung cancer patients without ILD, elevated KL-6 significantly predicted worse overall survival (hazard ratio 1.85, 95% CI: 1.21 to 2.83, P = 0.005). However, in patients with ILD, elevated KL-6 did NOT significantly predict OS (hazard ratio 1.27, P = 0.539).
Interpreting the ILD finding. The authors conclude that in patients with both lung cancer and ILD, elevated KL-6 levels primarily reflect the severity and activity of the fibrotic lung disease itself, rather than providing independent prognostic information about the cancer. The KL-6 signal is essentially "confounded" by the underlying ILD - making it difficult to separate cancer-driven KL-6 elevation from fibrosis-driven elevation.
A higher threshold may help. One study that used a much higher cut-off of 1,000 U/mL did find prognostic significance in ILD patients. This suggests that very high KL-6 elevations may carry a cancer-specific signal that can be detected above the elevated baseline established by ILD. However, this requires investigation in dedicated prospective studies with ILD severity-adjusted analyses before any clinical recommendations can be made.
Two measurement methods compared. The included studies used two different laboratory techniques to measure serum KL-6: electrochemiluminescence immunoassay (ECLIA) and enzyme-linked immunosorbent assay (ELISA). These techniques use different chemical detection principles, and ECLIA is generally considered to have superior analytical sensitivity and linear range compared to ELISA.
Striking performance difference. Subgroup analysis by detection method revealed a substantial difference in predictive performance. Studies using ECLIA showed a much stronger association between high KL-6 and worse OS (HR 3.30, 95% CI: 1.85 to 5.89) compared to studies using ELISA (HR 1.24, 95% CI: 0.78 to 1.94 - non-significant). This difference between methods was itself statistically significant (P = 0.009).
Clinical implication of the method difference. The superiority of ECLIA in detecting KL-6's prognostic value likely reflects its greater analytical sensitivity capturing a wider range of KL-6 concentrations and detecting differences that ELISA may miss. The authors recommend ECLIA as the preferred method for clinical KL-6 measurement when used for lung cancer prognostication.
Need for direct comparison studies. No study has yet directly compared ECLIA and ELISA for KL-6 measurement in the same patient samples. Direct head-to-head comparison studies would definitively establish the magnitude of the methodological difference and help standardize clinical practice. Until such studies exist, the ECLIA recommendation is based on the indirect subgroup analysis evidence from this meta-analysis.
KL-6 is produced by cancer cells. Evidence suggests that the elevated circulating KL-6 in lung cancer patients originates from the tumor itself. Studies have found that KL-6 expression in lung cancer tissues correlates with blood levels, and KL-6 levels fall significantly after surgical removal of the tumor. Similarly, patients who respond well to targeted therapy show declining KL-6 levels within weeks of treatment initiation - further supporting cancer as the source of elevated circulating KL-6.
MUC1's oncogenic role explains the association. MUC1 (the parent protein of KL-6) drives lung cancer progression through multiple mechanisms: promoting tumor cell survival, enhancing invasion and metastasis, and suppressing anti-tumor immune responses by upregulating PD-L1 on cancer cells. Cancers with higher MUC1/KL-6 expression are therefore more biologically aggressive, explaining why patients with higher blood KL-6 levels have worse outcomes.
Parallels in other cancers. KL-6 has demonstrated prognostic significance in other cancer types. In pancreatic cancer, positive KL-6 expression is associated with lymph node metastasis, tumor invasion, and significantly lower 5-year survival rates. Studies suggest KL-6 may promote cancer spread by suppressing cell adhesion proteins (E-cadherin and beta-catenin) that normally keep cancer cells from detaching and migrating.
Potential as a treatment response monitor. Beyond baseline prognosis, the observed decline in KL-6 levels following successful treatment suggests it could serve as a dynamic monitoring biomarker - rising with disease progression and falling with treatment response. This application of KL-6 for treatment monitoring deserves dedicated prospective investigation.
A simple pretreatment risk assessment tool. The main clinical promise of KL-6 is its simplicity: it requires only a standard blood draw before treatment begins, and a value above 500 U/mL identifies patients at substantially higher risk of disease progression and death. This information could immediately help clinicians select more aggressive treatment strategies, schedule more frequent monitoring, or enroll high-risk patients in clinical trials of novel therapies.
Currently limited to Asian populations. All 13 included studies were from Asian populations (primarily Japan, with two from Korea). The 500 U/mL cut-off was established from Asian reference ranges and clinical practice. Importantly, there is evidence of significant ethnic variation in baseline KL-6 levels driven by genetic polymorphisms in the MUC1 gene. Egyptian patients with liver cancer have been shown to have significantly higher KL-6 levels than Japanese patients, suggesting population-specific reference ranges may be needed.
Predicting treatment-related lung injury. An additional clinical application emerged from one study: elevated pretreatment KL-6 levels predicted higher risk of developing severe treatment-related interstitial lung disease (a potentially fatal complication of lung cancer therapy). If validated, this could help identify patients who should receive alternative treatment approaches or require particularly close pulmonary monitoring during therapy.
Limitations for current clinical use. The evidence is primarily from retrospective studies in Asian populations, most of which had NSCLC rather than small cell lung cancer. Optimal cut-off values are not definitively established, and the confounding effect of ILD comorbidity requires careful consideration. The authors recommend that clinicians interpret KL-6 results in light of whether the patient has underlying ILD, and await prospective validation before incorporating KL-6 into routine prognostic algorithms.
A validated prognostic biomarker for a specific population. This meta-analysis of 13 studies and 1,723 patients establishes that elevated pretreatment serum KL-6 (above 500 U/mL) is a biomarker of poor prognosis in Asian lung cancer patients without ILD. The associations with both progression-free and overall survival are statistically robust and consistent across subgroup analyses.
Key practical recommendations. Based on the evidence, the authors recommend electrochemiluminescence immunoassay (ECLIA) as the preferred testing method for KL-6 measurement, given its substantially stronger predictive performance compared to ELISA. The 500 U/mL threshold is supported by the available evidence for Asian patients, though it requires prospective validation and may need adjustment for other ethnic groups.
The ILD caveat is critical. Clinicians must assess whether lung cancer patients have concurrent ILD before interpreting KL-6 results. In ILD patients, elevated KL-6 reflects fibrotic disease severity rather than cancer aggressiveness, and the standard 500 U/mL threshold loses its cancer-specific prognostic meaning. Higher thresholds may be needed in this subgroup, but specific cut-offs await prospective validation.
Future research priorities. Large-scale prospective studies in diverse ethnic populations are needed to establish population-specific reference ranges and cut-off values. Research should extend to small cell lung cancer (currently underrepresented in existing studies), investigate KL-6 as a dynamic treatment response biomarker, and directly compare ECLIA and ELISA in the same patient samples. Prospective validation of KL-6 for predicting treatment-related lung injury is also warranted.