As the most common hereditary CRC, LS has received increasing attention from colorectal surgeons, because hereditary background and molecular subtypes are significant factors in the prognosis of CRC patients [22]. However, due to LS’s greater difficulty of diagnosis and its different behavior from sporadic CRC, it is difficult to compare the long term prognoses for LS and SCRC. Using propensity score matching analysis, with the clinical characteristics of LS as baseline and a 1:2 matching ratio between LS and SCRC patients, we achieved comparability between the two groups in the current study. We found that OS period in LS patients was significant longer than that in SCRC patients, which may indicate that long-term prognosis for LS patients is better than that of SCRC patients. We used tumor progression of primary CRC to calculate PFS and included metachronous tumor occurrence into the calculations for TFS, because although LS patients have higher metachronous tumor occurrence rates, the occurrence of such second primary tumors is different from primary tumor progression in clinical significance. We found that PFS for LS patients is comparable to that for SCRC group, whereas the high proportion of metachronous CRCs and extra-colonic cancers in LS group have remarkably shortened TFS time, resulting in higher TFS rates for the SCRC group.
By analyzing genetic testing results for suspected LS patients, we found that the Chinese population’s LS genotype distribution is similar to distributions reported in western review,6 with MLH1 and MSH2 being the majority. However, due to insufficient sample size, no significant differences in the phenotypes and long-term prognosis of each genotype can be found. The insufficient sample size relates to the wider issue that, the detection rate and genetic counseling of LS in the Chinese population are lower than those in western populations, which is already considered insufficient [23]. There are a few possible reasons. Firstly, for patients with high hereditary CRC risk, existing screening policies and procedures largely originate from Western guidelines and thus may not be fully suited for Asian populations. Secondly, hereditary CRC research in Asia is in its early stages, so there are only a few retrospective reports on the subject. Therefore, prospective studies with larger samples are needed to better understand the clinical characteristics, genotypes, and phenotypes of Asian LS patients. This would improve screening and follow-up policies for Asian patients, which would allow higher detection rates and even larger samples for more future prospective studies.
While comparing primary tumor progression rates, we found that the liver, lungs, and abdominal lymph nodes remain the most common sites for CRC metastasis. Studies have confirmed that LS patients have lower risk of tumor recurrence or distant metastasis, which implies that LS patients should have longer PFS [24, 25]. However, in the current study, the LS and SCRC groups show no statistical difference in PFS, despite showing a significant tendency of dispersion in their PFS survival curves (Fig. 1). A possible reason is that because the tumor stages in both samples are relatively early, both groups have longer TFS. As a result, our follow-up time is not long enough to record tumor progression beyond TFS and find further statistical difference in PFS. Therefore, future studies should increase patient sample size and extend follow-up time to yield more accurate comparative results.
A higher probability of metachronous tumor occurrence, including endometrial cancers and other cancers in the LS tumor spectrum, is the most prominent feature of LS [6]. Thus, treating the first primary tumor is often only the first step of treatment for LS patients. In the current study, metachronous tumor occurrence was present in a large portion of patients in the LS group, which significantly shortened the LS group’s TFS. Standardized postoperative monitoring and follow-ups enabled us to detect these metachronous tumors at earlier stages, most of which were cured after complete resection. Especially after any second primary tumor occurrence, we largely performed extensive radical resections and individualized postoperative adjuvant treatments, including chemotherapy, targeted therapy, etc., according to each patient’s condition. As a result, in terms of survival rate, metachronous tumor occurrence should have no significant impact on OS if it can be cured through resection. This result highlights the importance of establishing and improving the registration system for hereditary CRC in clinical work. Creating standardized and individualized diagnosis and treatment plans with respect to each patients’ heterogeneity, including in age, tumor, and molecular characteristics, etc., is key to ensuring good prognoses. Only through strict monitoring and follow-ups can we detect and treat metachronous tumors in a timely manner, and minimize their impact on the long-term survival of LS patients after treatment.
