To understand the impact of alternative translation initiation on a proteome, we performed the first large-scale study of protein turnover rates in which we distinguish between N-terminal proteoforms pointing to translation initiation events. Using pulsed SILAC combined with N-terminal COFRADIC we monitored the stability of 1,941human N-terminal proteoforms, including 147 proteoform pairs with heterogeneous N-termini originating from the same gene that result from alternative translation initiation and incomplete processing of the initiator methionine. N-terminally truncated proteoforms were on average less abundant than canonical proteoforms, many had different stabilities and exhibited both faster and slower turnover rates compared to their canonical counterparts. These differences in stability did not depend on the length of truncation but on individual protein characteristics. In silico simulation of N-terminal proteoforms in macromolecular complexes revealed possible consequences for complex integrity such as replacement of unstable canonical subunits. The extent of intrinsic disorder in N-terminal protein structures correlated with turnover times, indicating that a change in the structural flexibility of protein N-termini in truncated proteoforms might impact proteoform stability. Interestingly, removal of the initiator methionine by methionine aminopeptidases reduced the stability of processed proteoforms while susceptibility for N-terminal acetylation, another common co-translational modification, did not seem to impact on turnover rates. Taken together, our findings reveal differences in protein stability between N-terminal proteoforms and point to a role for alternative translation initiation and co-translational initiator methionine removal in the overall regulation of proteome homeostasis.