OS is the most important indicator for evaluating long-term prognosis, and the current study confirms what many studies have already proposed, that LS patients have better long-term prognosis than SCRC patients [7–17].The current study has confirmed that apart from factors intrinsic to the initial primary tumor, such as tumor characteristics, pathological type, and differentiation grade, tumor progression such as recurrence and metastasis is independent prognostic factor affecting long-term survival. Thus, given that microsatellite instability (MSI) tumors experience lower rates of local tumor recurrence, especially at distant sites, than do microsatellite-stable (MSS) tumors [24].CRC displaying MSI that are diagnosed at early stages have a better prognosis compared to CRC displaying MSS [24]. In line with these expectations, our results show that the 5-year survival rate of LS patients is 97.6%, which is significantly higher than the 82.6% of SCRC patients. In addition, our multivariate analyses show that the etiology of the two groups (LS vs. Sporadic) is an independent factor in long-term survival after surgery. Thus, the current study further confirms that LS’s long-term prognosis is better than that of SCRC.
With respect to oncologic outcomes, LS-associated tumors are associated with better prognoses and therapeutic responses, in part because the DNA MMR system, which is missing in CRCs with MSI, is involved in triggering cell death after chemotherapy-induced DNA damage [7]. We did not include chemotherapy into our analysis of prognostic factors, however, because during survival analysis, we found that chemotherapy and TNM staging overlapped in their effects on prognosis, since only patients with late staging (II/III) received chemotherapy. Nonetheless, during subgroup analysis, we found that for patients with distant metastasis, although all received chemotherapy combined with targeted therapy such as bevacizumab and cetuximab, the LS group had longer OS than the SCRC group. This phenomenon further affirms the aforementioned explanation that LS patients, who has dMMR and thus MSI, should be more responsive to chemotherapy than SCRC patients. In fact, several prospective-retrospective analyses collecting data from different randomized trials demonstrated that the use of adjuvant 5-FU-based chemotherapy was effective in stage II CRC displaying MSI [24–27].In particular, stage III MSI cases did benefit from adjuvant 5-FU-based chemotherapy, albeit only if the MSI was associated with LS [24].Although numerous western guidelines on LS treatment exist today, there is no recognized consensus on treatment for recurrence and metastasis, due to significant differences between individual patients. Therefore, targeted combined chemotherapy should at least be offered to LS patients with recurrence or metastasis, regardless of whether complete resection can be performed. Simultaneous, corresponding prospective clinical trials and basic research will also contribute significantly to future treatment selection for preventing postoperative LS tumor progression and improving long-term survival.
During the analysis of factors in long-term survival, we found that microvascular invasion, cancer nodule occurrence, and the late TNM staging of the first primary CRC are three independent risk factors that affect PFS and OS. These factors have already been widely recognized as adverse prognostic factors [28, 29], which this study has once again confirmed. Given that all three factors occur later in a tumor’s course of disease, early diagnosis and treatment can mitigate these adverse factors and ensure better survival prognoses. Thus, after an LS patient is diagnosed, the patient’s relatives (esp. first- and second-degree relatives) should undergo genetic counseling and standardized follow-up monitoring in accordance to guidelines, as they have a higher risk of cancer than the general population.
Lastly, because early-onset tumors (< 50 y/o) is another important characteristic of LS, we conducted a subgroup analysis and comparison of LS and SCRC early-onset CRC patients. However, we found no significant difference in comparison of OS, which indicates that early-onset CRC is an indicator of hereditary CRC in general. Moreover, current domestic epidemic reports in China show a trend of CRC incidence at an increasingly younger age, which indicates that early-onset CRC is a growing concern [30]. Thus, early-onset cancer patients are recommended to undergo genetic testing in order to increase the detection rate of hereditary CRCs, including LS, so that more precise treatment and follow-ups can be performed for the patient and his/her relatives
The current study has the following limitations. Firstly, this study is retrospective. Even though propensity score matching was used, selection bias is still hard to avoid. Secondly, the LS group contains patients with synchronous primary tumors, a factor which could not be balanced on the baseline and thus may affect the results. Lastly, the sample size for the LS group is relatively small, and would ideally require further accumulation and longer follow-up times.